No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Capacitor ESR and C Monitoring in Modular Multilevel Converters
Abstract
The capacitor is one of the weakest components in the module multilevel converter (MMC). The rise of equivalent series resistance (ESR) is a prominent character to monitor the lapsed capacitor, but current researches only focuses on capacitance and neglects ESR of capacitors in the MMC. This paper proposed a sorting-based monitoring strategy for capacitors in the MMC, which monitors not only the capacitance but also the ESR of capacitor. This paper reveals the relationship among the capacitor's ESR, capacitance, current and energy. Based on the relationship, the ESRs and capacitances of submodule (SM) capacitors in the arm are indirectly sorted, respectively, and only the capacitor with biggest ESR and the capacitor with smallest capacitance in the arm are monitored. The proposed strategy not only realizes both ESR and capacitance monitoring in the MMC, but also proposes a simplified monitoring algorithm for MMCs with large number of capacitors. The simulation and experimental results confirm the effectiveness of the proposed monitoring strategy for MMCs.
... To monitor the ESR in MMC, Deng et al. built a model between the harmonic components and the ESR, so the ESR can be estimated with the model [13]. Xia et al., measured the charging transient and used the wavelet transformation to estimate the ESR [14], in which the charging transient is the time when the submodule changes between the bypassed states and the inserted states. ...
... The median values for the 20 monitored capacitors are in the range of 1 ± 1 % and 1 ± 10 % for capacitances and ESRs, respectively. The acceptable estimation ranges are chosen according to the state-of-the-art [13] and they are colored with a white background located in the middle of the boxplots, while the other ranges are marked with a red background. In total, the average estimation error of C and ESR are 0.18 % and 5.47 %. ...
... The methods using the assumption of neglecting ESR in C estimations are illustrated in dash lines, including a recursive least-square method (RLS) [4], a model-based method that monitors only C [6], and a model-based method that monitors C and ESR separately [13]. The estimation errors of these three methods increase as the ESR increases. ...
Reliability of capacitors is a critical factor in ensuring the optimal performance of modular multilevel converters (MMC), which can be enhanced through health status monitoring and predictive maintenance. However, existing literature on MMC submodule capacitor condition monitoring primarily focuses on capacitance while overlooking the influences of the equivalent series resistance (ESR). Furthermore, the second problem is that the ESR is assumed to be negligible in the capacitance estimation. The assumption is not always valid, and may introduce significant errors in capacitance estimation. To improve existing condition monitoring methods, this paper uses a capacitor-voltage equation to model the coupling effect of capacitance and ESR, based on which, particle swarm optimization (PSO) is used to update the estimations of capacitance and ESR together. The proposed method offers more reliable health monitoring with two health indicators and derives a better estimation accuracy when the ESR is not negligible. Furthermore, to ensure the existence of the global optimal solution, the convexity of the data-driven problem is studied. The effectiveness and feasibility of the proposed method are validated with simulations, experiments, and an open-source dataset.
... The method in [21] not only minimizes the burden but also introduces ESR monitoring for the first time. The approach establishes mathematical relationships between ESC, ESR, and the fundamental frequency component, enabling the ranking of SMs and focusing monitoring only on potentially anomalous SMs. ...
... Denoting Saui as the switching signal of the ith SM, vSM as the voltage of the SM capacitor, and iaui as the current flows through the ith SM. The expressions of the ith SM capacitor voltage and SM current in the upper arm (phase-a) are given by Equations (7) and (8) [15][16][17][18][19][20][21][22][23][24]: ...
... Denoting S aui as the switching signal of the ith SM, v SM as the voltage of the SM capacitor, and i aui as the current flows through the ith SM. The expressions of the ith SM capacitor voltage and SM current in the upper arm (phase-a) are given by Equations (7) and (8) [15][16][17][18][19][20][21][22][23][24]: ...
The sub-module capacitor is the most vulnerable component in a modular multilevel converter (MMC), and its aging poses a significant challenge to system stability. To accurately monitor capacitor aging, this article utilizes capacitor voltage fluctuations to recognize the inserted window for capacitance calculation using nearest-level modulation. Additionally, a time-slicing method is developed to improve accuracy. The proposed method, which combines the inserted window recognition method with the time-slicing algorithm, offers a simple, easy-implementation approach. Simulations and experimental results validate that the method achieves high accuracy (less than 0.5%). Moreover, it does not require additional sensors, precise extraction of switching signals, or interruption to the system’s normal operation, making it highly suitable for MMC systems with a large number of sub-modules. Furthermore, the proposed method also demonstrates strong robustness in dynamic conditions and can be extended to all sub-modules.
... In [104], a method is described that involves deriving the relationship between ESR and the capacitance among the arms of a submodule in an inverter circuit. (94)-(97) are derived to establish this relationship. ...
... Steps involved in the technique proposed in[104]. ...
... Characteristic variable calculation of Capacitor monitoring strategy in[104]. ...
Capacitors play a critical role in power electronic systems, and their health and performance directly impact system reliability and efficiency. This article provides an overview of advanced techniques developed for capacitor monitoring, focusing on diagnosing, estimating, and predicting capacitor health in various power electronic converters. The condition monitoring techniques are categorized based on the type of converter, and a comprehensive review and comparison of various methods are provided within each category. The article covers various techniques, including impedance spectroscopy, frequency response analysis, harmonic analysis, wavelet analysis, short-time Fourier transform, and algorithm-based predictive models. These techniques enable early detection of capacitor faults, accurate estimation of capacitance and equivalent series resistance (ESR), and prediction of the remaining useful life of capacitors. By implementing these advanced monitoring techniques, engineers and researchers can enhance system reliability, prevent unexpected failures, and optimize maintenance and replacement strategies for capacitors in power electronic systems.
... In this method, the second-order components of the capacitor voltage and current are extracted by the band-pass filter (BPF), and the SM capacitance is estimated by the recursive least square technique. This idea is further discussed in [18] to monitor both the equivalent series resistance (ESR) and the capacitance of the SM capacitor. In [19] and [20], the second-order components of the SM capacitor voltage and current are extracted by the BPF. ...
... In addition, it is noted that the initial value W0 is a constant value, and it will not influence the capacitance estimation results, according to (18). ...
... By the FFT extraction of FF components, the sampling noise can be canceled [16]. The methods in [17], [18], [19], and [20] use the second-order components of the SM capacitor voltages to estimate its capacitance, where a second-order BPF is applied to reduce noise interferences in these methods. However, the second-order components are not dominant, and ...
To reduce noise interferences and improve the steady-state estimation accuracy of submodule (SM) capacitors, a novel SM capacitance monitoring approach has been proposed for the modular multilevel converter (MMC) in this paper. Firstly, the fundamental frequency (FF) component is analyzed to be dominant of the capacitor voltage squared. Then, based on the energy equation W = Cv
2
/2, half the FF capacitor voltage squared is selected as the input variable, and the FF capacitor energy calculated by power integral is selected as the feedback component. By proper design of the estimation law, closed-loop capacitance monitoring can be achieved with reduced noise interference. Furthermore, based on the Lyapunov stability criterion, the estimation parameter is designed so that the estimation error asymptotically converges to zero. Compared with the conventional methods, the proposed approach is easy to implement and requires no operation interruption of the monitored SM. Simulation and experimental results indicate that by the proposed approach, the estimation error of the SM capacitance can be limited to 1%.
... For instance, ref. [7] utilizes the inherent circulating current to estimate the capacitance, while refs. [8], [9], [13] inject additional highfrequency circulating current to enhance the health-relevant signals. Meanwhile, some literature derive the capacitor health status based on specific modulation schemes, such as the estimation of the capacitance based on the phase-shifted carrier (PSC) [10], [14], and the nearest level modulation (NLM) [11], [12]. ...
