Study on Power Transformers Fault Diagnosis Based on Wavelet Neural Network and D-S Evidence Theory

>Transformer faults are quite complicated phenomena and can occur due to a variety of reasons.There have been several methods for transformer fault synthetic diagnosis,but each of them has its own limitations in real fault diagnosis applications.In order to overcome those shortcomings in the existing methods,a new transformer fault diagnosis method based on a wavelet neural network optimized by adaptive genetic algorithm(AGA)and an improved D-S evidence theory fusion technique is proposed in this paper.The proposed method combines the oil chromatogram data and the off-line electrical test data of transformers to carry out fault diagnosis.Based on the fusion mechanism of D-S evidence theory,the comprehensive reliability of evidence is constructed by considering the evidence importance,the outputs of the neural network and the expert experience.The new method increases the objectivity of the basic probability assignment(BPA)and reduces the basic probability assigned for uncertain and unimportant information.The case study results of using the proposed method show that it has a good performance of fault diagnosis for transformers.

Surface charge inversion algorithms based on integral equation method

Accurate surface charge inversion can guide the research on surface modification of insulators in GIS/GIL.The current inversion algorithms have disadvantages of high computational cost and low accuracy.Based on that,the integral equation method(IEM)is proposed to calculate the transformation matrix.Compared with the traditional analytical method(AM),IEM has a simple calculation process.The calculation speed of IEM is much faster than that of AM.To suppress the numerical divergence in IEM,the Tikhonov regularisation method is introduced and Tikhonov-IEM is proposed.For square insulators,compared to IEM,the peak-mean square error(PMSE)is reduced by about 40 percent.However,Tikhonov-IEM is not suitable for basin insulators.Therefore,the least square method(LSM)is introduced and the LSM-IEM is proposed.For basin insulators,compared to IEM,the PMSE is reduced by about 30 percent.Finally,the accuracy of the algorithms is verified by physical tests.

Robust Load Frequency Control for Isolated Microgrids Based on Double-loop Compensation

Frequency is an important indicator for the oper-ation of microgrids.However,the randomness and uncertainty of renewable energy and load variability may lead to frequency undulation.So,a robust load frequency control(LFC)is pro-posed for isolated wind-diesel microgrids considering time delay and parameter uncertainty.The control strategy can suppress frequency fuctuation and optimize frequency dynamic response.First,the double compensation loop,including feedforward control and integral sliding mode control(SMC),is devised to provide anti-disturbance compensation for the diesel generator system and ameliorate the frequency stability of independent microgrids.Secondly,a dynamic fuzzy controller,composed of wind speed and load demand,is designed to provide real-time response reference power for doubly fed induction generator systems(DFIGs),which can promote the effective participation of a wind turbine system for frequency regulation.Then,the proportional differential(PD)parameters of a dynamic fuzzy controller and the frequency adjustment compensation of DFIGs can be obtained by using a particle swarm optimization(PSO)algorithm.Thirdly,load demand is an important index of the robust dynamic load frequency control method;the radial basis function(RBF)neural network observer(NNO)based on the LFC model is presented to obtain more accurate load deviations and improve the control precision of LFC.The performance of the proposed LFC method is tested under different operation cases.Index Terms-Load frequency control,microgrid,neural network observer,sliding mode,time delay and parameter uncertainty.