Method for controlling temperatures of reactors of pressurized water reactor nuclear power stations by aid of RBF (radial basis function) neural networks

Abstract

The invention relates to a method for controlling temperatures of reactors of pressurized water reactor nuclear power stations by the aid of RBF (radial basis function) neural networks. The method is used for controlling the temperatures of the reactors of 900MW pressurized water reactor nuclear station units by the aid of the RBF neural networks, and includes testing characteristics of reactor temperature control systems on pressurized water reactor nuclear power unit simulation test platforms; fitting analytic and transfer functions of the systems according to data acquired from tests, solving the reactor temperature transfer functions and building high-precision mathematical models; designing traditional PID (proportion, integration and differentiation) controllers and controlling the temperatures of the reactors. The method has the advantages that the RBF neural networks are combined with the traditional PID controllers on the basis of the traditional PID controllers, so that the temperatures of the reactors can be compositely controlled, and obvious control effects can be realized as compared with control effects of the traditional PID control and fuzzy PID control; the method has high practical application value in the aspect of reactor temperature control, research and application of advanced intelligent control on nuclear power units can be promoted, and accordingly the control level on the units can be improved.

Description

technical field [0001] The invention relates to one, in particular to an RBF neural network control method for the reactor temperature of a pressurized water reactor nuclear power plant. Background technique [0002] Since the 21st century, with the recovery of the world economy and the increasingly serious energy crisis, nuclear energy has been favored for its advantages as a clean energy source, and it has become the three pillars of the world's electric energy together with thermal power and hydropower. There are many types of nuclear reactors, such as pressurized water reactors, boiling water reactors, heavy water reactors, and fast neutron reactors. The pressurized water reactor has become the most widely used nuclear reactor in the world so far due to its unique advantages such as compact structure, mature technology, low infrastructure cost, and short construction period. What the present invention studies is the temperature regulation system of the pressurized water...

AI Technical Summary

Problems solved by technology

[0013] The present invention aims at the problem that the reactor temperature control has strict requirements in the production process of the current nuclear power plant, and the current control cannot adapt to the large-scale change of the system parameters. When operating under working conditions, when parameters change in a small range, conventional PI or PID control effects are better, but they cannot adapt to large-scale changes in system parameters. The present invention combines the RBF neural network with traditional PID control to achieve composite control. The effect is better than traditional PID control and fuzzy PID control

Images