... 1) The relationship between the ESR and the capacitor charging transient voltage is established, and a CM method Signal conditioning [7] Second-harmonic impedance + + N N N [8], [9], [13] High frequency source injection + unknown Y N [10] Fundamental frequency response in PSC + + N N N [11] Capacitor voltage variation in NLM + + N N N [12] Submodule switching times in NLM + N N N [14] Capacitor voltage phase in PSC + + N N N [15] Recursive weight least square method + + + N N Y [16] Kalman filter + + + N N Y [17], [18] Reference submodule + Y Y Y [19] Tunnel Magnetoresistance sensor + + + Y Y Y [20] Adaptive observer + + + N N Y [21] Fast-affine projection algorithm ...
... The minimum sampling frequency can be preset with a given maximum error according to (13). However, the practical sampling frequency must be higher than the theoretical value considering the additional noise, which is expressed by The estimation error would change when a capacitor is degraded. ...
This paper proposes a capacitor condition monitoring (CM) method for modular multilevel converter (MMC) in motor drive applications. The proposed method is based on wavelet decomposition of transient voltage signals, which is independent of modulation schemes and control strategies. The equivalent series resistance (ESR) change can be detected with a moderate computation requirement. Both simulation and experimental results have verified the performance of the proposed method.
... φ2n, φ2p, φ4n, φ4p are respectively the phase angles of these harmonics. According to [27], the SM capacitor current in upper arm of phase a can be expressed as Normally, the circulating current control has little effect on the yref_ua [28]. Therefore, it can be expressed as yref_ua=m·sin(ωgt), where m is the modulation index of MMC. ...
... Therefore, it can be expressed as yref_ua=m·sin(ωgt), where m is the modulation index of MMC. Combining (24) and (25), the rated capacitor voltage fluctuation ∆ucuaN can be obtained as (26 Fig. 12 shows the relationship between ε and Csm in the MMC of the proposed system, which is derived from (24)~ (27) and Table I. It can be seen that along with the increase of Csm, ε will be reduced; along with the reduction of Csm, ε will be increased. ...
... The voltage fluctuation ratio ε is 2.57% under the Csm of 2.7 mF. According to (24)~ (27) and experiment parameters in Table II, Fig. 22 shows the required Csm under various ε. The estimated ε is 2.6% when Csm is 2.7 mF, and accordingly the experimental result is almost consistent with the theoretical analysis. ...
Low-speed drive is one of the challenges for modular multilevel converters (MMCs) due to large capacitor voltage fluctuation. In this paper, a cascaded MMC and cycloconverter (CCV) based machine drive system is proposed to ensure stable operation of medium- -voltage and low-speed machine. The MMC provides medium-frequency ac voltage for the CCV, and the CCV converts the medium-frequency ac input to low-frequency ac output required by the machine. Detailed analysis about MMCs operation frequency, device current stress and submodule (SM) capacitance are given in this paper. Proposed drive system can operate at zero/low frequency under rated load torque, while the SM capacitance is much smaller than that in existing methods. Simulation and experimental studies are conducted, and the results verify the effectiveness of proposed system.
... However, the measurements and calibrations of individual capacitors in the first stage could be challenging due to the large number of capacitors [69]. The sorting technique is also applied in [74] to sort the SM, which features the highest ESR or the lowest capacitance in each arm of the MMC. The sorting process is based on the relationship between the energy and ESR and the current and capacitance of each SM; it is expressed as [74]: The sorting technique is also applied in [74] to sort the SM, which features the highest ESR or the lowest capacitance in each arm of the MMC. ...
... The sorting technique is also applied in [74] to sort the SM, which features the highest ESR or the lowest capacitance in each arm of the MMC. The sorting process is based on the relationship between the energy and ESR and the current and capacitance of each SM; it is expressed as [74]: The sorting technique is also applied in [74] to sort the SM, which features the highest ESR or the lowest capacitance in each arm of the MMC. The sorting process is based on the relationship between the energy and ESR and the current and capacitance of each SM; it is expressed as [74]: ...
... The sorting technique is also applied in [74] to sort the SM, which features the highest ESR or the lowest capacitance in each arm of the MMC. The sorting process is based on the relationship between the energy and ESR and the current and capacitance of each SM; it is expressed as [74]: The sorting technique is also applied in [74] to sort the SM, which features the highest ESR or the lowest capacitance in each arm of the MMC. The sorting process is based on the relationship between the energy and ESR and the current and capacitance of each SM; it is expressed as [74]: ...
Capacitors are critical components of power converter systems as they influence the cost, size, performance, and scale of such systems. However, capacitors exhibit the highest degeneration and breakdown rates among all power converter components due to their wear-out failures and short lifespans. Therefore, condition monitoring is a vital process to estimate the health status of capacitors and to provide predictive maintenance for ensuring stability in the operation of power converter systems. The equivalent series resistance (ESR) and the capacitance of the capacitor are two widely used parameters for evaluating the health status of capacitors. Unlike the ESR, the capacitance of a capacitor is suitable for the health monitoring of various types of capacitors; therefore, it is more preferable for large-scale systems. This paper presents an overview of previous research addressing this aspect of capacitors and provides a better understanding of the capacitance monitoring of capacitors utilized in power converter systems.
... Capacitor fault is another fault type in the MMC as well, which is mainly caused by the capacitance drop due to the capacitor aging [14]. In practical projects, electrolytic capacitors still have a wide range of applications due to its large capacity and low cost. ...
... The electrolyte of capacitor reduces with its working time, which makes the capacitance of capacitor gradually decreases [14,15]. Due to the cascade structure of MMC, there are large numbers of capacitors in MMC, whose capacitances could be different because of different operation conditions. ...
... Supposing that the 2 nd -order harmonic component of the circulating current has been suppressed, the arm currents can be described as sin (14) where au i and au i au i a are the dc and ac components of iau, respectively. Imau and α are the amplitude and phase angle of (12), it can be obtained ...
... In these cases, if the condition monitoring method only monitors capacitance, the converter might operate with a degraded capacitor that has a high ESR, and the end-oflife warning is not triggered until the capacitance reaches the end-of-life criteria which makes the estimated health status unreliable. Considering the importance of monitoring ESR in reflecting the capacitor degradation, recently published research has started to monitor the ESR for capacitors in an MMC, e.g., by using the wavelet decomposition to analyze the turn-on transient voltage step which requires extra components to measure the fast and small voltage step [21] and the harmonic-based method [22]. ...
... For instance, if the swarm size is 5, the interquartile ranges of capacitance and ESR estimations are 0.21 % and 14.47 % of 1 p.u., respectively; while if the swarm size is 10, these values are 0.11 % and 2.95 %, respectively. Considering the acceptable error ranges for C and ESR are ±1 % and ±10 %, respectively [22], the IQR of ESR when the swarm size is 5 is 14.47 % which is higher than the acceptable error range ±10 %. Therefore, it is better to have a larger swarm size, and the swarm size is selected as ten, which gives stable estimations in the acceptable error range. ...
... Combining the advantages of these two indicators can significantly enhance the robustness of capacitor parameter estimation methods against various uncertainties. Although ref. [9] is capable of simultaneously estimating the C and ESR, it fundamentally employs separate methods for the estimation of C and ESR, which could entail significant computational effort in practical applications. Therefore, it is of great challenge to develop a method that can simultaneously obtain both C and ESR information to enhance the universality of capacitor parameter estimation. ...
... • Harmonic injections [1], [2], [9], [10] • Utilizing low-frequency operational features [11], [12] • Using mid-frequency transients [8], ...
This paper proposes a capacitor parameter estimation method based on wavelet transform and convolution neural network (CNN). By fully utilizing wavelet transforms and the inherently non-ideal properties of bandpass filters, the lowfrequency and mid-frequency band features contained in capacitor voltages are extracted with high resolution. Leveraging these features, a subsequent CNN network simultaneously estimates two crucial aging indicators of capacitors, i.e., capacitance and equivalent series resistance (ESR). While most existing methods can only identify either capacitance or ESR, the proposed method stands out by addressing both. The integration of two different frequency features enables the proposed method to exhibit broader applicability across different modulation schemes and control strategies, and is less sensitive to load conditions and sampling frequency. Experiment results based on a modular multilevel converter case study prove the effectiveness of the proposed method
... For sub-microsecond and lower component time scales, the condition monitoring of power electronics systems is achieved through monitoring capacitances, equivalent series resistance (ESR), and semiconductors. Most methods for monitoring capacitors and ESR use voltage sensors to determine the voltage and ripple of the capacitors [20]- [23]. Additionally, some current sensorless methods for online monitoring of capacitance and ESR have been proposed [24] and [25]. ...
