Common models for form error evaluation using a refined GA

The problem of form error evaluation was essentially a nonlinear optimization problem. A refined genetic algorithm(GA)-based method was proposed to search for the optimum solution of the problem. Firstly the error zone variables of six kinds of basic geometric features were analyzed under the minimum zone conditions. The common mathematical model was given and the fitness function of GA was provided according to the specific problem. Then the key techniques in the process of the algorithm implementation were discussed in detail such as the population size, initial population, selection strategy, genetic operators and termination criterion, etc., Finally the method was verified by examples. The computational results indicate that the model is reasonable and can give an accurate and robust solution.