No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Non-synchronous Signal Monitoring Based on Simulated Annealing Neural Network
Abstract
A neural network method combined with simulated annealing algorithm is proposed for power system harmonic analysis. This method is aimed at the system in which the sampling frequency cannot be locked on the actual fundamental frequency. By updating the relevant parameters including the learning rate of fundamental frequency, fundamental frequency, harmonic phases and amplitudes, the accurate harmonic estimating results can be obtained. The simulating results show that the harmonic estimation accuracy by the proposed approach is relatively better than that by the conventional harmonic analysis methods in the asynchronous case.
... Further, four objective functions to measure the performance of the detection platform were set as follows: the magnetic flux density, the uniformity of the magnetic flux density, the Bx component of magnetic flux density, and the uniformity of the Bx component of magnetic flux density in the circular wafer area where the MNP device was located. The MNP detection platform applies a static uniform magnetic field (10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20) to the MNP device through the Helmholtz coil structure [27], and the magnetic field generated by a current-carrying conductor can be calculated through the Biot-Savart law, as shown in Equation (1). ...
... For the detection positions of group 1-90 (Position1-Position90), the evaluation value of the group was greater when the MAE was smaller, the detection accuracy of the number of MNPs was higher and the average response magnetic field intensity was also higher. Based on the obtained MAE, detection accuracy of the number of MNPs and average response magnetic field intensity, combined with the linear weighting method, the evaluation function H(x) was constructed, which is shown in Equation (18). ...
... In Equation (18), m represents the number of index functions, w i represents the weighting factor, f i (x) represents the index function, and H(x) represents the evaluation function. ...
In recent years, magnetic nanoparticles (MNPs) have been widely used as a new material in biomedicine and other fields due to their broad versatility, and the quantitative detection method of MNPs is significantly important due to its advantages in immunoassay and single-molecule detection. In this study, a method and device for detecting the number of MNPs based on weak magnetic signal were proposed and machine learning methods were applied to the design of MNPs number detection method and optimization of detection device. Genetic Algorithm was used to optimize the MNPs detection platform and Simulated Annealing Neural Network was used to explore the relationship between different positions of magnetic signals and the number of MNPs so as to obtain the optimal measurement position of MNPs. Finally, Radial Basis Function Neural Network, Simulated Annealing Neural Network, and partial least squares multivariate regression analysis were used to establish the MNPs number detection model, respectively. Experimental results show that Simulated Annealing Neural Network model is the best among the three models with detection accuracy of 98.22%, mean absolute error of 0.8545, and root mean square error of 1.5134. The results also indicate that the method and device for detecting the number of MNPs provide a basis for further research on MNPs for the capture and content analysis of specific analyte and to obtain other related information, which has significant potential in various applications.
... At present, BP NN is widely used in harmonic analysis of power system. Such as in [10], by using f s as the sampling frequency of the harmonic signal sampling, the discrete model based on BP NN is expressed as: ...
The research is aimed at the application of back propagation neural network (BP NN) on the world's first AP1000 third generation 1250 MVA nuclear half-speed (4-pole) turbo-generator in harmonic analysis. Harmonic distortion has caused many power quality problems in the large-capacity generator unite of power system since the discovery of alternating current system. Taken AP1000 nuclear turbo-generator which has many problems needed solving as the analysis example, this paper presents the BP neural network technique which can realize the ability of self-learning to estimate and calculate the harmonics in AP1000 turbo-generator under various negative-sequence loss conditions. Considering the BP neural network application and harmonic effects in the large-capacity turbo-generator, overall simulation is performed in this work. Simulation results indicate that the method of using BP neural network is effective and accurate, and the change regulation of each harmonic is also obtained.
