Article

A Physics-Constrained TD3 Algorithm for Simultaneous Virtual Inertia and Damping Control of Grid-Connected Variable Speed DFIG Wind Turbines

Authors:
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

This paper proposed a physics-constrained twin delayed deep deterministic policy gradient (TD3) algorithm for simultaneous virtual inertia and damping control of a grid-connected variable speed doubly-fed induction generator (DFIG) wind turbine using a combined deep reinforcement learning (DRL) and quadratic programming as a novel solution to suppress frequency fluctuations caused by the control mechanism which decouples the active power from the system frequency, thus hiding the rotating kinetic energy of the wind generator. The optimization stage modifies the action of the DRL agent, thus preventing the agent from taking certain unsafe actions. We tested the effectiveness of the proposed scheme under various scenarios through simulations on an IEEE 9-bus test system in MATLAB/Simulink. Compared with other virtual inertia controls, the results show that the proposed scheme achieved improved dynamic performance with the lowest system frequency deviation and fastest frequency recovery under wind and load variations and severe grid faults. A further test on the IEEE 39-bus system shows that the grid size does not affect the performance of our proposed technique. Note to Practitioners —Integrating the wind turbine systems into the utility grid results in power quality problems such as frequency fluctuation, voltage dip, power loss, and severe power outages. The unpredictability and uncontrollability of the wind pose a serious problem in integrating wind energy conversion systems. This problem becomes worse with the increasing number of connected wind turbines. Therefore, new control strategies are required to mitigate this issue. The droop-based virtual and damping control method traditionally provides frequency support in grid-tied wind turbine systems. However, the fixed droop gain is a significant drawback of this method. In this paper, we proposed a novel physics-constrained deep-reinforcement learning-based virtual inertial and damping control. The proposed control agent is constrained from unsafe actions and rewarded for maintaining the grid frequency within operational limits. Simulation results with the IEEE 9 bus system validated the feasibility and effectiveness of our proposed approach. A comparison of our method with the conventional control scheme, adaptive droop-based virtual inertia control, etc., carried out under various operational scenarios verified the enhanced performance of our proposed strategy.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... Further, the training process aims to minimize this loss function by adjusting the model's weights using optimization algorithms, such as Stochastic gradient descent (SGD) or Adam. These optimization techniques help in minimizing the difference between the predicted and actual stability states, improving the model's ability to accurately classify V/F stability [62][63][64][65][66]. ...
Article
Full-text available
Frequency response of wind turbine generators (WTGs) is a crucial technology to ensure safe integration of increasing future wind power into power grids. WTGs' regulation capabilities are variable compared to conventional generations due to uncertain wind speed and dynamic deloading factors, which affects the design of their frequency response strategy. This paper proposes a novel deep reinforcement learning (DRL)‐based frequency response framework of WTGs, combining a deep deterministic policy gradient (DDPG)‐based control and a proportional power sharing (PPS) strategy. The DDPG‐based control dynamically determines the optimal total regulation power of WTGs and the PPS strategy allocates the total regulation power to each WTG. Benefiting from the ability of DDPG algorithm to handle multi‐dimensional continuous system states and actions, the proposed DDPG‐control enables a wind farm to dynamically determine the optimal total frequency regulation power under WTGs' variable regulation capabilities. Besides, the application of deep neural networks (DNN) in DDPG algorithm makes the DDPG‐based control possible to be implemented in complex nonlinear systems. Simulations on an IEEE 39‐bus system demonstrate that effectiveness of the proposed frequency response strategy of WTGs.
Article
Full-text available
Due to the shortage and environmental pollution of fossil fuel sources, conventional generating units are substituted with renewable energy sources. Thus, the total inertia decreases while the rate of change of frequency (ROCOF) increases, affecting the system frequency and stability. In this paper, a supplementary control, washout filter, is added to the primary frequency control of a doubly-fed induction generator (DFIG)- based wind turbine for better frequency regulation. In order to achieve the best performance (lowest frequency deviation with keeping the system stable), a new optimization technique called Mayfly Algorithm (MA) is used to find the optimum values of the washout filter’s parameters. Compared with the traditional method, a great improvement has been presented in the system frequency response under different load perturbations. Also, the results show that the washout filter’s parameters have an adverse impact on the frequency response and system stability if they aren’t determined accurately. MATLAB/SIMULINK program is used to test the performance of the improved method under different scenarios.
Article
Full-text available
With large-scale integration of renewable generation and distributed energy resources, modern power systems are confronted with new operational challenges, such as growing complexity, increasing uncertainty, and aggravating volatility. Meanwhile, more and more data are becoming available owing to the widespread deployment of smart meters, smart sensors, and upgraded communication networks. As a result, data-driven control techniques, especially reinforcement learning (RL), have attracted surging attention in recent years. This paper provides a comprehensive review of various RL techniques and how they can be applied to decision-making and control in power systems. In particular, we select three key applications, i.e., frequency regulation, voltage control, and energy management, as examples to illustrate RL-based models and solutions. We then present the critical issues in the application of RL, i.e., safety, robustness, scalability, and data. Several potential future directions are discussed as well.
