Based On The Main Beam Of The Bridge Crane Annealing Genetic Algorithm To Optimize The Design And Software Implementation

The General overhead crane is an important lifting equipment to the nationaleconomy construction,to our country's modernized production andinfrastructure construction plays a vital role.The bridge crane metal structureaccounted for more than60%of the whole weight,used since the design theoryof crane metal structure is conservative,mainly in the high safety coefficient,thelarge self-weight of structure,such a man-made cause the overall size and weightof the crane is too large.If you can guarantee under the premise of thestrength,stiffness and stability, reduce its weight as much as possible,is not onlybeneficial to the transportation and installation,also can make the manufacturingcost greatly reduced.Therefore,uses the reasonable modern design methods tooptimize the metal structure of the bridge crane,is an essential part of the cranedesign.So far,many scholars at home and abroad in the crane metal strucureoptimization problem is studied,and have achieved certain results,sum up theused methods mainly have: Penalty function method,Random directionmethod,Complex shape method,and Genetic Algorithm,etc.However,thesemethods mainly in the improvement of optimization objectives do a lot ofwork,not on the optimize performance(optimize speed and global covergence)for any discussion,but a truly practical and effective optimization method notonly has good optimization results,should has more excllent optimizeperformance.Currently, used for the crane metal structure optimization designon the Genetic Algorithm,most of the problems are classified as a mixedvariable optimization problems,cause must be round the optimized designvariables;In addition, the basic Genetic Algorithm there are precocious,slowconvergence speed and low global convergence defects.In view ofthis,considering the production practice,the accuracy of the human eye canidentify and the thickness size of the steel plate prescribed in the nominalthickness "1." of the single rolling steel plate,re-positioning the crane metal structure optimization design belongs to constrained,nonlinear,discrete variableoptimization design problems,avoid must be round the optimized designvariables phenomenon;And aiming at the inherent defects of the GeneticAlgorithm put the Simulated Annealing Algorithm who has strong local searchability assimilated into the improved Genetic Algorithm, forms a hybridalgorithm——Adapative Simulated Annealing Genetic Algorithm (ASAGA forshort), owing to the new algorithm strengthen their respectiveadvantages,weakening their respective shortcomings,reaching behaviorcomplementary,through an example to verify its better optimize performance.Based on the Visual C++6.0development plateform,to the main beamsectional area of the General overhead crane as the objective function,establishthe mathematics model of the General overhead crane box girder optimizationdesign,application of the formed hybrid algorithm prepared the software of "TheCrane Box Girder Optimization",for the General overhead crane girdercross-section optimization,and make the optimize results compared withengineering cases,in meet the various constraint conditions required by the mainbeam,make the optimized girder cross-section reduced by19.9%,illustrating thatthe application of this software can reduce the weight of the crane,cut theconsumption of steel,to achieve the purpose of saving costs and improve thedesign efficiency.