In the grinding process, a large number of PID controllers are needed. In this paper, the improved algorithm of global optimization is studied in depth, and combined with the PID controller, the global optimization PID controller is formed. The simulation results of MATLAB show that the intelligent PID controller has good effect and strong stability.
Intelligent PID controller is a new type of controller [5] [6], which uses intelligent algorithm and applies control strategy to the actuator. If the controlled object is replaced by the optimized objective function and the given value of the system is set to a sufficiently small value, the function value of the optimized objective function will keep approaching the given value until it reaches its global minimum under a certain control strategy. This is the global optimization PID algorithm. Applying the algorithm to the PID controller can effectively improve the efficiency of adjusting parameters, and it is also applied to the grinding process [1] [4]. At present, most of the numerical solutions of global optimization algorithms for optimization problems can only get local optimal solutions, but in practical engineering, most of them need global optimal solutions, such as control in grinding process [1].
However, many local optimal solutions exist in the optimization models of practical problems, which makes it impossible for some classical numerical methods to obtain the global optimal solution of the problem.
Therefore, it is necessary to develop global optimization algorithms to solve these problems.
In this paper, an improved global optimization algorithm based on linear relaxation is adopted.
First, a linear relaxation programming for the original generalized geometric programming problem is established by the following methods. Then a linear relaxation programming for the original GGP problem is constructed by the above method.
By solving the RLP problem [3], a lower bound of the global optimal value of the GGP problem is obtained. Then, thermostatic element the feasible region is divided and the global optimal solution of GGP problem is obtained by combining branch and bound method.
The above PID controller is used to form the control loop of the system.
In order to prevent single system from falling into local minimum, multi-control system parallel computing is adopted. The given value of each system is R, and the control strategy is PID algorithm, but the starting point of iteration calculation of each system is different. Multiple systems perform optimization iteration calculation at the same time, and each system must learn from the optimum values of itself and all systems. Experiments and simulation results: mixed programming with C language and MATLAB language. The algorithm program is written in C, and the subprogram of objective function, fitness function and output subprogram are written. The flow chart is as follows: Fig. 1. Through the above simulation figure 2, it can be clearly shown that the stability of the PID controller is very strong, the convergence speed is improved and the premature trapping into local optimum is prevented. The improved global optimization is used to set the parameters, which overcomes the shortcomings and shortcomings, and some improvement measures are put forward. Different coding strategies are used to improve the accuracy of the algorithm, and iteration is used to prevent premature falling into local optimum. The simulation results show that these improvements improve the stability and convergence speed of the PID.
The parameter tuning method based on global optimization algorithm proposed in this chapter is effective.