Article
Full-text available
As the increasing of wind power penetration, the fluctuations of wind power results in serious power system frequency turbu-lences. This paper proposes a low-pass virtual filter (VF) for wind energy conversion systems (WECSs) to smoothen output power. Firstly, a linearized model is established for a WECS. Then, a low-pass VF is derived for the active power control loop in WECS to modulate the kinetic energy stored in rotating mass for output smoothing. The principle is to modify the traditional wind power controller to emulate a physical energy storage system controlled as a low-pass filter to smooth the wind power output. The pro-posed low-pass VF does not need the physical deployment of energy storage system but still can filter the high-frequency fluctuations in the output wind power. Furthermore, in order to ensure operation stability of the wind turbine (WT), a stabil-ity-constrained coefficient is determined according to the pro-posed stability criteria to ensure the stability of WTs. Simulation and experiment results verifies the effectiveness of the proposed VF in WECS for output wind power fluctuation reduction and the stability-constrained coefficient for stability ensuring.
Article
Full-text available
We investigate the important problem of certifying stability of reinforcement learning policies when interconnected with nonlinear dynamical systems. We show that by regulating the partial gradients of policies, strong guarantees of robust stability can be obtained based on a proposed semidefinite programming feasibility problem. The method is able to certify a large set of stabilizing controllers by exploiting problem-specific structures; furthermore, we analyze and establish its (non)conservatism. Empirical evaluations on two decentralized control tasks, namely multi-flight formation and power system frequency regulation, demonstrate that the reinforcement learning agents can have high performance within the stability-certified parameter space and also exhibit stable learning behaviors in the long run.
Article
Full-text available
Large inrush currents emerge abruptly whenever a doubly fed induction generator (DFIG) attempts to connect with the grid. Moreover, a DFIG is highly vulnerable to grid disturbances owing to its direct stator-grid contact, which induces a sizable electromotive force in its rotor circuit. Therefore, to ensure reliable and stable system operations, it is unavoidable to improve system performance through suitable and corrective control actions to counteract such issues effectively. In this context, this study focuses on establishing a stable and smooth grid connection for a DFIG by developing dynamic models of the rotor-side converter controls that generate suitable rotor-current references for both the synchronizing and running modes of a DFIG. Furthermore, a smooth transition between both modes is also precisely made with trivial inrush currents. Dynamic simulations confirm the validity of the proposed synchronization method in MATLAB/SIMULINK. Finally, a comparative study with a conventional synchronization method is also performed.
Article
Full-text available
The recent power cut incident in the UK indicated that frequency control to reduce frequency nadir and eliminate frequency second dip is highly desirable for power grids with high penetration of wind energy. This paper proposes a fast frequency support scheme for wind turbine systems (WTSs) that can enable frequency nadir to be significantly reduced and close to the settling frequency and eliminate frequency second dip. In the proposed frequency support scheme, in order to achieve similar frequency support performance and ensure stability of WTSs under varying wind speeds, different levels of wind power penetration and system conditions, an adaptive gain, which is a function of real-time rotor speed and wind power penetration level, is proposed. In the proposed scheme, rotor speeds of WTSs are proposed not to be recovered to the optimal operating points during the primary frequency control, but recovered during the secondary frequency control. Simulation results on the IEEE two-area power system with a doubly fed induction generator (DFIG)-based wind farm and the IEEE 39-bus power system with permanent magnetic synchronous generator (PMSG)-based wind farms using real-time digital simulator (RTDS) and Dymola are presented to verify the effectiveness of the proposed scheme.
Article
Full-text available
Due to the influence of wind speed disturbance, there are some uncertain phenomena in the parameters of the nonlinear wind turbine model with time in an actual working environment. In order to mitigate the side effects of uncertainties in speed models of wind turbines, researchers have designed a variety of controllers in recent years. However, traditional control methods require more knowledge of dynamics. Therefore, based on reinforcement learning and system state data, a robust wind turbine controller that adopts adaptive dynamic programming (ADP) is proposed. The ADP algorithm is a combination of temporal-Difference (TD) algorithm and actor-critic structure, which can guarantee the rotor speed is stable around the rated value to indirectly adjust the wind energy utilization coefficient by changing the pitch angle in the area of high wind speed and achieve online learning in real-time. In addition, the variation of the pitch angle command of the proposed controller is relatively gradual, which can reduce the energy consumption of the variable pitch actuator, and extend the service life of the equipment. Finally, the wind speed model is simulated by combined wind speed based on Weibull distribution, the comprehensive simulation results show that the proposed controller has better control effect than some existing ones.
Article
Full-text available
The present study applies reinforcement learning strategy for controller design to improve the low voltage ride through (LVRT) capability of a hybrid power system through convertible static compensator (CSC). The study considers different configurations of CSC such as static synchronous series compensator (SSSC), one static synchronous compensator (STATCOM), two STATCOMs, unified power flow controller (UPFC). Both Q-learning and dynamic fuzzy Q-learning (DFQL) based controllers are designed and their performances were compared with classical PID controller considering a 3-machine system consisting of 2-synchronous and 1-wind energy systems. The results of simulation show that the performance of DFQL based controller is better compared to other 2-controllers in improving the LVRT capability. Further, it is shown that the UPFC and two STATCOMs configurations of CSC provide higher voltage support compared to other configurations.