... This paper [6], [7], [9], [10], [18], [19] [12], [14], [15], [16] [17], [20], [23], [24], [25], [ switching frequency are basically the same, indicating that the RTDM method can accurately map the frequency features of power conversion under the timescale of the switching period. Additionally, the output signal range in the RTDM platform is 0.5V to 1.5V when the output impedance of the DAC module is 50Ω. ...
Operation condition monitoring of power electronics systems can improve analysis visibility for device administrators and researchers. However, the current sensor-based monitoring methods have limitations in accurately monitoring variables at multi-timescales. This article presents a real-time digital mapped (RTDM) method for the sensorless multi-timescale operation condition monitoring in power electronics systems using FPGA simulators. The mathematic models of power converters are deployed in the simulator considering multi-timescale characteristics. Both the digital simulator and the physical power converter are controlled by the same control signals, allowing the simulator to map the accurate operation conditions of the power converters without additional physical sensors. The RTDM method can monitor variables on the switching period timescale to the system transient timescale due to the consideration for the ideal switch modeling of the semiconductor switch. The effectiveness of the proposed method is demonstrated using a 30kW DAB converter as an example, through the integration of an experimental prototype and a simulator to create an RTDM platform. This method offers an easy and sensorless way to monitor multi-timescale variables in power electronics systems, providing a more comprehensive and visual analysis of their operation.
... However, the estimation of the ESR with these methods seems to be difficult because they utilize lowfrequency components in which the capacitance is dominant in impedance. The method presented in Ref. (19) calculates the ESR from the capacitor loss that is a product of the capacitor voltage and the current. This method, however, requires the strict synchronization of the detection timing of the capacitor voltage and the current. ...
... That is complex for the MMC because it is often controlled by hierarchical controllers (24) (25) . Moreover, Ref. (19) does not mention the sampling frequency for capacitor monitoring. ...
A cell capacitor is one of the most fragile components in a modular multilevel converter (MMC). Capacitance decreases and equivalent series resistance (ESR) increases, as the capacitor deteriorates. This paper proposes an online monitoring method for both the capacitance and ESR of the cell capacitors in an MMC without any additional components or operation. The proposed method measures the impedance of the cell capacitor not only at a low frequency but also at a switching sideband-frequency of the cell without current sensors for capacitor current. The high-frequency components around the switching frequency improve the estimation accuracy compared with the previous methods that measure only impedance at low frequencies. The experimental results confirm that the proposed method estimates the capacitance and ESR with errors of 3.1% and 8.8%, respectively.
... The cell capacitors are either Metallized Polypropylene Film Capacitors (MPPF-Caps) or Aluminum Electrolytic Capacitors (Al-Caps) and nowadays the MPPF-Caps are more adopted in industrial MMCC cells due to higher lifespan and self-healing capability [109], [110] while Al-Caps are especially used in prototypes and demonstrators [111]. However, according to [112], [113] Al-Caps are still advantageous in some MMCC applications where design objectives are high energy density and low costs. ...
... The impact of aging on the performance of the capacitor is usually associated with the loss of dielectric properties, a decrease of capacitance, and an increase of the ESR [112] and power losses, which can cause a destructive shortcircuit [150] (see Tab. 4). These last two effects force the cell to increase the commutation frequency of the IGBTs to maintain the voltage ripple within suitable boundaries and also to achieve voltage balancing among the cluster cells [160]. ...
Modular Multilevel Cascade Converters (MMCCs) are considered a promising power electronics topology in industry. Their scalability allows to reach (ultra/very) high voltage levels with low harmonic content and high efficiency and makes MMCCs an ideal solution for high-power applications; such as electrical drives, solid-state transformers and high-voltage direct-current (HVDC) transmission systems. However, the high levels of thermal, electrical and mechanical stress on the power electronics devices and the large number of components (e.g. capacitors or semiconductors) make MMCCs prone to faults reducing its reliability. In this first part of the paper, a comprehensive overview of reliability on MMCCs, failure mechanisms and fault impact analysis in MMCCs, including failure rates and fault modes is presented. Also a set of tables which collect all information to easily detect and identify faults in MMCCs is presented.
... Another common capacitor failure mode is a gradual increase of ESR and a decrease of capacitance beyond allowable design tolerance levels [20]. The ESR of a capacitor can be calculated using (1) [22]. ...
... As discussed in [22], the resistive components of Figure 2 represent losses due to leakage current (R p ) and resistance of the physical components of the capacitor (R 1 ). The total ESR of a capacitor is capable of increasing for a number of reasons, the most important cause for this study being due to the evaporation of electrolyte [23], which decreases the value of C 1 thereby increasing ESR as per (1), noting that both R p and R 1 are not affected by the evaporation of electrolyte. ...
Low voltage devices connected to public electricity supply networks can be subjected to a wide range of voltage quality, including sustained overvoltage. Assessing the impact of sustained overvoltage on devices connected to the distribution network is important in determining the appropriate variance of standard voltage levels without unduly impacting the utility and customers from a technical and economic perspective. Therefore, a clear understanding of the impact that sustained overvoltage has on the lifetime of the connected equipment is a necessary task. This paper investigates the impact of sustained overvoltage through a series of accelerated-life testing experiments on a custom designed test device representative of a common interface between the power supply and distribution network, a switch mode power supply. The switch mode power supply with combined rectifier and filtering capacitor represents one of the most common front ends of LV equipment in modern appliances, and the work here concentrates on the impact of overvoltage on capacitor ageing. The results of the testing indicate that there is an accelerated ageing impact correlated with the applied voltage magnitude. Furthermore, analysis shows that mismatches between appliance voltage rating and sustained network voltage, leading to accelerated ageing, may result in premature device failure without the consumer being aware of the root cause.
... Reliability is one of the important issues for MMCs. The MMC consists of numerous submodules (SMs) and each SM contains several power switches, where each switch could be a potential failure point and the switch open-circuit fault would affect the safe operation of MMCs [3], [4], and accordingly it is crucial to localize the faulty SM after the fault occurrence. ...
Fault localization is one of the most important issues for modular multilevel converters (MMCs) consisting of numerous switches. This paper proposes an Isolation Forest (IF) based SM switch open-circuit fault localization method for MMCs. Based on the continuous sampling SM capacitor voltages, a number of Isolation Trees (ITs) are produced to construct the IFs for MMCs. Through the comparison of continuous IFs outputs, the faulty SM can be effectively localized. The proposed IF based fault localization method only requires SM capacitor voltages in the MMC to construct concise low-data-volume tree models, and uses sparsity and difference properties of outlier data to localize fault, and accordingly it simplifies calculation complexity. In addition, it does not require the MMCs mathematical models and manual setting of empirical thresholds. Simulation and experiment are conducted, and the results confirm the effectiveness of proposed method.
... The electrolytic capacitor is expected to provide a low-impedance path. However, due to the aging process, the equivalent series resistance (ESR) of the electrolytic capacitor increases which can cause higher core loss [6]- [8]. The aging process is mainly caused by the evaporation of electrolytic liquid resulting in the rise in ESR [9]. ...
Electrolytic capacitors are very popular for use in the dc-link in power electronics applications and especially in renewable energy systems. However, their life expectancy and reliability can be severely affected by the ripple in the capacitor current that can cause increased thermal stresses. Further, the switching of the power electronics switches and the presence of harmonic components can increase the ripple in the capacitor current worsening the life expectancy. This paper proposes the use of a shunt active filter to maintain an almost constant dc-current in the capacitor and hence to mitigate the events that can cause the capacitor life expectancy to reduce. This filter reduces the ripple current irrespective of the frequency spectrum. The design principle and the resulting operating condition have also been derived. The results from simulation studies using Matlab/Simulink and the experimental results from a laboratory prototype have demonstrated the effectiveness of the proposed technique.
