摘要
分析了粒子群算法和组合预测的特点。将组合预测和粒子群算法结合,建立了一种组合形式的供热负荷预测模型。同时针对粒子群算法易于陷入局部最优解、进化后期收敛慢等缺点对粒子群算法进行改进,解决了组合预测中权重难以确定的问题。改善了预测模型的拟合能力,提高了预测精确度。最后选取大庆油田某一天供热数据作为测试数据,结果表明组合预测误差较小,精确度高于其他单项预测方法40%以上。
This study analyzed the characteristics of the particle swarm algorithm and the combination forecast.By combining the combination forecasting with the improved particle swarm optimization,a combination of heating load forecasting model was established.For the particle swarm algorithm is liable to fall into local optimal solution,and due to its slow convergence in the later stage of evolution and other shortcomings,the standard particle swarm optimization was improved,which solved the difficulties in determining the weight coefficients.And the fitting ability of forecasting model was effectively improved as well as the accuracy of prediction.Adopting the data from daily run of a thermal substation in Daqing oilfield to verify the forecasting effect of the model,the results show that the accuracy of the combination forecasting is higher than other single forecasting methods by 40%;it has the minimum predication error.
出处
《信息与电子工程》
2011年第5期655-659,共5页
information and electronic engineering
关键词
供热负荷预测
组合预测
粒子群算法
权重
预测精确度
heating load forecasting
combination forecasting
particle swarm optimization
weight coefficient
accuracy of prediction