Urban instantaneous water consumption prediction method based on LSTM

Abstract

The invention discloses an urban instantaneous water consumption prediction method based on LSTM, and the method comprises the steps: firstly building a prediction mechanism through employing a long-short-term memory artificial neural network as a basic regression algorithm, and then collecting the actual normal operation data of a part of urban water consumption units as a training sample; learning and training to obtain an LSTM prediction model based on a long short-term memory artificial neural network, and finally using the urban water consumption in the next time period as prediction input to obtain a prediction value output by the LSTM prediction model. According to the method, the urban water consumption data sets with similar volatility are adopted to participate in LSTM training, the defect of low precision of a historical urban water consumption prediction algorithm can be solved, and the problem of slow algorithm convergence can be well solved.

Description

technical field [0001] The invention belongs to the technical field of urban water consumption prediction, and in particular relates to an LSTM-based urban instantaneous water consumption prediction method. Background technique [0002] With the continuous development of society and economy, urban water demand is rising, but the water supply is very limited. In order to solve this outstanding contradiction, it is necessary to analyze and predict the utilization of local water resources, so as to carry out long-term unified planning and management of local water resources and water conservancy projects. Commonly used urban water consumption forecasting methods are divided into three categories: intuitive forecasting method, time series forecasting method and simulation model forecasting method. Intuitive prediction method is a kind of qualitative prediction, which refers to the method of judging the future water use condition by relying on human's intuitive judgment ability,...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide an LSTM-based urban instantaneous water consumption prediction method, which solves the problem that most of the neural networks used in the existing urban water consumption prediction use the BP algorithm, which is easy to fall into local extremum and requires a long training time.

Images