基于改进PSO和DE优化神经网络的电能质量扰动分类
|
摘要
针对电能质量问题提出了基于改进粒子群优化算法(PSO)和差分进化算法(DE)相结合优化神经网络的分类方法。首先用Matlab仿真几种典型的电能质量扰动信号,再利用小波变换进行多尺度的分解,得到各尺度的能量信息作为特征向量输入BP神经网络分类器中对扰动信号进行快速、准确的分类识别。并针对传统BP算法收敛耗时长速度慢,不能保证获得全局最优等缺点,在种群分类基础上提出了一种混合粒子群与差分进化算法的新型PSO-DE算法,并利用其对神经网络进行改进。这种混合PSO-DE算法在很大程度上能弥补BP神经网络的不足,采用该算法对网络进行优化后完成电能质量扰动信号的自动分类。
Aiming at the power quality problem,an improved classification method based on improved particle swarm optimization( PSO) and differential evolution algorithm( DE) optimization neural network is proposed. Firstly,several typical power quality disturbances are simulated by Matlab,and multi-scale decomposition of wavelet transform is used to obtain the energy features of each scale,which is used as input feature vectors of neural networks. Then BP network is used to train the output samples,and the input feature vectors are identified and classified. In view of the shortcomings of the traditional BP algorithm,such as easiness to fall into local minimum and slow convergence speed,a new PSO-DE hybrid algorithm is introduced to improve the neural network. The algorithm is based on improved particle swarm optimization and differential evolution algorithm for population classification. This hybrid PSODE algorithm can make up for the deficiency of BP neural network to a great extent. After the optimization of the network,the algorithm can automatically classify the power quality.
Aiming at the power quality problem,an improved classification method based on improved particle swarm optimization( PSO) and differential evolution algorithm( DE) optimization neural network is proposed. Firstly,several typical power quality disturbances are simulated by Matlab,and multi-scale decomposition of wavelet transform is used to obtain the energy features of each scale,which is used as input feature vectors of neural networks. Then BP network is used to train the output samples,and the input feature vectors are identified and classified. In view of the shortcomings of the traditional BP algorithm,such as easiness to fall into local minimum and slow convergence speed,a new PSO-DE hybrid algorithm is introduced to improve the neural network. The algorithm is based on improved particle swarm optimization and differential evolution algorithm for population classification. This hybrid PSODE algorithm can make up for the deficiency of BP neural network to a great extent. After the optimization of the network,the algorithm can automatically classify the power quality.
引文
[1]张逸,林焱,吴丹岳,等.电能质量监测系统研究现状及发展趋势[J].电力系统保护与控制,2015,43(2):138-147
[2]徐永海,赵燕.基于短时傅里叶变换的电能质量扰动识别与采用奇异值分解的扰动时间定位[J].电网技术,2011,35(8):174-180
[3]De Yong D,Bhowmik S,Magnago F.An effective powerquality classifier using wavelet transform and support vector machines[J].Expert Systems with Applications,2015,42(15-16):6075-6081
[4]吴兆刚,李唐兵,姚建刚,等.基于小波和改进神经树的电能质量扰动分类[J].电力系统保护与控制,2014,42(24):86-92
[5]Peng L,Jing G,Duo X,et al.Hilbert-Huang transform with adaptive waveform matching extension and its application in power quality disturbance detection for microgrid[J].Journal of Modern Power Systems and Clean Energy,2016,4(1):19-27
[6]黄南天,张卫辉,蔡国伟,等.采用改进多分辨率快速S变换的电能质量扰动识别[J].电网技术,2015,39(5):1412-1418
[7]尹柏强,何怡刚,朱彦卿.一种广义S变换及模糊SOM网络的电能质量多扰动检测和识别方法[J].中国电机工程学报,2015,35(4):866-872
[8]李晓娜,沈兴来,薛雪,等.基于改进HHT和决策树的电能质量扰动辨识[J].电力建设,2017,38(2):114-121
[9]张明,李开成,胡益胜.基于多域特征提取和自适应神经-模糊推理系统的电能质量扰动识别[J].电力系统保护与控制,2011,38(24):6-13
[10]Martin V R,de Jesus Rene R T,Alfredo O R R,et al.Detection and classification of single and combined power quality disturbances using neural networks[J].IEEE Transactions on Industrial Electronics,2014,61(5):2473-2482
[11]俞晓冬,周栾爱.基于改进SVM模型的电能质量扰分类[J].电力系统保护与控制,2010,38(3):15-19
[12]Kennedy J,Eberhart R.Particle swarm optimization[C].In:Proceedings of IEEE International Conference on Neural Networks,Perth,Australia,1995,4(8):1942-1948
[13]Storn R,Price K.Differential evolutiona simple and efficient heuristic for global optimization over continuous spaces[J].Journal of Global Optimization,1997,11(4):341-359
[14]易文周,张超英,王强,等.基于改进PSO和DE的混合算法[J].计算机工程,2010,36(10):2
[2]徐永海,赵燕.基于短时傅里叶变换的电能质量扰动识别与采用奇异值分解的扰动时间定位[J].电网技术,2011,35(8):174-180
[3]De Yong D,Bhowmik S,Magnago F.An effective powerquality classifier using wavelet transform and support vector machines[J].Expert Systems with Applications,2015,42(15-16):6075-6081
[4]吴兆刚,李唐兵,姚建刚,等.基于小波和改进神经树的电能质量扰动分类[J].电力系统保护与控制,2014,42(24):86-92
[5]Peng L,Jing G,Duo X,et al.Hilbert-Huang transform with adaptive waveform matching extension and its application in power quality disturbance detection for microgrid[J].Journal of Modern Power Systems and Clean Energy,2016,4(1):19-27
[6]黄南天,张卫辉,蔡国伟,等.采用改进多分辨率快速S变换的电能质量扰动识别[J].电网技术,2015,39(5):1412-1418
[7]尹柏强,何怡刚,朱彦卿.一种广义S变换及模糊SOM网络的电能质量多扰动检测和识别方法[J].中国电机工程学报,2015,35(4):866-872
[8]李晓娜,沈兴来,薛雪,等.基于改进HHT和决策树的电能质量扰动辨识[J].电力建设,2017,38(2):114-121
[9]张明,李开成,胡益胜.基于多域特征提取和自适应神经-模糊推理系统的电能质量扰动识别[J].电力系统保护与控制,2011,38(24):6-13
[10]Martin V R,de Jesus Rene R T,Alfredo O R R,et al.Detection and classification of single and combined power quality disturbances using neural networks[J].IEEE Transactions on Industrial Electronics,2014,61(5):2473-2482
[11]俞晓冬,周栾爱.基于改进SVM模型的电能质量扰分类[J].电力系统保护与控制,2010,38(3):15-19
[12]Kennedy J,Eberhart R.Particle swarm optimization[C].In:Proceedings of IEEE International Conference on Neural Networks,Perth,Australia,1995,4(8):1942-1948
[13]Storn R,Price K.Differential evolutiona simple and efficient heuristic for global optimization over continuous spaces[J].Journal of Global Optimization,1997,11(4):341-359
[14]易文周,张超英,王强,等.基于改进PSO和DE的混合算法[J].计算机工程,2010,36(10):2