Article
Full-text available
When a large amount of wind power is connected to the power grid, certain control methods need to be taken for wind turbines (WTs) so that WTs can respond to system frequency changes and maintain system frequency stability. Based on the frequency control method of a single doubly fed induction generator, a deloading power coordinated distribution method for frequency regulation by wind farms considering wind speed differences is proposed to better utilize the wind farm frequency regulation ability. A reasonable distribution of the deloading power of a wind farm for participation in grid frequency regulation to better utilize the wind farm frequency regulation ability. A quantitative calculation formula for the wind speed, total deloading power and single WT deloading power is established by using a variable parameter deloading power coordinated distribution coefficient based on the wind speed. The deloading power is distributed according to the wind speed of the WT units to adapt to the wind speed distribution differences between WTs in cluster wind farms and fully utilize the wind farm frequency regulation ability. A typical doubly fed induction generator (DFIG) wind farm is used as a simulation case. The experimental results show that the control strategy can distribute the output of each WT, improve the frequency response of the wind farm and enhance the frequency stability of the system.
Article
Full-text available
Electric vehicles (EVs) have experienced a rapid growth due to the economic and environmental benefits. However, the substantial charging load brings challenging issues to the power grid. Modern technological advances and the huge number of high-rise buildings have promoted the development of distributed energy resources, such as building integrated/mounted wind turbines. The issue to coordinate EV charging with locally generated wind power of buildings can potentially reduce the impacts of EV charging demand on the power grid. As a result, this paper investigates this important problem and three contributions are made. First, the real-time scheduling of EV charging is addressed in a centralized framework based on the ideas of model predictive control, which incorporates the volatile wind power supply of buildings and the random daily driving cycles of EVs among different buildings. Second, an EV-based decentralized charging algorithm (EBDC) is developed to overcome the difficulties due to: 1) the possible lack of global information regarding the charging requirements of all EVs and 2) the computational burden with the increasing number of EVs. Third, we prove that the EBDC method can converge to the optimal solution of the centralized problem over each planning horizon. Moreover, the performance of the EBDC method is assessed through numeric comparisons with an optimal and two heuristic charging strategies (i.e., myopic and greedy). The results demonstrate that the EBDC method can achieve a satisfactory performance in improving the scalability and the balance between the EV charging demand and wind power supply of buildings.
Conference Paper
Full-text available
Control theory can establish properties of systems which hold with all signals within the system and hence cannot be proven by simulation. The most basic of such property is the stability of a control subsystem or the overall system. Other examples are statements on robust control performance in the face of dynamical uncertainties and disturbances in sensing and actuation. Until now these theories were developed and checked for their correctness by control scientist manually using their mathematical knowledge. With the emergence of formal methods, there is now the possibility to derive and prove robust control theory by symbolic computation on computers. There is a demand for this approach from industry for the verification of practical control systems with concrete numerical values where the applicability of a control theorem is specialised to an application with given numerical boundaries of parameter variations. The paper gives an overview of the challenges of the area and illustrates them on a computer-based formal proof of the Small-gain theorem and conclusions are drawn from these initial experiences.
Conference Paper
Full-text available
Frequency regulation in interconnected networks is one of the main challenges posed by wind turbines in modern power systems. The wind power fluctuation negatively contributes to the power imbalance and frequency deviation. This paper presents an intelligent agent based load frequency control (LFC) for a multi-area power system in the presence of a high penetration of wind farms, using multi-agent reinforcement learning (MARL). Nonlinear time-domain simulations on a 39-bus test power system are used to demonstrate the capability of the proposed control scheme.
Article
The high system complexity and strong wake effects bring significant challenges to wind farm operations. Conventional wind farm control methods may lead to degraded power generation efficiency. A reinforcement learning (RL)-based approach is proposed in this paper to handle these issues, which can increase the long-term farm-level power generation subject to strong wake effects while without requiring analytical wind farm models. The proposed method is significantly distinct from existing RL-based wind farm control approaches, whose computational complexities usually increase heavily with the increase of total turbine numbers. In contrast, our method can greatly reduce training loads and enhance learning efficiency via two novel designs: (1) automatic grouping and (2) multi-agent-based transfer learning (MATL). Automatic Grouping can divide a large wind farm into small turbine groups by analyzing the aerodynamic interactions between turbines and utilizing some key principles from the graph theory. It enables the separated conduction of RL algorithms on small turbine groups, avoiding the complex training process and high computational costs of applying RL on the entire farm. Based on Automatic Grouping, MATL can further reduce the computational complexity by allowing agents (i.e. wind turbines) to inherit control policies under potential group changes. Case studies with a dynamical simulator show that the proposed method achieves clear power generation increases than the benchmark. It also dramatically reduces computational costs compared with typical RL-based wind farm control methods, paving the way for the application of RL in general wind farms.