... However, two SMs have to be out of service during the monitoring period. Reference [26] presents a sorting-based capacitance monitoring method to reduce computational burden in the MMC with a large number of SMs, where only the SM with the smallest capacitance is judged and estimated. In [27], the monitored SM is controlled to be bypassed from the arm and the capacitor is discharged through the bleeding resistor, where the time of the capacitor voltage decreasing to a predefined value is used to estimate the SM capacitance. ...
The capacitance monitoring is one of the important issues for the MMC to obtain high reliability. Since the pulse-width modulation (PWM) is usually implemented in the Field Programmable Gate Array (FPGA), the produced precise switching states in the FPGA cannot be directly access by the top Digital Signal Processor (DSP) controller, which poses challenges on the existing capacitance monitoring methods based on the precise switching states of the MMC. This paper presents a SM capacitance monitoring strategy for the MMC with a simple algorithm, where the fundamental frequency components of the SM capacitor voltage and current are extracted to estimate the SM capacitance based on reference, but not the precise switching states. The proposed scheme not only simplifies the implementation and calculation, but also avoids heavy communication burden between DSP and FPGA as well as control limitations. Besides, the proposed strategy effectively eliminates noise impact from sensors and increases accuracy. Simulation and experimental studies are implemented, and the results confirm the effectiveness of the proposed strategy.
... This method is further extended to estimate C and ESR using a sorting-based capacitance monitoring. In this method, capacitance of only one SM capacitor having the biggest ESR and smallest capacitance of SM capacitor in the arm is estimated [29]. An improved RSM-based capacitance estimation method is presented to enhance the SM capacitor monitoring accuracy by utilising the voltage sensor measurement range [30]. ...
Impedance characteristics of the floating capacitor in the submodule (SM) of a modular multilevel converter (MMC) change due to its chemical and aging characteristics. This can result in a rise of voltage ripple across SM capacitors, uneven power distribution among the SMs and could also interrupt the operation with prolonged use of degraded capacitors. Existing balancing approaches are presented for the MMC with equal SM capacitance and ignored the effect of capacitor degradation as well as parameter mismatch. Hence, a closed‐loop control scheme that incorporates the effect of capacitor degradation is much necessitated. This study aims to overcome this gap by proposing a new power balancing control algorithm for SM capacitors, which effectively controls and evenly distributes the power among SMs. The efficacy of the proposed strategy is substantiated through PLECS simulations and OPAL‐RT/OP4510 controller‐based experiments on a single‐phase MMC laboratory prototype.
... Besides, the ESR monitoring error is ~5% on an average [133] , which is still not high enough. Thus, a sorting-based monitoring method is reported; it monitors the capacitance together with the ESR of capacitors [137] . In this method, only capacitors with the biggest ESR or smallest capacitance are monitored. ...
In the present scenario, modular multilevel converters (MMCs) are considered to be one of the most promising and effective topologies in the family of high-power converters because of their modular design and good scalability; MMCs are extensively used in high-voltage and high-power applications. Based on their unique advantages, MMCs have attracted increasing attention from academic circles over the past years. Several studies have focused on different aspects of MMCs, including submodule topologies, modeling schemes, modulation strategies, control schemes for voltage balancing and circulating currents, fault diagnoses, and fault-tolerant control strategies. To summarize the current research status of MMCs, all the aforementioned research issues with representative research approaches, results and characteristics are systematically overviewed. In the final section, the current research status of MMCs and their future trends are emphasized.
... i dc is the dc-link current of the MMC. According to [30] and [31], the capacitor current i caui in the ith SM of the upper arm of phase A can be expressed as follows: ...
The capacitor voltage balancing is one of the most important issues for safe and reliable operation of the MMC. This paper proposes a currentless submodule individual voltage balancing control (SMIVBC) to realize capacitor voltage balancing in the MMC. Through regulating the dc component in each SM capacitor current by modulation index m-based SMIVBC or phase angle m-based SMIVBC, each SM capacitor voltage can be individually controlled to follow reference value, which can realize capacitor voltage balancing in the MMC. The proposed SMIVBC not only requires no sorting technique to select the SMs in the MMC avoiding sorting algorithm, but also requires no knowledge of current in the MMC, which reduces the sensors and sampling signals, saves the cost and improves the reliability. Simulation studies with professional tool PSCAD/EMTDC and experiment studies with a down-scale prototype in the laboratory are both conducted and their results confirm the effectiveness of the proposed SMIVBC for the MMC.
... Reliability is one of the substantial issues for the MMC consisting of a large number of power electronic switches and each switch could be a potential failure point [17]- [19]. Normally, it is desired that MMC can continue operating without any interruption, even though some of the SMs malfunction [20]. ...
The modular multilevel converter (MMC) is an attractive option for high-voltage and high-power applications. Fault localization is one of the most important challenges for the MMC consisting of a large number of power electronic switches. Existing literature has not provided a complete fault localization strategy for the MMC under submodule (SM) lower switch open-circuit fault, which only considers the MMC in inverter mode and neglects the MMC in rectifier mode. This paper proposes a fault localization strategy for the MMC under SM lower switch open-circuit fault, which is suitable for the MMC not only in inverter mode, but also in rectifier mode. This paper analyzes the MMC fault characteristics under SM lower switch open-circuit fault, which considers the MMC in inverter and rectifier mode, respectively. Based on two modes' common fault characteristic that the capacitor is charged all the time during the period when the arm current is positive, the faulty SM can be effectively located by the proposed fault localization strategy. The proposed strategy is not only implemented in simulation with professional tool PSCAD/EMTDC, but also verified with a down-scale MMC prototype in the laboratory. The results confirm the effectiveness of the proposed strategy.
Metallized film capacitors (MFCs) have been widely used in power electronics such as DC transmission, electric vehicles, and wind power converters, due to its high reliability. MFCs has the unique self-healing discharge characteristics, which can remove internal defects and achieve stable operation. The self-healing discharge signal, which correlates with the degradation of MFCs, can be utilized as a characteristic parameter for assessment. In this paper, an intelligent monitoring approach for assessing the degradation status of MFCs based on the ultrasonic eigenfrequency of selfhealing discharge signals is proposed. 1D-convolutional neural network (1D-CNN) is used for feature extraction and classification. The experimental results reveal that the method can quickly extract the ultrasonic eigenfrequency feature of a period of time, and the accuracy of degradation status classification reaches 98.1%, showing superior performance in degradation status of MFCs monitoring conditions.
Capacitors are key components of modular multilevel converters (MMC), and capacitor condition monitoring is a crucial issue in MMC. This paper introduced a capacitor monitoring method based on submodule (SM) insertion windows to enhance the reliability of the MMC. Firstly, the voltage difference between the reference SM and other SMs is monitored to assess the capacitor condition of the SMs. Secondly, for the selected SM capacitors in abnormal state, an insertion window for the SM is extracted by introducing a time delay factor, and the subsequent capacitance calculation is performed based on the voltage changes of the reference SM and other SMs within the monitoring interval. The proposed monitoring method can effectively identify the abnormal capacitor state and accurately calculate the capacitance value. Moreover, the proposed method does not affect the normal operation of the MMC. Finally, simulation analysis and experimental validation are performed to verify the effectiveness of the proposed method.
Considering that control parameters are difficult to obtain due to industry confidentiality and intellectual property limitations, the existing white-box digital twin (DT) models of single-phase PWM rectifiers are difficult to apply in practical applications. Therefore, this article develops a gray-box DT model that can overcome the limitations of unknown control parameters and has higher accuracy. Firstly, the dead-time effect and the equivalent series resistance of capacitors are considered in the DT model to improve model accuracy. The operational characteristics of the DT model are closer to the actual system and have more comprehensive health monitoring capability. Then, based on the collected data from actual physical systems under dynamic and static operating conditions, the developed DT model is implemented on an independent server. The key parameters of the model are accurately monitored by using a hybrid algorithm of particle swarm and whale optimization. Finally, the proposed model and its parameter monitoring capability have been validated under different operating conditions, parameters, and control schemes, and the error impact has also been analyzed. This study contributes to the application of non-invasive health monitoring in practical power converters.
