摘要
为避免BP算法易陷入局部极小的缺陷,根据遗传算法具有全局寻优的特点,将二者结合起来形成GA-BP混合算法。以GA优化BP网络的初始权值和阈值,按负梯度方向修正网络权值及阈值,对网络进行训练。用matlab编写GA-BP计算程序,以多组数据进行测试,并与纯BP算法进行分析比较,结果表明该方法可以有效、准确的应用于短期电力负荷预测。
To avoid the BP algorithm’s shortcoming of trapping to a local optimum and to take advantage of the genetic algorithm’s globe optimal searching,a new kind of hybrid algorithm was formed based on GA and BP。 First,the initialized weights and biases of BP neural network was optimized with GA,and the network was trained by modifying the weights and biases。 Finally,the program code of GA-BP was made in Matlab language,and some data was used to test it。 Compare to BP algorithm,the results shows that GA&BP is an ef-fective and accurate method to predict the Short-Term Power Load Forecast.
出处
《电脑知识与技术(过刊)》
2009年第3X期1962-1964,共3页
Computer Knowledge and Technology
关键词
遗传算法
BP算法
神经网络
电力负荷预测
genetic algorithm
BP algorithm
neural network
prediction of power demand