Article
Reduction in system inertia and maintaining the frequency at the nominal value is a staple of today's and future power systems since their operation, stability, and resiliency are degraded by frequency oscillation and cascading failures. Consequently, designing a stable, scalable, and robust virtual inertia control system is highly relevant to skillfully diminishing the deviations during major contingencies. Therefore, considering the potential problems in predesigned nonflexible control systems with offline tuning techniques, we propose a variable fractional-order PID controller for virtual inertia control applications, which is tuned online using a modified neural network-based algorithm. The new proposed tuner algorithm is trained using a deep reinforcement learning strategy with a simplified deep deterministic policy gradient, which considers microgrid uncertainties. Compared with existing methods, all the tuning knobs of the discrete type and fully tunable variable FOPID controller (for both gain and order) can be captured based on the proposed hybrid algorithm, which inherits features from both classical and advanced techniques. To demonstrate the effectiveness of the training of the proposed controller, a comparative analysis with the standard FOPID and PID controllers is given under three different scenarios with a smooth (dis)connection of renewable energy sources and loads.
Article
Environmental and energy security concerns lead to the continuous displacement of traditional fossil fuel-based power generation to power electronics interfaced distributed generations (DGs). Increasing penetration of renewable energy sources and the substitution of synchronous generators with power electronic converters have resulted in reduced power system inertia and damping. This causes large frequency deviations and a higher rate of change of frequency. The fast and flexible nature of the energy storage system with a designed controller can achieve frequency stability in low inertia microgrids (MGs). The conventional proportional-integral-based virtual inertia controller is unable to eliminate frequency instability in low inertia MG. To enhance frequency stability, this article proposes a virtual inertia emulation strategy using a twin delayed deep deterministic policy gradient (TD3) algorithm for fast frequency regulation of MGs with inverter-based DGs. A comparative analysis of the TD3 scheme and the conventional method is made. The results show that the TD3-based virtual inertia control provides better tracking with a 56.6% reduction in frequency deviation and faster transient recovery than the conventional virtual inertia control against a broad range of operational scenarios. Performance metrics and simulation results are shown to demonstrate the feasibility of the proposed control scheme.
Article
In this study, two types of methods from different perspectives based on spectral normalization (SN) are described for ensuring the stability of a feedback system controlled by a neural network (NN). The first one is that the L2 gain of the feedback system is bounded less than 1 to satisfy a stability condition derived from the small-gain theorem. When explicitly including the stability condition, the first type of method may provide an insufficient performance on the NN controller due to its strict stability condition. To overcome this difficulty, the second type of method is proposed, ensuring local stability with a larger region of attraction. In this second type, the stability is ensured by solving linear matrix inequalities after training the NN controller. SN improves the feasibility of the a posteriori stability test by constructing tighter local sectors. Numerical experiments show that the second type of method provides sufficient performance compared with the first one and ensures sufficient stability compared with existing reinforcement learning algorithms.
Article
As wind turbines (WTs) become more prevalent, there is an increasing interest in actively controlling their power output to participate in the frequency regulation for the power grid. Conventional frequency regulation controllers use fixed gains, making it difficult for the WT to adjust its kinetic energy uptake to its operating conditions and to collaborate effectively with other WTs in the wind farm. In addition, the design of conventional frequency controllers does not consider their impacts on the mechanical structure. To address these issues, in this article, we model the cooperative frequency control problem for all the WTs in a wind farm as a decentralized partially observable Markov decision process and use a multiagent deep reinforcement learning algorithm to solve it. We also develop a grid-connected wind farm simulation model based on MATLAB/Simulink and OpenFAST, which can reflect the detailed interactions between the electrical and mechanical components of WTs. Simulation results show that the proposed strategy is effective in reducing frequency drops and has less impact on mechanical structure deflections compared with traditional methods.
Article
With the large-scale wind power integration into the power system, the inertia and damping characteristics of the system have been weakened, which is not conducive to fulfill the grid-friendly operation of wind power. Aiming at this, an adaptive control scheme combining virtual inertia control (VIC) and supplementary damping controller (SDC) is proposed in this paper. Firstly, the mathematical model of wind grid-connected power system with VIC is established. Secondly, the polytopic linear parameter varying (LPV) system is built to describe the uncertainty of system operating points, and the gap metric-based nonlinear evaluation method is used for polytope vertex configuration. Next, Youla-Kucera parametrization are extended to polytopic LPV system to design an adaptive damping control, in which VIC and SDC are coordinated to provide damping and virtual inertia simultaneously. Finally, the proposed adaptive control scheme is compared with several existing control techniques to validate its effectiveness and advantages in suppressing system oscillation in the modified IEEE 39-bus system integrated with 2 wind turbines.
Article
Emergency frequency control is one of the most critical approaches to maintain power system stability after major disturbances. With the increasing number of grid-connected renewable energy sources, existing model-based methods of frequency control are facing up with challenges of computational speed and scalability for large-scale systems. In this paper, the emergency frequency control problem is formulated as a Markov Decision Process (MDP) and solved through a novel Distributional Deep Reinforcement Learning (DDRL) method, namely the distributional soft actor critic (DSAC) method. Compared with other RL methods that only estimate the mean value, the proposed DSAC model estimates the distribution of value function over returns. This advancement can lead to more insights and knowledge for the agent, with the benefit of a much faster and more stable learning process, and the improved frequency control performance. The simulation results on IEEE 39-bus and IEEE 118-bus systems demonstrate the effectiveness and robustness of proposed models, as well as the advantage compared to other state-of-the-art DRL algorithms.