In modular multilevel converters (MMCs) of HVDC system, metallized film capacitors (MFCs) are commonly used as DC-link capacitors. Based on the unique feature of the self-healing discharge, MFCs can clear the inner defects and realize the operation stability. However, the self-healing discharge can cause a continuous aging of MFCs due to the accumulation of the electrode area loss. In recent years, although electrical parameters have been sufficiently investigated to monitor operating states of MFCs, the correlation between discharge characteristics and the aging states is still not able to be clarified based on these parameters. To address the issues, this paper provides a novel perspective by proposing the discharge monitoring and aging assessment method based on the acoustic signal. First, an accelerated aging platform was established to obtain MFC samples in different aging degrees. Then, the acoustic signal and the pulse current signal of MFC samples were both detected to demonstrate the feasibility in terms of using acoustic signals as characteristic parameters. Further, the processed discharge data were subject to cluster analysis to investigate the relationship between discharge statistic characteristics and aging states of MFCs. The results indicate that the aging state of MFC can be assessed using the frequency spectrum characteristics and the cluster ratio of AE signals, which would instruct the future research on the monitoring method of MFC using non-electrical parameters.
Modular multilevel converters (MMC) have the characteristics of high modularity, good availability and high-power quality. Thus, they are widely used in medium and high-power applications. To meet large capacity application requirements, a large number of capacitors is applied in parallel and series. However, capacitors are one of the most vulnerable components in MMCs. To identify abnormal capacitors, a condition monitoring method for capacitors is proposed in this paper using the cumulative sum detection of the sliding window algorithm. First, the bilateral cumulative sum algorithm of the sliding window is proposed to extract the switch-on time and switch-off time of the submodules (SMs). Second, the capacitance is calculated using the monitored capacitor voltage and arm current during the switch-on status. Finally, the accuracy and effectiveness of the proposed method are verified by an MMC system operating condition simulated experiment platform. The proposed condition monitoring method simplifies the calculation of capacitance with a high accuracy by removing the requirement of measuring the switching signal. In addition, it does not have an adverse influence on the normal operation of MMC systems.
Metal-oxide-semiconductor field-effect transistors (MOSFETs) undergo fatigue degradation under high thermal and electrical stresses. This process results in changes in their parasitic parameters, which can be detected using frequency domain reflectometry (FDR). Frequency domain impedance analysis is employed to characterize the various quality states of Si and SiC MOSFETs obtained from accelerated aging experiments. Results demonstrate a consistent increase in parasitic resistance as the devices degrade. By determining the drain-source parasitic resistance at the self-resonant frequency
and the drain-source on-resistance for MOSFETs with varying degradation degrees, positive linear numerical fitting equations
(14)
–
(15)
are established to predict MOSFET degradation under zero DC bias voltage. In addition, FDR technology is used to identify the drain parasitic resistance at the
of MOSFET samples with different sizes of defects in the sintered silver layer. These results reveal a positive correlation between the quality of the sintered silver layer and
. The proposed approach is an effective quality screening technology for power semiconductor devices without requiring power-on treatment.
DC-link capacitor is the important device and the vulnerable electronic part of the train converter. The capacitance decline of the DC-link capacitor is accelerated because of the influence of voltage, current, charge-discharge frequency, temperature and other factors. This will lead to the circuit performance degradation and even capacitor failures. Therefore, the capacitance estimation of DC-link capacitors is necessary and of great significance to the safe operation of the train. However, the voltage sensors used in the train converter have large noise and low sampling frequency, which makes it difficult to obtain accurate capacitance estimations with existing methods. In this paper, an effective capacitance estimation method is proposed to solve these problems. The pre-charging process is analyzed and an estimation model is built to weaken the impact of the sensor noise. Based on the model, a three-stage iteration algorithm is used for capacitance estimation, which can effectively search the optimal solution at a low sampling frequency and further reduce the noise influence. The proposed method was verified on a metro vehicle and compared with other algorithms. The experimental results show that the aforementioned problems can be well solved by the proposed method and the estimation error is within 2%.
The capacitor condition monitoring is an important issue for the reliable operation of modular multilevel converters (MMCs) in high-voltage applications. The numerous capacitors make the monitoring for the capacitor condition complex and computation-consuming. This paper proposes a hierarchic capacitor condition monitoring strategy to assure the reliability of high-voltage MMCs. In the strategy, the switching times of the switching signals are counted to detect the condition of the capacitor in each submodule (SM). Then, the subsequent capacitance calculation method is only conducted to the abnormal capacitors indicated by the above condition detection algorithm. Meanwhile, the variations of the measured and reference capacitor voltages are employed to calculate the capacitance in the proposed method. The condition detection algorithm shrinks the scope where the abnormal capacitors locate, which avoids calculating the total capacitances in MMCs. In the proposed capacitance calculation method, the complex operations are further saved. Both the two measures reduce the computation burden effectively. The implementation of the whole monitoring strategy is simple and fit for high-voltage MMCs. Moreover, the strategy has no adverse influence on the normal operation of MMCs. The effectiveness of the proposed strategy is verified by the simulation studies with the professional tool Matlab/Simulink.
DC-link capacitor is one of the significant parts of traction converters. Due to the impact of electrical stress, high temperature and humidity, its capacitance degrades faster than expected. Accurate capacitance estimation is critical to the condition monitoring of DC-link capacitors, which is also the basis of online capacitor life prediction. Currently, there is a big challenge for capacitor capacitance estimation in railway applications. The noise fluctuation of the voltage sensor may be nearly equal to that of ripple voltage, leading to the considerable errors of the existing capacitor condition monitoring methods. Therefore, a capacitance estimation method based on pre-charging model and noise evaluation is proposed in this paper. The capacitance is estimated by improved recursive extended least square (RELS). The influence of the noise can be effectively reduced and both the calculation accuracy and convergence speed can be well guaranteed through an independent noise estimation model. The proposed method has good immunity in a certain range of signal-noise ratio (SNR) or signal bias. It shows good performance in experiments.
With the emerging wide bandgap (WBG) semiconductor development, the increasing power density and efficiency of power electronic converters may cause more switching oscillation, electromagnetic interference noise, and additional power loss, further increasing the probability of device failure. Therefore, determining and quantifying the failure of a metal-oxide-semiconductor-field-effect transistor (MOSFET), which assembled using WBG semiconductor in some applications, is crucial to improving the reliability of a power converter. This study proposes a novel failure quantitative assessment approach based on MOSFET parasitic parameters. According to the two-port network theory, MOSFET is equivalent to some second-order RLC circuits composed of independent inductances, capacitances, and resistances in series. Then, the frequency-domain impedance associated with the physical failure of MOSFET is identified through frequency domain reflectometry. Accelerated aging and bond wires cut-off experiments are employed to obtain various quality states of the MOSFET device. Result shows that the MOSFET quality level and its number of bond wire lift-offs can be quantified effectively. Drain-to-source on-resistance ( R DS(on) ) that normally represents the MOSFET quality shows a positive linear function relationship on drain-to-source parasitic resistance ( R D + R S ) during the quality degradation proceeding. This finding matches with the correlation established between R DS (on) and R D + R S in theory. Meanwhile, source parasitic inductance ( L S ) increases with the severity of bond wires faults, and even the slight fault shows a high sensitivity. The proposed approach would be an effective quality screening technology for power semiconductor devices without power on treatment, which can effectively avoid the impact of junction temperature and test conditions (current and voltage) on test results, and does not need to design additional test circuits. The test frequency range we used in this approach was 10–300 MHz, which to some extent is suitable for providing an on-line quality monitoring technology for high-frequency WBG power devices manufacturing.
Aluminum electrolytic capacitors are the most commonly used components in circuit systems but are the weakest part of the systems. To make better use of this kind of components, it is crucial to monitor their status and parameter variations more accurately. This brief puts forward a prototype to test the frequency-dependent characteristics of the capacitors, both the capacitor voltage and current are acquired by using the prototype, while a variable forgetting factor recursive least square (VFF-RLS) method based scheme is applied to identify the capacitance (C) and the equivalent series resistance (ESR). Experiments are present by using the proposed scheme and the results of which are compared with the LCR meter measurement results. The extensive comparison results show that the relative error between experimental results and measurements are less than 5.03% which are acceptable compared with some other existing schemes and indicate that the proposed scheme is effective for the online parameter identification of Aluminum electrolytic capacitors.
