Water supply prediction method and water supply prediction system

Abstract

The invention discloses a water supply prediction method and system. The method includes: analyzing historical data and establishing a water supply prediction model to obtain the functional relationship between the input layer parameters and the water supply volume output by the output layer; according to the established water supply prediction model and the required Real-time water supply forecasting on the forecast date; the process of analyzing historical data and establishing a water supply forecasting model includes but is not limited to the process of preprocessing the input historical data by using the moving average method, and using the neural network algorithm to determine the network topology of the water supply forecasting model. The water supply forecasting model process is determined according to the preprocessed historical data and the corresponding network topology. The present invention comprehensively adopts the moving average method and the neural network algorithm to predict the water supply volume, has low calculation complexity, has obvious periodicity and strong nonlinearity, has good real-time performance, and can accurately describe the influence of various factors on the water supply volume. The method can accelerate the convergence speed of the neural network model, and can be widely used in the field of water supply analysis.

Description

technical field [0001] The invention relates to the field of water supply analysis, in particular to a water supply prediction method and system. Background technique [0002] Urban water supply forecasting refers to the use of past information to calculate the urban water supply load for a certain period of time in the future. Urban water supply forecasting is an important preliminary work for urban water supply system planning, construction and renovation optimization. Its accuracy will directly affect the size of the project, project cost, reliability and practicability of scheduling decisions. [0003] Urban water supply forecasting can be broadly classified into two types in terms of time: short-term water supply forecasting and medium- and long-term water supply forecasting. The former refers to predicting the water supply in a short period of time in the future (such as hourly forecast, daily forecast or weekly forecast) based on historical water supply data combined...

AI Technical Summary

Problems solved by technology

However, these existing water supply prediction methods have relatively large defects or deficiencies: the exponential smoothing model requires a large amount of data, and the calculation complexity is high; The linear feature is weakened to a certain extent; the autoregressive (AR) forecasting model has obvious hysteresis. When the actual data changes abnormally, the forecast data cannot reflect the abnormal change due to the smoothing effect of the model, making it predict some abnormalities. Large errors or even distortions are caused when the value is large; the gray forecasting model is difficult to accurately describe the influence of various factors on the water supply; the traditional neural network model converges slowly and is easy to fall into the situation of local minimum

Images