Article
The control performance of conventional analytic algorithms for double-fed induction generator-based wind turbines is a fixed feature, which needs to be optimized by optimization processes. To avoid the optimization processes and update control strategies online, this paper proposes an online control algorithm based on the quantum process, deep belief networks, and reinforcement learning for double-fed induction generator-based wind turbines. The proposed approach is named quantum deep reinforcement learning, consisting of deep belief networks, one reinforcement learning framework, and multiple subsidiary reinforcement learning parts with quantum processes. The quantum deep reinforcement learning can update the control strategy online with general initialization for dynamic systems, avoid optimal local solutions, and predict the next systemic states of double-fed induction generator-based wind turbines. The proportional–integral–derivative, fractional-order proportional–integral–derivative, active disturbance rejection controller, reinforcement learning, and the quantum deep reinforcement learning are compared under four cases, i.e., variable wind speed with step magnitude, variable wind speed with sine magnitude, voltage dropout of power grids, and both variable wind speed with random magnitude and voltage dropout of power grids. Consequently, the proposed quantum deep reinforcement learning can obtain better control performance for double-fed induction generator-based wind turbines with smaller integrated absolute error, integral squared error, and integrated time-weighted absolute error of the control error than other compared methods.
Article
In this article, a deep reinforcement learning (RL)-based control approach with enhanced learning efficiency and effectiveness is proposed to address the wind farm control problem. Specifically, a novel composite experience replay (CER) strategy is designed and embedded in the deep deterministic policy gradient (DDPG) algorithm. CER provides a new sampling scheme that can mine the information of stored transitions in-depth by making a tradeoff between rewards and temporal difference (TD) errors. Modified importance-sampling weights are introduced to the training process of neural networks (NNs) to deal with the distribution mismatching problem induced by CER. Then, our CER-DDPG approach is applied to optimizing the total power production of wind farms. The main challenge of this control problem comes from the strong wake effects among wind turbines and the stochastic features of environments, rendering it intractable for conventional control approaches. A reward regularization process is designed along with the CER-DDPG, which employs an additional NN to handle the bias of rewards caused by the stochastic wind speeds. Tests with a dynamic wind farm simulator (WFSim) show that our method achieves higher rewards with less training costs than conventional deep RL-based control approaches, and it has the ability to increase the total power generation of wind farms with different specifications.
Article
A model-free deep reinforcement learning (DRL) method is proposed in this paper to maximize the total power generation of wind farms through the combination of induction control and yaw control. Specifically, a novel double-network-based DRL approach is designed to generate control policies for thrust coefficients and yaw angles simultaneously and separately. Two sets of critic-actor networks are constructed to this end. They are linked by a central power-related reward, providing a coordinated control structure while inheriting the critic-actor mechanism's advantages. Compared with conventional DRL methods, our method can adapt to the distinctive and incompatible features of different control inputs, guaranteeing a reliable training process and ensuring superior performance. Also, the prioritized experience replay strategy is utilized to improve the training efficiency of deep neural networks. Simulation tests based on a dynamic wind farm simulator show that the proposed method can increase the power generation for wind farms with different layouts.
Article
This article aims to address the wind-farm power tracking problem, which requires the farm's total power generation to track time-varying power references and, therefore, allows the wind farm to participate in ancillary services such as frequency regulation. A novel preview-based robust deep reinforcement learning (PR-DRL) method is proposed to handle such tasks which are subject to uncertain environmental conditions and strong aerodynamic interactions among wind turbines. To our knowledge, this is for the first time that a data-driven model-free solution is developed for wind-farm power tracking. Particularly, reference signals are treated as preview information and embedded in the system as specially designed augmented states. The control problem is then transformed into a zero-sum game to quantify the influence of unknown wind conditions and future reference signals. Built upon the HH_\infty control theory, the proposed PR-DRL method can successfully approximate the resulting zero-sum game's solution and achieve wind-farm power tracking. Time-series measurements and long short-term memory networks are employed in our DRL structure to handle the non-Markovian property induced by the time-delayed feature of aerodynamic interactions. Tests based on a dynamic wind-farm simulator demonstrate the effectiveness of the proposed PR-DRL wind-farm control strategy.
Article
This paper proposes a frequency-smoothing scheme of a doubly-fed induction generator (DFIG) that can significantly eliminate the frequency fluctuations caused by continuously fluctuating wind speeds. To achieve this, a proportional-integral (PI) control loop depending on the frequency deviation instead of a proportional control loop used in conventional schemes is added to the maximum power point tracking loop. The P control loop in the proposed frequency-smoothing scheme eliminates the alternating component in the frequency deviation, whereas the I control loop eliminates the slowly-varying component. The gains for the P control loop and I control loop are suggested for effectively using the rotating masses of a DFIG while avoiding over-deceleration of the rotor speed. The simulation results evidently demonstrate that the proposed scheme nearly removes the frequency deviation under various wind conditions, by using the proposed PI control loop, especially in higher penetration levels of wind.
