论文部分内容阅读
为保证城市供水优化运行的安全性和可靠性,提出了基于时间序列和神经网络理论的城市用水量预测的SIMULINK仿真模型。基于时间序列预测法的SIMULINK仿真模型依据回归算法确定模型参数,得到预测结果和误差,可通过调整SIMULINK模块参数提高仿真精度;在基于神经网络的SIMULINK仿真模型中,根据BP神经网络原理分别建立输入层、隐含层和输出层模型,得到预测结果和误差,可通过增加训练样本数提高仿真精度。仿真结果表明:基于时间序列和神经网络的水量预测SIMULINK仿真模型,不仅预测精度达到要求,而且还具有模块直观、参数易调和结果可视化等优点。
In order to ensure the safety and reliability of urban water supply optimization operation, a SIMULINK simulation model of urban water consumption prediction based on time series and neural network theory is proposed. The SIMULINK simulation model based on time series prediction method can determine the model parameters according to the regression algorithm and get the prediction results and errors. The SIMULINK module parameters can be adjusted to improve the simulation accuracy. In the SIMULINK simulation model based on neural network, the input is established according to BP neural network principle Layer, hidden layer and output layer model, the prediction results and errors can be obtained, and the simulation accuracy can be improved by increasing the number of training samples. The simulation results show that the SIMULINK simulation model of water quantity prediction based on time series and neural network not only meets the requirements of prediction accuracy, but also has the advantages of intuitionistic module and easy parameter adjustment and visualization.