城市燃气管网日负荷预测的灰色神经网络模型

Creation of humanistic environment"High-rise serried-core" of city

将灰色预测理论和人工神经网络理论结合起来 ,利用灰色静态预测模型来弱化数据的随机性并建立规律的累加数据 ,再利用神经网络模型来解决数据的非线性 ,建立了既反映其时间序列的周期性变化趋势 ,又包括天气、气温等影响因素的燃气日负荷预测灰色神经网络模型 .对哈尔滨市燃气管网系统的日燃气用量进行了预测 ,表明模型不仅有较高的收敛速度和精度 。

Applying the theory of Grey forecast to the artificial neural network, the data randomness is weakened and the disciplinary accumulated data are developed using the static Grey forecast model, and the data linearity is then solved with the neural network model. The city gas network daily load forecast model reflects the variation trend with the periodicity of time serial, the weather and temperature. The results show that the model has better convergence and forecast preciseness, better applicability and flexibility.