Article
The load frequency control (LFC) aims to keep the frequency fluctuation of the power grids within certain specified limits, under various load disturbances. However, with the increased usage of renewable energy sources (RESs) in smart grids, it is essential to regulate the conventional power plants, based on renewable energy penetration levels. Moreover, with the decentralized nature of the control operation in smart grids, the communication network between the control center and actuator faces the challenge of random communication delays and packet drops in the form of cyberattacks. In this article, the conventional thermal power plant operations within an LFC have been modified using energy storage elements with an emphasis on maximizing the RES utilization while tackling the problems associated with cyber-physical systems, such as packet drops and random time delays. A filtered proportional–integral–derivative (PID) controller is tuned in the LFC using the particle swarm optimization (PSO) algorithm, including random time delays and cyberattacks modeled as random packet drops. The tuned PID control performance in the LFC scheme is tested with synthetic stochastic as well as real profiles of RES and load demands. The numerical analysis has been conducted on two-area LFC model with Monte Carlo simulations of stochastic demand and generation profiles. Note to Practitioners —We are moving toward more renewable energy-based cyber-physical power grids, and it is becoming increasingly important to understand the limits of the balance between more renewable energy integration versus changing the prime-mover in thermal power generation units to meet uncertain load demands. This article proposes a new load frequency control (LFC) scheme employing a nonlinear dead-zone element between the control signals and the actuators (prime movers), which allows the utilization of the available renewable energy sources (RESs) to meet the load and then send a set-point change command in the thermal power generators if the RES does not meet the load. The robustness of the smart grid stability is also verified with the denial-of-service (DoS)-type cyberattack, which is represented in the form of high probability of packet drops and stochastic time delays in the communication channels between the control center and the generation units. It is also essential to understand how different stochastic profiles may affect the stability and performance of the tuned LFC loops, under random time delays and packet dropouts. These are quantified using the uncertainty bounds of grid frequency, its rate of change, control inputs, and power exchange between the two areas that are analyzed using Monte Carlo simulations with different types of nonstationary load and RES profiles and also using real data to show the effectiveness of the LFC scheme with communication constraints and the resulting imperfections.
Article
With increasing levels of wind generation in power systems, guaranteeing continuous power and system’s safety is essential. Frequency control is critical which requires a supplementary inertial control strategy. Since wind power generation depends directly on wind conditions, this creates an immense challenge for a conventional inertial controller with parameters suitable for all power grid operations and wind speed conditions. Therefore, tuning the controller gains is absolutely critical for an integrated conventional/renewable power system. Here, a fuzzy-logic adaptive inertial controller scheme for online tuning of the proportional-derivative-type (PD) inertial controller parameters is proposed. The proposed controller adapts the control parameters of the supplementary inertial control of the doubly fed induction generator (DFIG) wind turbine so that with any disturbance such as load changes, the active power output can be controlled to mitigate the frequency deviation. Simulation results indicate that the proposed adaptive controller demonstrates a more consistent and robust response to load changes compared to a conventional controller with fixed parameters.
Article
As wind energy is becoming a significant utility source, minimizing the Operation and Maintenance (O&M) expenses has raised a crucial issue to make wind energy competitive to fossil fuels. Wind Turbines (WTs) are subject to unexpected failures due to operational and environmental conditions, aging, etc. An accurate estimation of time to failures assures reliable power production and lower maintenance costs. In recent years, a notable amount of research has been undertaken to propose prognosis techniques that can be employed to forecast the Remaining Useful Life (RUL) of wind farm assets. This paper provides a recent literature review on modeling developments for the remaining useful life prediction of critical WT components including physics-based, Artificial Intelligence (AI)-based, stochastic-based, and hybrid prognostics. In particular, the pros and cons of the prognosis models are investigated to assist researchers in selecting proper models for certain applications of WT RUL forecast. Our comprehensive review has revealed that hybrid methods are now the leading and most accurate tools for WT failure predictions over individual hybrid components. Strong candidates for future research include the consideration of variable operating environments, component interaction, physics-based prognostics, and Bayesian application in the development of WT prognosis methods.
Article
A new DC current-voltage droop control method for a hybrid high-voltage direct-current (HVDC) system featuring a voltage-sourced converter (VSC) rectifier and a line-commutated converter (LCC) inverter is presented. The proposed method utilizes DC voltage droop control to reduce the AC voltage fluctuation at the inverter side, to exploit the hybrid HVDC system for connecting offshore wind farms to weak grids without necessitating continuous reactive power compensation facilities. Additionally, an optimal droop coefficient calculation method is proposed to keep the AC voltage constant for wind power fluctuations. To evaluate the stability of the proposed method, a small-signal state-space (SS) model of the target system is derived and the root locus analysis is given. Simulation case studies are performed using an electromagnetic transient model in PSCAD and an SS model in MATLAB. The results demonstrate that a hybrid HVDC system with the proposed method successfully integrates offshore wind energy and weak grids without additional facilities because the AC voltage can be maintained at a near constant during wind fluctuations.
