The Working Device Optimization Of Mining Cabin Cleaning Divice Based On Genetic Algorithm

The direction of the engineering machinery development is high efficiency and low consumption. The purpose of the article is to solve the defect of traditional optimization which can not find out the overall best consequence easily. By using stochastic optimizing and coding genetic algorithms to optimize mining cabin cleaning device.Under the premise of meeting the mechanical function, the research stare from the analysis of the movement of loading machine, build up a multi-parameter optimizing mathematic model. The objective function of the model is the climax force when mobile arms oil box and turn over oil box at the working position, the designing variable of the model is the rod length of connecting rod device, the bind function of the model are the capacity of working device and the design requirement. According to the practical condition of the optimizing design , the research improved the genetic algorithms as follows: combining standard genetic algorithms with combinatorial punish function, designed a reasonable operation when byte string of chromosome is too long at binary coding which lowered the effect of hybridize and variation , increased the working time and the possibility of sink into part optimize, the research adopted real number coding in order to decrease the decode procedure of standard genetic algorithms, the research adopted three genetic operators: the tournament model, arithmetic crossover and uneven mutation. All the program were organized by VB6.0. The program is consisted of crossover mutation, population replacement, selection main modules, input, output, coding, adaptive computation assist modules.By virtue of analysis of concrete, carry out the parameter optimizing setting of improved genetic algorithms, select out the reasonable population size, supreme genetic algebra, selective probability, crossover mutation probability and unevenmutation parameter. The result showed that under the condition that the supreme unload height, the minimum excavate depth, unload character, automatic setting character were all satisfied, impetus capacity of loading machine increased about 9 percent, the time of optimizing design was largely shorten, the executing efficiency is three times larger than combining punish function.Improved genetic algorithms solve the problem of seek solution during the loading machine optimizing design because of objective function can not be linear, parameters are multiple and restrain function are multiple. Improved the level of engineering design and design efficiency, shortened the period of design, reduced the engineering investment.