This work is the application of back propagation neural network (BP NN) on the world's first AP1000 third generation 1250 MVA nuclear half-speed (4-pole) turbo-generator in harmonic analysis. Large-capacity generator is the mainstream of nuclear power future, which leads the trend of nuclear power development and also faces many problems need to be tackled as soon as possible. Harmonic distortion is a predominant factor influent turbo-generator's output power quality and power system operations. In order to solve the problem, this paper presents an application control strategy to estimate the harmonics in the AP1000 nuclear turbo-generator unit using BP neural network, by which the neural structure can be used for harmonic analysis and power quality control. Simulation results prove that the method can realize the ability of self-learning. Meanwhile, the application results identify that BP neural network is an effective technology to calculate and analyze the harmonics in AP1000 large-capacity generator system under various negative-sequence loss conditions, and pave the path of the further theory research and the application practice a good solid foundation at the same time.
The wide use of non-linear components in power system gives rise to not only integer harmonics, but also non-integer harmonics in the power system. The conventional harmonic measurement algorithm Fast Fourier Transform (FFT) is suitable to be used in integer harmonic analysis, but is not fit to analyze non-integer harmonics due to its leakage and picket fence effects, which brings about large errors in practical applications. The engender of the leakage effect is caused by the different characters between the theoretic implementation of the Fourier Transform which deals with infinite signals and the practical implementation of Fourier Transform which deals with finite signals. These differences give rise to measurement errors of non-integer harmonics of FFT algorithm. In order to reduce the leakage errors and improve the measurement precision, this paper presents amending algorithm based on the analysis of the FFT algorithm. Through simple transforms of FFT algorithm, the amending algorithm can reduce satisfactorily the leakage error, and obtain accurate analysis results. Simulations validate the high precision of this novel algorithm. The amending algorithm have the characteristics of easy implementation and high precision, and will be a practical method for harmonic analysis in power system.
The mechanism of the impact of learning rate on the error-drop-curve of BP networks was studied. Then a limited supervising method for learning rate regulation was proposed, by which the learning rate of BP networks is capable for self-adaptive adjustment. This method was validated by demonstration study and engineering application.
It is improved in two aspects to the traditional window and interpolation algorithm, which is now widely employed to estimate power harmonics parameters. Firstly, a new kind of windows, called as Rectangular Self-Convolution Windows (RSCW), is presented. The m-order RSCW is developed by convolving m rectangular sequences; and in the zero points of its amplitude-frequency characteristic, the value of m-1 order derivatives is zero. As a result of this character, the interferences between the harmonics due to spectrum leakage can be reduced furthest by applying RSCW, therefore, the precision of harmonics estimation is boosted. Secondly, an interpolation algorithm suitable for RSCW is given, and some novel measures are taken to reduce the computational load. At last, the feasibility and effectiveness of the improved algorithm based on RSCW is verified by the results of both simulations and practice.
Fourier transform (FT) based techniques have been traditionally used for power system harmonics analysis. A recent alternative approach of wavelet transform (WT) is applied for non-sinusoidal signal analysis which has been found very useful in power quality studies. The problem of root mean square (RMS) value calculation is an important issue in harmonic analysis. This paper uses a statistical approach for RMS calculation. The proposed method simplifies wavelet based harmonic measurement process and increases measurement accuracy. Case studies from practical power systempsilas voltage and current signals are demonstrated. The measurement results using the discrete wavelet transform (DWT) method for various wavelet families are compared and discussed with the results of IEEE Std. 1459-2000 which forms the basis to propose the most suitable mother wavelet for power system harmonics analysis.
A back-propagation neural network method is proposed for accurate frequencies, amplitudes and phases estimation from periodic signals in power systems, and the convergence theorem shows that the proposed algorithm can be convergent asymptotically to its global minimum. The method is aimed at the system in which the sampling frequency cannot be locked on the actual fundamental frequency. Some simulating examples are given and the results show that the accuracy of the estimates provided by the proposed approach in the asynchronous case is relatively better than that of the estimates obtained with the conventional harmonic analysis methods.