Article
This paper presents an intelligent wind speed sensor less maximum power point tracking (MPPT) method for a variable speed wind energy conversion system (VS-WECS) based on a Q-Learning algorithm. The Q-Learning algorithm consists of Q-values for each state action pair which is updated using reward and learning rate. Inputs to define these states are electrical power received by grid and rotational speed of the generator. In this paper, Q-Learning is equipped with peak detection technique, which drives the system towards peak power even if learning is incomplete which makes the real time tracking faster. To make the learning uniform, each state has its separate learning parameter instead of common learning parameter for all states as is the case in conventional Q-Learning. Therefore, if half learned system is running at peak point, it does not affect the learning of unvisited states. Also, wind speed change detection is combined with proposed algorithm which makes it eligible to work for varying wind speed conditions. In addition, the information of wind turbine characteristics and wind speed measurement is not needed. The algorithm is verified through simulations and experimentation and also compared with perturbation and observation (P&O) algorithm.
Article
Cooperative wind farm control is a complex problem due to wake effect and it is hard to find the proper model. Reinforcement learning can find the optimal policy in a dynamic environment using trial-and-error, but may damage the machine and cause high cost during the learning process. In order to address this challenge, this paper proposes the knowledge assisted reinforcement learning framework by combining the low-fidelity analytical model with reinforcement learning framework. Moreover, the Knowledge Assisted Deep Deterministic Policy Gradient (KA-DDPG) algorithm and three kinds of knowledge assisted learning methods are proposed based on the framework. The proposed methods are tested in four different scenarios of WFSim. The simulation results show that the KA-DDPG algorithm can reach the maximum power output and ensure safety during learning. In addition, the learning cost is reduced by accelerating the learning process.
Article
Conjoint participation of wind generation with conventional power plant, necessitate wind-farms to participate in automatic generation control(AGC) for frequency regulation. This implicates wind farms operator to monitor commands received from the transmission system operator(TSO). Consequently, advanced control techniques are deployed, which assuredly helps in power tracking, but increases pitch angle controller dynamics. This has resulted in mechanical stress on equipment involved. In order to improve wind-farms response, a synergistic frequency regulation control mechanism(SFRCM) is proposed by the authors. The scheme considers the response time and reserve availability depending on forecasted wind data and also examine load curve pattern to obtain the reference signal for the wind-farm controllers. The work provides a distinct solution to address the following: 1) Optimal pitch dynamics regulated operating point tracking with revised-pitch angle control(R-PAC), 2) Maximization of rotational kinetic energy viz attuned-rotor speed control(A-RSC), to increase stored kinetic energy in the rotor. Case studies at constant and variable wind speed are presented to show the effectiveness of proposed algorithm. To test robustness of the technique, transient operation is also conducted. The simulations, performed in the real time environment, depict the fulfillment of the above objective in an efficient manner compared with other conventional approaches.
Article
This paper proposes a variable proportion coefficient based control scheme for the doubly-fed induction generator based wind turbines (DFIG-WTs) to implement frequency support by regulating rotor speed. To reduce the adverse impact caused by regulating DFIG-WT on the dynamic characteristic of the grid frequency, a two-stage switching control scheme is employed for the proposed method. In the first stage, the variable proportion coefficient is designed for emulating the virtual inertia of the DFIG-WT to provide inertia response. In the second stage, a fuzzy control scheme is employed to design the variable proportion coefficient for both quickly restoring the maximum power point tracking (MPPT) operation of DFIG-WTs and avoiding the secondary frequency dip to system. Case studies are undertaken based on WSCC 9-bus and IEEE 39-bus power systems, respectively. Simulation results show that the proposed control scheme can not only make grid-connected DFIG-WTs provide the friendly frequency support, but also help them to fully use the frequency regulation ability of synchronous generators for quickly restoring the MPPT operation.
Article
The wind turbine ( WT ) is a renewable energy conversion device for transformation of kinetic energy from the wind to mechanical energy for subsequent use in different forms. This paper focuses on wind turbine control design strategies. The content is divided into the following parts: 1 ) An overview of the recent advances that have been made in the application of adaptive and model predictive control strategies for wind turbines. 2 ) Summarizes some important aspects of modeling of wind turbines for control studies. 3 ) Provides an outlook on the application of adaptive model predictive control for uncertain systems to stimulate new research interests for wind turbine systems. We provide an overall picture of the research results with evaluation of the merits / demerits.
Article
In the literature, many droop control loops have been proposed to enhance the frequency control capability of the wind farm (WF) when a large frequency disturbance occurs. However, some loops are not feasible to be implemented in the dynamic equivalent model (DEM) of the WF. The kinetic energy (KE) based droop control loop is an example. The droop gain of this loop is expressed as a quadratic function of the wind turbine (WT) rotor speed. However, it is not feasible to implement a nonlinear function in DEM of WF. Therefore, in this study, a new linear-gain droop control loop is proposed for the doubly-fed induction generator (DFIG) based WF. In the proposed control loop, the droop gain is a linear function of the WT rotor speeds. By selecting the proper coefficients of the linear function, the proposed linear droop gain can achieve a good approximation to the quadratic droop gain. The performance of the proposed droop control loop is demonstrated based on three initial conditions. To verify the responses of system frequency and WF power output, four indices are developed. The simulation results demonstrate that the proposed linear-gain droop control loop is capable of approximating closely to the KE-based droop control loop.
