Water supply pipe network on-line modeling method based on ARX

Abstract

The invention discloses a water supply pipe network on-line modeling method based on ARX. The water supply pipe network on-line modeling method based on ARX includes the steps that real-time operating data of a pipe network system are set up, and an input and output process model of the system is built, wherein a real-time operating data base comprises real-time input data and real-time output data of the system, the input data comprise supply water pressure of a water plant, the amount of water supply of the water plant, the pressure and flow of a booster pump station in a pipe network, and the switch-on degree of a controllable valve on a pipe, and the output data comprise the pressure of a pressure monitor point, the flow of a flow detection point and the water level of an impounding reservoir; on the basis of a real-time process operating data set of the water supply pipe network, an ARX model in a discrete difference equation mode is built; finally, parameters of the model are estimated in real time. The parameters of the model can be estimated online and in real time, state changes of the water supply pipe network are tracked in real time and predicted, the model is simple in structure and high in robustness, and by combination of the prediction control theory and the self-adaptation control theory, the water supply pipe network can be optimized and dispatched in real time.

Description

technical field [0001] The invention belongs to the technical field of automatic control, in particular to an ARX (Auto Regressive Exogenous)-based online water supply pipe network modeling method. Background technique [0002] The water supply pipe network system is a dynamic network system with complex structure, large scale, and strong randomness of water consumption. Whether it can quickly and accurately simulate and predict the operating conditions of the pipe network is the key to optimal scheduling of the water supply pipe network. Water supply network system models can be roughly divided into two categories: micro model and macro model. The microscopic model is a mechanism model, which needs to know specific parameters such as the topological structure of the pipe network, pipe length, pipe diameter, pipe material and node water consumption, and is generally a nonlinear system equation composed of a mass balance equation and an energy balance equation, which ignores ...

AI Technical Summary

Problems solved by technology

Clearly, traditional models are incapable of

Images