【摘要】 针对BP算法收敛速度慢,存在易陷入局部极小值,不能有效地搜索到全局极小点等缺点。采用遗传算法优化神经网络权系值的方法,设计了神经网络供热负荷预测模型,并用实际数据进行了仿真检验,结果表明该模型不仅在一定程度上避免了学习中的局部极小问题、改进了网络性能、提高了学习的效率,而且对供热负荷预测具有较高的精度和可靠性。更多还原

【Abstract】 BP alogrithm has the weakness such as slow convergent speed, easy getting into local minimum and being insurable to find global minimum value point. This paper introduces the genetic alogrithm to optimize the weights of the neural network, and designs a neural network heating load forecasting model, and carrieds out simulation with the practical data.The results show it not only can avoid getting into local minimum to some degree and modify the capacity of the network and enhances the efficiency, but also gets high precision and reliability in the heating load forecasting.更多还原