Article
This paper proposes an optimal efficiency control scheme for a wind system with doubly fed induction generator (DFIG). The suggested control scheme combines loss minimization (LM) in the DFIG and maximum power point tracking (MPPT) in the wind turbine and therefore maximum electrical energy generation, by the same wind energy potential, is achieved. Moreover, since the cut-in wind speed is reduced, extension of the exploitable wind speed range toward the lower speed region is attained. The LM is achieved by properly controlling the flux-linkage of the DFIG with respect to the stator current and the MPPT is accomplished by regulating the rotor speed through the rotor current. The parameters of the LM and MPPT controllers can be determined experimentally and thus, the knowledge of the wind energy conversion system (WECS) model is not required. For the implementation of the proposed control strategy, a new structure of the WECS has been adopted. However, the hardware requirements of the WECS and considerably the cost has not been considerably affected compared to the conventional configuration. Selective simulation and experimental results are presented to validate the effectiveness of the proposed control strategy and demonstrate the operational improvements. IEEE
Article
This paper proposes a fuzzy-based speed controller for the doubly-fed induction generator (DFIG)-based wind turbines with the rotor speed and wind speed inputs. The controller parameters are optimized using the particle swarm optimization (PSO) algorithm. To accelerate tracking the maximum power point (MPP) trajectory, the conventional controller is augmented with a feed-forward compensator, which uses the wind speed input and includes a high-pass filter. The proposed combined speed controller is robust against wind measurement errors and as the accuracy of anemometers increases the speed regulation tends towards the ideal controller. The cutoff frequency of the applied filter is determined considering a compromise between the sensitivity to measurement errors and speed of regulation process. We also design an auxiliary frequency controller to equip the DFIGs with an inertial frequency response. In the proposed controller, two important constraints are taken into account: the feasible rotor speed range during the injection period, and the minimum time to recover the DFIG's speed. The impacts of the proposed controllers are evaluated through extensive time domain simulations on an IEEE 9-bus test system using the DIgSILENT/PowerFactory software. Results confirm the effectiveness of the proposed controllers in serious transients and load disturbances. Index Terms—Doubly-fed induction generator (DFIG), fuzzy logic, speed controller, auxiliary frequency controller, inertial response.
Article
Preservation of the environment has become the main motivation to integrate more renewable energy sources (RESs) in electrical networks. However, several technical issues are prevalent at high level RES penetration. The most important technical issue is the difficulty in achieving the frequency stability of these new systems, as they contain less generation units that provide reserve power. Moreover, new power systems have small inertia constant due to the decoupling of the RESs from the AC grid using power converters. Therefore, the RESs in normal operation cannot participate with other conventional generation sources in frequency regulation. This paper reviews several inertia and frequency control techniques proposed for variable speed wind turbines and solar PV generators. Generally, the inertia and frequency regulation techniques were divided into two main groups. The first group includes the deloading technique, which allow the RESs to keep a certain amount of reserve power, while the second group includes inertia emulation, fast power reserve, and droop techniques, which is used to release the RESs reserve power at under frequency events.
Article
If a large disturbance occurs in a power grid, two auxiliary loops for the inertial control of a wind turbine generator have been used: droop loop and rate of change of frequency (ROCOF) loop. Because their gains are fixed, difficulties arise in determining them suitable for all grid and wind conditions. This paper proposes a dynamic droop-based inertial control scheme of a doubly-fed induction generator (DFIG). The scheme aims to improve the frequency nadir (FN) and ensure stable operation of a DFIG. To achieve the first goal, the scheme uses a droop loop, but it dynamically changes its gain based on the ROCOF to release a large amount of kinetic energy during the initial stage of a disturbance. To do this, a shaping function that relates the droop to the ROCOF is used. To achieve the second goal, different shaping functions, which depend on rotor speeds, are used to give a large contribution in high wind conditions and prevent over-deceleration in low wind conditions during inertial control. The performance of the proposed scheme was investigated under various wind conditions using an EMTP-RV simulator. The results indicate that the scheme improves the FN and ensures stable operation of a DFIG.
Article
The electrical frequency of an interconnection must be maintained very close to its nominal level at all times. Excessive frequency deviations can lead to load shedding, instability, machine damage, and even blackouts. There is rising concern in the power industry in recent years about the declining amount of inertia and primary frequency response (PFR) in many interconnections. This decline may continue due to increasing penetrations of inverter-coupled generation and the planned retirements of conventional thermal plants. Inverter-coupled variable wind generation is capable of contributing to PFR and inertia; however, wind generation PFR and inertia responses differ from those of conventional generators, and it is not entirely understood how this will affect the system at different wind power penetration levels. The simulation work presented in this paper evaluates the impact of the wind generation provision of these active power control strategies on a large, synchronous interconnection. All simulations were conducted on the U.S. Western Interconnection with different levels of wind power penetration levels. The ability of wind power plants to provide PFRand a combination of synthetic inertial response and PFRsignificantly improved the frequency response performance of the system. The simulation results provide insight to designing and operating wind generation active power controls to facilitate adequate frequency response performance of an interconnection.