The capacitors are one of the most important components in modular multilevel converter (MMC). Due to the large numbers of capacitors applied in MMC, the primary technical challenge about the capacitors is reliability. To identify the abnormal capacitors and improve the performance of MMCs, the capacitor condition monitoring strategy is proposed in this paper. In the monitoring algorithm, the relationship of the capacitances in an arm is deduced. The actual and nominal sums of switching signals for submodules (SM) are used to estimate the capacitances. Furthermore, a compensation for dead time is added to enhance the monitoring accuracy. The proposed condition monitoring algorithm not only simplifies the calculation of capacitances with high accuracy, but also has no adverse influence on the normal operation of MMCs. The feasibility of the proposed strategy is verified by the simulation results in Matlab/Simulink.
Online monitoring of dc-link capacitor's equivalent series resistance (ESR) could provide valuable and timely information for power converter failure prognosis and predictive maintenance. The conventional ESR estimation methods are mainly based on pulsewidth modulation (PWM) voltage and current ripples (ESR≈Δu
c
/Δi
c
), but the performance is affected by operating conditions (especially at light-load), quasi-resonance effect and sensor amplitude error. To overcome these difficulties, an online ESR monitoring method is proposed using the damping characteristic of dc capacitor switching ringings. The intrinsic high-frequency (HF) resonance incidents between the power device fast switching and converter parasitic parameters are utilized in this article. First, the mechanism of dc capacitor switching ringing is analyzed by the HF resonance equivalent circuit. The analytic relationship between the damping characteristic and the ESR is established. Then, the condition monitoring (CM) scheme, noncontact HF sensor, and half-power bandwidth algorithm for ESR estimation are presented. Finally, an experimental study is carried out to validate the feasibility and effectiveness. The damping of switching ringing is inherently sensitive to the ESR of dc capacitor under the condition of resonance. Both theoretical analysis and experimental results confirm that the proposed method can achieve mΩ-level accuracy and good robustness under different conditions.
The harmonic transfer of the modular multilevel converter (MMC) leads to the harmonics intersection between the AC- and DC- sides and introduces additional harmonics, which worsen the power quality. Especially, the low-order harmonics, which are difficult to filter out, may cause harmonic resonance and the aging of equipment. In this paper, the mechanism of harmonic transfer for MMC is revealed more easily and clearly based on the relation bewteen arm currents and voltages. The successive approximation method (SAM) is extended for the quantitative analysis of harmonic transfer. A digital-iteration scheme (DIS) is proposed to derive the general quantitative expressions of harmonics transfer of the MMC. And the DIS is appropriate for both conditions with and without considering the effect of the control system. By the iterative calculation, the harmonics can be calculated more accurately, especially the inner-harmonics. The effectiveness of the quantitative analysis is verified by comparing the calculation, simulation and experiment results.
Modular multilevel converter (MMC) with reduced number of voltage sensors is conducive to the reduction of equipment cost and hardware complexity, however, usually resulting in the performance degradation. For example, when only one voltage sensor is equipped for the measurement of arm voltage, estimation of capacitor voltages may be not accurate enough, moreover, condition monitoring and fault diagnosis of submodules (SMs) are hard to implement. Thus, capacitor voltage estimation with capacitance self-updating based on grouping measurement (GM) is proposed in this article. By grouping and updating SM capacitance periodically, the accuracy of voltage estimation is greatly increased, meanwhile, severely aging or faulty capacitors can be detected, convenient for prompt replacement. Furthermore, GM-based fault diagnosis and fault disposal are proposed. Utilizing the measured and estimated group voltages, single or multiple SM open-circuit failures can be located in several sampling cycles and the isolating time of faulty SMs can be easily obtained even considering the time delay of bypass switch. Therefore, capacitor overvoltage and current distortion during SM failures are restrained effectively, which enhances the reliability and achieves seamless operation of MMC. Finally, the effectiveness of the proposed scheme is confirmed by simulation and experimental results.
The modular multilevel converter (MMC) is attractive for medium- or high-power applications because of its high modularity, availability and high power quality. In this paper, the dc-link current ripple of the three-phase MMC derived from the phase-disposition pulse width modulation and circulating current control is analyzed. A variable carrier phase-shifted angle control strategy under the two-group carriers-based modulation is proposed to suppress the dc-link current ripple of the three-phase MMC. Through the regulation of the carrier phase-shifted angles under the two-group carriers-based modulation in each carrier period, the high-frequency current injected into the dc link from the three phases of the MMC can be counteracted, which can effectively suppress the dc-link current ripple of the MMC, and therefore improves the dc-link power quality. Simulations and experimental studies of the MMC are conducted, and the results confirm the effectiveness of the proposed control strategy for the MMC.
The modular multilevel converter (MMC) is attractive for medium/high-voltage and high-power applications because of the advantages of its high modularity, availability and high power quality. Due to chemical process, aging effect, etc., the capacitor in the submodule (SM) of the MMC would gradually deteriorate and the capacitance would drop, which would cause unbalanced SM power losses distribution in the same arm of the MMC and affect the reliability of the MMC. This paper proposed an equivalent-reference (ER) control method, which can effectively realize balanced SM power losses distribution in the same arm of the MMC through the voltage-balancing control for the virtual capacitor voltages in the MMC under the capacitor deterioration. The proposed ER control can effectively improve the reliability of the MMC with the balanced SM power losses distribution in the MMC under capacitor deterioration. The simulation studies with the time-domain professional tool PSCAD/EMTDC are conducted and a down-scale MMC prototype is also tested with the proposed control strategy. The study results confirm the effectiveness of the proposed control strategy.
The modular multilevel converter (MMC) is attractive for medium or high-power applications because of the advantages of its high modularity, availability, and high power quality. Reliability is one of the most important challenges for the MMC consisting of a large number of submodules (SMs). The capacitor monitoring is one of the important issues in the MMC. This paper proposed a reference submodule (RSM)-based capacitor monitoring strategy for the capacitance estimation in the MMC, where the capacitances in the monitoring SMs can be estimated based on the capacitor voltage relationship between the RSM and the monitoring SMs. The proposed monitoring strategy does not rely on the information of all capacitor voltage and current, which effectively simplifies the algorithm for capacitance estimation. The simulation studies with the time-domain professional tool PSCAD/EMTDC are conducted and a down-scale MMC prototype is also tested in the laboratory with the proposed capacitor monitoring strategy. The study results confirm the effectiveness of the proposed capacitor monitoring strategy.
As the penetration of distributed energy resources (DERs) keeps increasing, a single microgrid is incapable of controlling and consuming the large number of DERs. Therefore, to interconnect multiple microgrids and organise them as an integrated system is a promising way to solve the identified problem. Here, an advanced and comprehensive design of microgrid interface for multiple microgrids is proposed in order to facilitate the interconnection of multiple microgrids and reinforce the system controllability. The microgrid interface, named hybrid unit of common coupling (HUCC), employs modular multilevel converter and provides both AC and DC connection. Energy storage system is also designed in the HUCC. The advanced HUCC not only provides a unified microgrid interface, but also improves the flexibility and coordination of multiple microgrids. A case study of three interconnected HUCC‐based microgrids is carried out in PSCAD/EMTDC. The simulation results validate the effectiveness of the HUCC.
