论文部分内容阅读
分析了传统的粒子群优化(particle swarm optimization,PSO)算法和径向基(radial basis function,RBF)神经网络的优缺点,提出一种自适应变系数粒子群优化算法(adaptive variable coefficients particle swarm optimizer,AVCPSO)。该算法与RBF神经网络结合形成自适应变系数粒子群-径向基(AVCPSO-RBF)神经网络混合优化算法。基于此优化算法,建立了短期电价预测模型,并利用贵州电网历史数据进行短期电价预测。仿真计算结果表明,AVCPSO-RBF混合优化算法在短期电价预测中优于传统RBF神经网络法和PSO-RBF神经网络方法,克服了上述2种方法的缺点,改善了RBF神经网络的泛化能力,具有输出稳定性好、预测精度高、收敛速度快等特点,使用该方法得到的各日预测电价的平均百分比误差可控制在2%以内,平均绝对误差最大值为1.652RMB/MW·h。
The advantages and disadvantages of traditional particle swarm optimization (PSO) algorithm and radial basis function (RBF) neural network are analyzed. An adaptive variable coefficients particle swarm optimizer , AVCPSO). The algorithm combined with RBF neural network to form an adaptive variable coefficient particle swarm optimization (RBF) neural network hybrid optimization algorithm. Based on this optimization algorithm, a short-term electricity price forecasting model is established and the short-term electricity price forecast is made based on the historical data of Guizhou power grid. The simulation results show that the AVCPSO-RBF hybrid optimization algorithm outperforms the traditional RBF neural network method and the PSO-RBF neural network method in the short-term price forecasting, overcomes the shortcomings of the above two methods and improves the generalization ability of the RBF neural network, With the advantages of good output stability, high prediction accuracy and fast convergence rate, the average percentage error of the forecast electricity price obtained by this method can be controlled within 2% and the average absolute maximum error is 1.652RMB / MW · h.