The effects of phorbol-12,13-dibuterate (PDBu) on total sodium current (INa-total), tetrodotoxin-resistant sodium current (INa-TTXr), 4-AP-sensitive potassium current (IA) and TEA-sensitive potassium current (IK) in trigeminal ganglion (TG) neurons were investigated. Whole-cell patch clamp techniques were used to record ion currents in cultured TG neurons of rats. Results revealed that 0.5 μmol/L PDBu reduced the amplitude of INa-total by (38.3±4.5)% (n=6, P<0.05), but neither the G-V curve (control: V
0.5=−17.1±4.3 mV, k=7.4±1.3; PDBu: V
0.5=−15.9±5.9 mV, k=5.9±1.4; n=6, P>0.05) nor the inactivation rate constant (control: 3.6±0.9 ms; PDBu: 3.6±0.8 ms; n=6, P>0.05) was altered. 0.5 μmol/L PDBu could significantly increase the amplitude of INa-TTXr by (37.2±3.2)% (n=9, P<0.05) without affecting the G-V curve (control: V
0.5=−14.7±6.0 mV, k=6.9± 1.4; PDBu: V
0.5=−11.1±5.3 mV, k=8.1±1.5; n=5, P>0.05) or the inactivation rate constant (control: 4.6±0.6 ms; PDBu: 4.2±0.5 ms; n=5, P>0.05). 0.5 μmol/L PDBu inhibited IK by (15.6±5.0) % (n=16, P<0.05), and V
0.5 was significantly altered from − 4.7±1.4 mV to −7.9 ±1.8 mV (n=16, P<0.05). IA was not significantly affected by PDBu, 0.5 μmol/L PDBu decreased IA by only (0.3±3.2)% (n=5, P>0.05). It was concluded that PDBu inhibited INa-total but enhanced INa-TTXr, and inhibited IK without affecting IA. These data suggested that the activation of PKC pathway could exert the actions.
The fast Fourier transform (FFT) has been widely used for the signal processing because of its computational efficiency. Because of the spectral leakage and picket-fence effects associated with the system fundamental frequency variation and improperly selected sampling time window, a direct application of the FFT algorithm with a constant sampling rate may lead to inaccurate results for continuously measuring power system harmonics and interharmonics. An improved FFT-based algorithm to measure harmonics and interharmonics accurately is proposed. In the proposed algorithm, a frequency-domain interpolation approach is adopted to determine the system fundamental frequency, and the interpolatory polynomial method is applied to reconstruct the sampled time-domain signal; it is followed by using the FFT to calculate the actual harmonic components. Then, the frequency-domain interpolation is again applied to find the interharmonic components. The performance of the proposed algorithm is validated by testing the actual measured waveforms. Results are compared with those obtained by directly applying a typical FFT algorithm and by the IEC grouping method. It shows that the solutions determined by the proposed algorithm are more accurate, and a reasonable computational efficiency is maintained.
The interharmonics, which are the same as integer harmonics, have serious impacts on the design of harmonics compensation equipment. Therefore, it is important to measure interharmonics accurately. The interharmonics cannot be obtained accurately by using fast Fourier transform (FFT) for an unsynchronized sample sequence analysis because of large errors in such cases. To reduce the errors caused by using FFT for an unsynchronized sample sequence analysis and to improve the measuring precision of interharmonics in electric power systems, an interharmonics estimation method based on the interpolation FFT algorithm is proposed. An algorithm for estimating interharmonics parameters based on the Blackman-Harris window, Rife-Vincent (I) window, and Rife-Vincent (III) window is given. Simulation results demonstrate that by using the proposed algorithm for the unsynchronized sample sequence analysis, the estimated interharmonics frequency, amplitude, and phase are highly accurate
Optimal design for learning rate of BP neutral network
- Y H Liu
- R Chen
- W Peng
- L Zhou
Y.H. Liu, R. Chen, W. Peng and L. Zhou, " Optimal Design for Learning Rate of BP Neutral Network ", Journal of Hubei University of Technology, China, Volume 22, Issue 3, 2007, pp. 1-3.
Modern optimization computing method, Tsinghua university press
- W X Xin
- J X Xie
Morlet wavelet based detection of noninteger harmonics
- H Xue
- R G Yang