The Mechanical Reliability Optimization Based on the Improved Artificial Bee Colony Algorithm

The reliability is the factor that can't be ignored in the optimization design of the mechanics. In the large complex mechanical systems, especially in the closet-state function systems, the reliability based optimization has the features that the convergence speed is slow, and process easily get trapped in the local optima because of the stronger nonlinearity of the systems. In order to conquer these weaknesses in the reliability optima process, the paper proposes an improved artificial bee colony algorithm, this new algorithm uses the genetic hybrid genes and the chaotic-search strategy, and make the optima process convergence quickly and avoid get trapped in the local optima easily. This paper achieves the best design objective value base on the high reliability in the mechanical system with the parameterized virtual prototyping technology and the improved artificial bee colony. At last, paper proofs this method's effectiveness and practically through an engineering example.