论文部分内容阅读
为了提高电力市场环境下的电价预测精度,在研究短期电价预测中采用了粒子群和反向传播神经网络相结合的混合算法,先利用粒子群算法确定初值,再采用神经网络完成给定精度的学习。对我国四川电网电价进行预测的结果表明,粒子群优化的神经网络算法收敛速度快于神经网络算法,预报精度显著提高,平均百分比误差可控制在2%以内,平均绝对误差最大值为1.87$/MWh。该算法可有效用于电力系统的短期电价预测。
In order to improve the prediction accuracy of electricity price in electricity market environment, a hybrid algorithm combining particle swarm optimization and backpropagation neural network is used in the research of short-term electricity price forecasting. The initial value is determined by particle swarm optimization algorithm firstly, then the neural network is used to achieve the given accuracy Learning. The results of forecasting the electricity price of Sichuan power grid in China show that the particle swarm optimization neural network algorithm converges faster than the neural network algorithm, and the prediction accuracy is significantly improved. The average percentage error can be controlled within 2%, the average absolute maximum error is 1.87 $ / MWh. The algorithm can be effectively used in the short-term price forecasting of power system.