本文選題：粒子群算法 + BP神經(jīng)網(wǎng)絡(luò)��； 參考：《計算機科學(xué)》2017年S2期

【摘要】：短期負荷預(yù)測是電力系統(tǒng)正常運行的關(guān)鍵環(huán)節(jié),合理的發(fā)電計劃依靠準確的負荷預(yù)測,因此提出交變粒子群算法來優(yōu)化BP網(wǎng)絡(luò)模型以預(yù)測電力短期負荷。針對依靠先前的經(jīng)驗來確定BP神經(jīng)網(wǎng)絡(luò)的權(quán)值缺少理論依據(jù)的問題,采用交變粒子算法優(yōu)化BP神經(jīng)網(wǎng)絡(luò)權(quán)值,以減少通過神經(jīng)網(wǎng)絡(luò)預(yù)測模型求解電力短期負荷預(yù)測帶來的誤差。實驗證明,經(jīng)過優(yōu)化的BP神經(jīng)網(wǎng)絡(luò)預(yù)測模型比傳統(tǒng)的BP神經(jīng)網(wǎng)絡(luò)預(yù)測模型的誤差更小,更加接近實際電力負荷。
[Abstract]:The short-term load forecasting is the key link in the normal operation of the power system. The reasonable generation plan relies on the accurate load forecasting. Therefore, the alternating particle swarm optimization (PSO) algorithm is proposed to optimize the BP network model to predict the short-term load of the electric power system. The problem of the lack of theoretical basis for determining the weight value of the BP neural network is determined by the previous experience, and the alternating particle is adopted. The algorithm optimizes the weights of the BP neural network to reduce the error caused by the prediction model of the neural network. The experiment shows that the optimized BP neural network prediction model is smaller than the traditional BP neural network prediction model and is closer to the actual electrical load.
【作者單位】： 廣西大學(xué)電氣工程學(xué)院;廣西職業(yè)技術(shù)學(xué)院機械與汽車技術(shù)系;
【分類號】：TM715;TP18

【相似文獻】

相關(guān)期刊論文 前10條

1 鄭永康;陳維榮;蔣剛;郝文斌;;基于混沌理論的短期負荷局域多步預(yù)測法[J];電力系統(tǒng)及其自動化學(xué)報;2007年04期

2 王辛，孟昭敦;短期負荷預(yù)報最優(yōu)算法的模糊判據(jù)[J];電力系統(tǒng)自動化;1995年12期

3 劉遵雄,鐘化蘭,張德運;最小二乘支持向量機的短期負荷多尺度預(yù)測模型[J];西安交通大學(xué)學(xué)報;2005年06期

4 雷紹蘭;孫才新;周nv;鄧群;劉凡;;電力短期負荷的多變量混沌預(yù)測方法[J];高電壓技術(shù);2005年12期

5 傅書，

本文編號：1978039