Optimum Design Study Of Sturactural System Based On Reliability

In practical engineer, some randomness is appeared in the stiffness of structural system and external load is subjected to in structure. So during structure analyze and design, the randomness of structural system and load is considered is according to engineer practice.And for random mechanical analysis of structural system and reliability design based on probability idea, it is necessary to consider the randomness factor of practical engineer. Simultaneously, with the increase of structure complex extent and continuous update of structure materials, the structure bearing force become more complex and the difficulty of reliability analysis is also increased, all which are appeared as many indeterminate factors in structure design. And complex function relation is also appeared after structure bearing load, in addition, some high-technology products such as airship, rocket, ship and so on, all which must operate independently under complex and bad condition, so the enough stiffness must be ensured. On the other side, in order to increase speed and decrease cost, the sides and weights of these members are smallest possible under the enough intension are satisfied. At present, all consider generally that it has many superiority to take reliability parameters as control structural design parameters, so the structural system reliability or failure probability as restrained condition are combined in optimization design to carry out structural system optimization design based on reliability. Some studies are made on structural system reliability analysis and optimization design in this thesis, are included as the following:1. Based on the theory of structural system fatigue reliability, reliability function for future active structural system is given, the method to analyze remaining fatigue life for exiting structural system is proposed. Results indicated that in predicting activity with increase of load frequency the structural system fatigue reliability index decrease, under the condition of load frequency is invariant, with the structural system activity increase the structural system fatigue reliability index decrease continuously.2. In structure design, randomness of structural system and load is considered, sensitivity analysis of random structure safety margin and reliability indexβ,are finished, so the reliability index become restrained condition of function. Encode of GA and genetic operators are improved, so they are more appropriate for reliability optimization of stochastic structure.At last, the better global search capability of GA and local search capability of optimum vector algorithm are utilized to construction hybrid algorithm based on GA and optimum vector algorithm, at the same time the niche technology is induced to increase operational efficiency and solution quality, so the local optimum solution is avoid. Calculated example indicate that the hybrid algorithm with high accuracy and operate stably, which supply theory basis to reliability optimization of stochastic structural system especially to large-scale structure reliability optimization.3. Failure mode of structural system under dead load and failure load is analyzed, reliability analysis method of structural system reliability under these two loads is given, and the improved GA as optimum method is adopted, based on structure reliability analysis theory under dead load and failure load to carry out optimization design, calculated example validate that the validity of GA on solving this question is proved, which supply a reasonable method on studied is question.4. Reliability of composite material lamination plate is analyzed, the strength theory is combined to establish safety margin expression of lamination plate under layer fracture and stratified fracture, the SFEM is utilized to carry out sensitivity analysis on safety margin, and reliability analysis method of composite material lamination plate is given. Based on this, the spread quantity and spread angle of lamination plate are optimum designed. Calculated example indicate that efficiency and precision of improved GA are increased than SGA, the improved integer code integrate individual code of spread quantity and spread angle into volume code, so the amount of work is decreased; the use of dynamic optimized operators ensure multiplicity and convergence of group; delete operators delete directly dummy spread created in optimization, so the validity of optimization is ensured. Due to reliability index is only as constraint condition is induced into optimization mathematics mode, it made no difference on sufficiency of GA, so the final results got from optimization are considered as global optimum solution.