A subsynchronous oscillation (SSO) phenomenon has been observed in a modular multilevel converter-based high-voltage DC (MMC-HVDC) transmission system for wind farm integration in the real world, which is independent of the type of wind turbine generator. This kind of oscillation appears different from those in DFIG-based wind farm with series-compensation line or wind farm integration through two-level VSC-HVDC transmission system, because the internal dynamics of the MMC may have significant impact on the oscillation. By far, however, very few papers have reported it. In this paper, the generation mechanism of the SSO phenomenon in an MMC-HVDC transmission system for wind farm integration is revealed from an impedance point of view. The harmonic state-space (HSS) modeling method is applied to model the multi-frequency behavior of the MMC, based on which, the ac-side small-signal impedance of the MMC is analytically derived according to harmonic linearization theory. As a general rule, the controller parameters of the wind power inverter and the HVDC converter are designed separately, to meet the performance requirements of the single converter under ideal conditions, but this practice does not guarantee the stability of the interconnected system. Therefore, an optimal design method for controller parameters is proposed in this paper in order to guarantee the small-signal stability of the interconnected system from a system point of view. Finally, time-domain simulations validate the effectiveness of the theoretical analysis and the proposed optimal design method.
The modular multilevel converter (MMC) is attractive for medium- or high-power applications because of the advantages of its high modularity, availability, and high power quality. Reliability is one of the most important challenges for the MMC consisting of a large number of power electronic devices. The diode open-circuit fault in the submodule (SM) is an important issue for the MMC, which would affect the performance of the MMC and disrupt the operation of the MMC. This paper analyzes the impact of diode open-circuit failures in the SMs on the performance of the MMC and proposes a protection scheme for the MMC under diode open-circuit faults. The proposed protection scheme not only can effectively eliminate the possible caused high voltage due to the diode open-circuit fault but also can quickly detect the faulty SMs, which effectively avoids the destruction and protects the MMC. The proposed protection scheme is verified with a downscale MMC prototype in the laboratory. The results confirm the effectiveness of the proposed protection scheme for the MMC under diode open-circuit faults.
For the application of variable speed drive, sub-module (SM) capacitors of the modular multilevel converter (MMC) suffer from large voltage fluctuations at low frequencies. This paper presents a control technique that regulates this fluctuation to a constant value at all speeds with rated load torque. Transformerless configuration of this drive is as follows. An MMC is used as a front-end converter and is directly connected to the grid. It keeps the grid current sinusoidal and in phase with the grid voltage. It generates regulated dc current at the output. Another MMC, connected in cascade with the earlier one, drives a three phase induction motor. Therefore, the input to this rear end MMC is constant dc current, the output is variable magnitude and frequency voltage source that drives the motor. This paper proposes a control strategy for this drive. The proposed control method and its performance have been analyzed and verified through experimental results in this paper.
Capacitors are one type of reliability-critical components in power electronic systems. In the last two decades, many efforts in academic research have been devoted to the condition monitoring of capacitors to estimate their health status. Industry applications are demanding more reliable power electronics products with preventive maintenance. Nevertheless, most of the developed capacitor condition monitoring technologies are rarely adopted by industry due to the complexity, increased cost, and other relevant issues. An overview of the prior-art research in this area is therefore needed to justify the required resources and the corresponding performance of each key method. It serves to provide a guideline for industry to evaluate the available solutions by technology benchmarking, as well as to advance the academic research by discussing the history development and the future opportunities. Therefore, this paper first classifies the capacitor condition monitoring methods into three categories, then the respective technology evolution in the last two decades is summarized. Finally, the state-of-the-art research and the future opportunities targeting for industry applications are given.
In power electronic systems, capacitor is one of the reliability critical components . Recently, the condition monitoring of capacitors to estimate their health status have been attracted by the academic research. Industry applications require more reliable power electronics products with preventive maintenance. However, the existing capacitor condition monitoring methods suffer from either increased hardware cost or low estimation accuracy, being the challenges to be adopted in industry applications. New development in condition monitoring technology with software solutions without extra hardware will reduce the cost, and therefore could be more promising for industry applications. A condition monitoring method based on Artificial Neural Network (ANN) algorithm is therefore proposed in this paper. The implementation of the ANN to the DC-link capacitor condition monitoring in a back-to-back converter is presented. The error analysis of the capacitance estimation is also given. The presented method enables a pure software based approach with high parameter estimation accuracy.
DC-link capacitors are an important part in the majority of power electronic converters which contribute to cost, size and failure rate on a considerable scale. From capacitor users' viewpoint, this paper presents a review on the improvement of reliability of dc link in power electronic converters from two aspects: 1) reliability-oriented dc-link design solutions; 2) conditioning monitoring of dc-link capacitors during operation. Failure mechanisms, failure modes and lifetime models of capacitors suitable for the applications are also discussed as a basis to understand the physics-of-failure. This review serves to provide a clear picture of the state-of-the-art research in this area and to identify the corresponding challenges and future research directions for capacitors and their dc-link applications.
The object of this paper is to present an economic electronic
module integrated on an electrolytic capacitor that is able to indicate
the moment when it must be changed. First, with a switchmode power
supply as an example, the high probability of electrolytic capacitor
failure with respect to other power components is noted. Second, the
authors recall that the increase of the equivalent series resistance
(ESR) of the capacitor is the best indicator of their faulty state. From
the measurements of the voltage ripple and the capacitor current, one
can deduce the ESR; the latter is compared to the ESR value of the sound
capacitor deduced from the component case temperature. Thus, the
capacitor deterioration can be diagnosed
Aluminum electrolytic capacitors (AECs) are the most common capacitors used in power electronics. They are frequently used for filtering and storage functions. However, AECs also represent the major cause for power electronics equipment breakdown, mainly due to their wear out through the vaporization of the electrolyte, due to both aging and temperature effects. The aging of AECs is expressed by the increase of their equivalent series resistance (ESR) and the decrease of their capacitance (C). Therefore, estimating these two parameters can provide a good indicator for a potential capacitor failure. Additionally, these changes in C and ESR parameters strongly influence the ratio between the capacitors voltage ripple and their current ripple. This ratio is dominated by C at low frequencies and by the ESR in the high frequency range. Hence, this paper suggests a method for on-line estimating and tracking of both C and ESR using that ratio, which is computed starting from some always existing harmonics at low and high frequencies. The proposed method uses the Short Time Least Square Prony's (STLSP) technique to determine and track the amplitudes of the target harmonics. It is shown that compared with the Discrete Fourier Transform (DFT) method, the STLSP technique is well suitable for such application, since it has the ability to determine and track the amplitude and frequency of any spectral component, even for noisy and non-stationary signals, using a small number of data samples. The method's effectiveness is proved by simulation and experimental tests using a boost DC-DC converter.
the flying capacitor modular multilevel converter (FC-MMC) overcomes the low/zero-speed operation issue of conventional MMC in medium-voltage motor drive application. However, the introduced flying capacitor (FC) in each phase leg brings great difficulties for circuit startup: 1) pre-charging of FC to nominal voltage is necessary for normal operation; 2) the oscillation between FC and inner buffer inductors need to be mitigated in the pre-charging process. In this paper, a startup method with simple control algorithm is proposed for solving the above two issues. The proposed method operates the switches of submodules (SMs) to make the FC connected in parallel with SM capacitors for charging up. Meanwhile, the oscillation caused by the equivalent LC circuit is effectively mitigated by properly implementing the damping modes. The suppressed oscillation prevents the FC from any negative voltage, which guarantees the safety of FC especially when electrolytic capacitor is utilized. The validation of the proposal is proved by simulations and experiments.
This paper addresses the protection of a high voltage direct current (HVDC) transmission system, utilizing the modular multilevel converter (MMC) topology in addition to incorporating a hybrid transmission corridor (transmission line including overhead line and cable sections). A solution is proposed for identifying the section within which a dc fault is located for the purpose of maintaining power delivery. A detailed model of the MMC-HVDC system is simulated using PSCAD and an in depth fault analysis is performed. A characteristic signal is discovered and then implemented into a novel solution. The end result is a fault section identification protection algorithm, implementing protective relay coordination to protect the system from false circuit breaker reclose as well as enabling fast system restart for nonpermanent faults. This restart protection algorithm is implemented without the use of a communications channel between converter stations, introducing novelty and quick restart response.
