Optimization of PID parameters for constant-pressure water supply control system based on multi-strategy integrated particle swarm optimization(MSIPSO)

Abstract The urban constant-pressure water supply system is a typical nonlinear complex system. For such systems, conventional PID control often fails to adapt to multiple operating conditions, and the standard Particle Swarm Optimization (PSO) algorithm is susceptible to local optima when tuning PID parameters. To address these issues, a multi-strategy integrated Particle Swarm Optimization (MSIPSO) algorithm is proposed. The MSIPSO algorithm enhances population diversity via Kent chaotic mapping, incorporates a nonlinear adaptive inertia weight to balance global exploration and local exploitation capabilities, and employs the Levy flight strategy to prevent premature convergence. Experimental results on the CEC2005 benchmark functions demonstrate that the MSIPSO algorithm outperforms several comparative algorithms in both convergence accuracy and stability. Subsequently, the MSIPSO algorithm was applied for PID parameter tuning in the constant-pressure water supply system. The results indicate that the proposed MSIPSO-PID controller effectively reduces the overshoot, shortens the settling time, and consequently enhances the system’s dynamic performance. This study thereby offers a viable solution for the optimal control of complex industrial processes.