Method for predicting and controlling water level of series of water conveyance canals on basis of fuzzy neural network

摘要

Disclosed is a method for predicting and controlling the water level of a series of water conveyance canals on the basis of a fuzzy neural network. The method comprises: conducting the relationship between the opening degree of a gate and the controlled water level of an open channel by means of a fuzzy neural network, and constructing a before-gate water level controller coupled to a predictive control algorithm; on the basis of a control objective of the before-gate water level controller, solving an optimal control rate of the before-gate water level controller by means of a gradient optimization algorithm; and on the basis of the solved optimal control rate, generating a control policy by collecting measured water level change information and multiplying same by the optimal control rate, and thereby achieving the aim of water level prediction and control.