Research of BPNN Application in Liquid Level Control of Nuclear Power Plant Steam Generator

The performance of steam generator (SG) water level control is a critical factor in ensuring the safe, efficient, and stable operation of nuclear power plants. Based on the basic structure of a Backpropagation Neural Network (BPNN) Proportional-Integral-Derivative (PID) controller and incorporating the physical characteristics of the SG, a dedicated BPNN PID controller suitable for SG water level control processes was developed. This PID controller features self-matching, self-adaptive, and self-tuning capabilities, enabling it to automatically calculate appropriate PID parameters in real time based on changes in the operating conditions of the controlled object during nuclear power plant operations. This ensures consistently excellent control performance of the controller. The BPNN PID controller model was subject to full-scope simulator tests for nuclear power plants. While maintaining control performance comparable to the original controller under full-power operation conditions of nuclear power plants, it demonstrated significantly improved control performance under low-power operation conditions.