Improved Chaos-GA-PID control in digital invert power supply
Duan Bin, Tongjing Sun, Zhenhua Li, Mei Gaoqing, Zhang Guangxian · 2009
For multi-parameter coupling nonlinear welding process, both chaos theory and genetic algorithm are used to achieve optimal matching of PID parameters, and the way to apply the algorithm for the welding machine is presented in this paper. First, requirements of invert welding machine control are analyzed. Encoding and decoding methods and fitness function are designed to form improved genetic algorithm. Then, chaos mechanism is researched to overcome randomness and avoid shortcomings of improved GA, such as local minimum value and prematurity. From the simulation experiments, it is shown that the improved chaos genetic algorithm is better than ANN-PID, fuzzy-PID and so on, has faster convergence speed and higher steady state accuracy, and can realize accurate regulation of welding current adaptively.