This paper employs a Kalman filter (KF) algorithm to estimate the capacitance of submodule (SM) capacitors in modular multilevel converters (MMCs). The proposed method can be used in the health condition monitoring technique of the capacitors. This proposed scheme estimates the capacitance of each individual capacitor from data of arm current and the corresponding voltage across the targeted capacitor. The arm current is used to calculate the capacitor current from the upper state of the SM switch. The proposed method has shown very good performance under normal operating conditions. The effectiveness of the proposed technique is validated via a MATLAB simulation analysis of a single-phase four-level MMC.
In the rapid development process of a high-speed electrified railway, power quality problems in traction power grid have become increasingly deteriorative. In order to ensure a three-phase balanced traction power grid, a modular multilevel converter based railway traction power conditioner (RTPC) is presented. The RTPC consists of four H-Bridge clusters and filter inductors, and it can be directly connected to traction feeders in a co-phase traction system without insulation transformers. According to the equivalent circuit analysis of RTPC system, it can be considered as four single-phase inversion systems. Based on the equivalent control model of each single-phase cluster, the relationship between the multilevel output voltage and current slopes in a control period is analyzed, and an improved model predictive control is proposed. Moreover, a linear combination of two different output levels is proposed to improve tracking performance of cluster current control, and enhance the waveform quality of ac current. Finally, both simulation and experiment results are presented to verify the effectiveness of the structure and its control method.
This paper presents an experimental technique that allows the determination of the reactance and ESR intrinsic values of aluminum electrolytic capacitors. The equivalent circuit of an electrolytic capacitor comprises an equivalent series inductance (ESL), an equivalent series resistance (ESR) and a capacitance (C), whose values change with frequency. Therefore, the behavior of an electrolytic capacitor varies from approximately a simple capacitor for low frequencies, to a resistor at its resonance frequency, and finally to an inductor for frequencies higher than its resonance frequency. The knowledge of the resonance frequency of the electrolytic capacitors, as well as their equivalent circuit at the operating frequency becomes of paramount importance for designing purposes. The ESR is one of the most critical problems of static converters that operate near the resonance frequency of the electrolytic capacitors. An increased ESR value leads to a significant increase in the alternate component of the output voltage, requiring a constant action of the control system. To validate the proposed technique and to demonstrate its applicability some experimental and simulated results are presented
In this paper, reliability evaluation of the capacitor bank for the Modular Multilevel Converter (MMC) based on two common types of the DC-Link capacitors i.e. Aluminum Electrolytic and Film capacitors is investigated. Because of the high current ratio through the capacitor in the MMC, it is shown that in the Film capacitor because of the low ESR, the power loss and the core temperature rising will be lower than the Aluminum Electrolytic Capacitors. Based on three-sample design using ALE and Film capacitors, the reliability function over time is calculated for each case study. The result shows that the Film capacitor bank design is the high reliability and the long life solution in each design. In order to improve life time and reliability of the Aluminum Electrolytic Capacitor, adding Film Capacitor to ALE capacitor bank is presented in this paper.
This paper proposes a condition monitoring scheme of submodule (SM) capacitors in the modular multilevel converters (MMC), which is based on the capacitance estimation using the information of the current and voltage ripples of the capacitors. In order to induce the voltage ripple in the capacitor, a controlled AC current is injected into the circulating current loop through the action of the average voltage control of each SM in the converter leg. The capacitor current is calculated from the arm current and its switching state. By processing these AC voltage and current with digital filters, the capacitance is estimated by a recursive least square (RLS) algorithm. The validity of the proposed scheme has been verified by simulation results for the 1-MW 7-level modular multilevel converter. Also, its feasibility has been proved by experimental results for a reduced-scale prototype.
The grid-connected Modular Multilevel Converter (MMC) is very likely to operate under asymmetric grid conditions. Such a converter features several unique characteristics, which make its analysis different from other types of power converters in similar cases. In this paper, the three well-established control techniques, i.e., balanced current and negative sequence current injection for active/reactive power oscillation elimination, are tailored for the MMC case taking into account grid current limitation under fault conditions. Since the focus is laid on the MMC design impact during grid asymmetries, the three methods are compared in terms of branch energy variation increase as well as maximum achievable active power transfer. Moreover, the effect of circulating current second-order harmonic injection for capacitive storage reduction as well as DC-link oscillation elimination is also studied. Finally, experimental results from a down-scaled laboratory prototype verify the theoretical studies.
The modular multilevel converter (MMC) has been a subject of increasing importance for medium/high-power energy conversion systems. Over the past few years, significant research has been done to address the technical challenges associated with the operation and control of the MMC. In this paper, a general overview of the basics of operation of the MMC along with its control challenges are discussed, and a review of state-of-the-art control strategies and trends is presented. Finally, the applications of the MMC and their challenges are highlighted.
In this paper, a novel scheme for the estimation of the equivalent series resistance (ESR) of the dc-link electrolytic capacitor in three-phase ac/dc pulsewidth-modulation converters is proposed for condition monitoring. First, a controlled ac current component is injected into the input. Then, it induces ac voltage ripples on the dc output. By manipulating these ac voltage and current components with digital filters, the value of the ESR can be calculated, where the recursive least squares algorithm is used for reliable estimation results. In addition, the value of the ESR is corrected by considering the temperature effect, for which a simple temperature-sensing circuit has been designed. The simulation and experimental results show that the estimation error of the ESR is within a reasonable range, thereby enabling the determination of the appropriate time for the replacement of the capacitor.
This paper presents a very simple technique to estimate the condition of aluminum-electrolytic-capacitors. The aging of aluminum-electrolytic-capacitors is expressed by the increase of their equivalent series resistance (ESR) and the reduction of their capacitance. Thus, the proposed technique is based on the estimation of both the ESR and capacitance values. To implement the proposed technique, the capacitor is fed by a sinusoidal voltage with the required power. The resultant impedance vector defined by the capacitor voltage and current gives enough information to estimate the condition of the capacitor.
A novel life-cycle monitoring and voltage-managing device for dc-link electrolytic capacitors in pulsewidth modulation converters is presented. The system performs online identification of the capacitor's equivalent series resistance (ESR) in order to detect the life-cycle status and permit preventive maintenance. The ESR detection is based on the capacitor's ac losses calculated from voltage/current measurements using a simple low-cost microcontroller. The unit is designed as small printed circuit board located directly at the capacitor's screw terminals in order to simplify the required temperature sensing and to minimize wiring effort. The minimized energy consumption allows a power supply taken out of the capacitor to be tested. Besides life-cycle monitoring, the unit further facilitates energy-efficient voltage balancing for capacitors in series arrangements avoiding any power resistors typically used in balancing circuits. Instead, the unit controls the capacitor voltage by influencing its power consumption. Each individual monitoring unit (one per each power capacitor of the converter) transfers the ESR results to the converter's main controller and receives control commands via a common optoisolated fieldbus. Alternatively, this data transfer is performed using wireless near-field communication leading to a completely autonomous monitoring unit without any wiring for power supply and data transmission.
This paper presents an experimental technique that allows the determination of both reactance and equivalent series resistance (ESR) intrinsic values of aluminum electrolytic capacitors.
The ESR is one of the most critical problems of static converters that operate near the resonance frequency of electrolytic capacitors. An increased ESR value leads to a significant increase in the alternate component of the output voltage, requiring
constant action of the control system. In addition, manufacturers usually do not give the ESR value, but do give the maximum dissipation factor (DF), at the operating frequency of 120 Hz.
From the DF, it is possible to determine the maximum ESR value at 120 Hz. Since the ESR value changes with frequency, and static converters operate at much higher ranges, designers need to compute its value at the operating frequency of these converters
to reach the best design proposal. Moreover, almost all capacitor manufacturers give the capacitance value with a tolerance of 20%, whereas the proposed technique permits computation of its value with higher accuracy. Additionally, at low frequencies, if the ESR
value of the capacitor is known, through its comparison with the maximum value given by the manufacturer, it is possible to estimate its condition. To validate the proposed technique and to demonstrate its applicability, some experimental results are presented
and compared with the ones obtained with an impedance gain-phase analyzer and the manufacturers’ specifications for four different capacitors.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication
-8993 (c) 2019 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TPEL.2019.2939185, IEEE
Transactions on Power Electronics
IEEE POWER ELECTRONICS REGULAR PAPER