CAJ | 학술논문

由于BP神经网络在解决非线性复杂系统中存在很大的优势，以江西省1991—2011年人口、经济和耗电量等数据为研究对象，利用BP神经网络构建耗电量预测模型。模型一利用1991—2009年人口、经济和耗电量等数据作为训练样本，以2010—2011年作为测试样本来验证网络的准确性，再根据历史人口、经济等数据来预测历史耗电量；模型二采用传统的多元回归分析法，对非线性多元函数进行多元线性回归，通过回归模型得到的参数来预测耗电量。结果表明，模型一收敛性较好，所得预测结果绝对误差较小，而模型二传统方法得到的预测结果误差较大，因此，利用BP神经网络预测的结果具有非常大的参考价值，证明BP神经网络应用在电力消耗中的应用是可行的。
유우BP신경망락재해결비선성복잡계통중존재흔대적우세，이강서성1991—2011년인구、경제화모전량등수거위연구대상，이용BP신경망락구건모전량예측모형。모형일이용1991—2009년인구、경제화모전량등수거작위훈련양본，이2010—2011년작위측시양본래험증망락적준학성，재근거역사인구、경제등수거래예측역사모전량；모형이채용전통적다원회귀분석법，대비선성다원함수진행다원선성회귀，통과회귀모형득도적삼수래예측모전량。결과표명，모형일수렴성교호，소득예측결과절대오차교소，이모형이전통방법득도적예측결과오차교대，인차，이용BP신경망락예측적결과구유비상대적삼고개치，증명BP신경망락응용재전력소모중적응용시가행적。
The BP neural network has great advantages in solving nonlinear complex system. This paper, using the population, economy and power consumption data during 1991-2011 in Jiangxi province as the research object, builds up electricity consumption forecasting models based on BP neural network. Model 1 adopts the annual popu?lation, economy and power consumption data during 1991-2009 as training samples, with those of 2010-2011 as test samples to verify the accuracy of the network. Then according to the historical data, it forecasts the power con?sumption based on factors of population and economy. Model 2 determines the multiple linear regression for non?linear multivariable functions through the regression analysis method, predicting electricity consumption through the parameters of the regression model. Results show Model 1 has good convergence with small prediction absolute error while Model 2 with the traditional method has larger errors. It proves that BP neural network is feasible in forecasting electric power consumption.