Reliability-based Robust Optimization Design For Structure Of Crane Based On Intelligence Algorithm

As a kind of mechanical equipment, the function of lifting machinery is to carry on the material handling operation, which is playing a more and more important role in modern industrial production. With the extensive use of cranes and the improvement of the production efficiency, cranes are used more frequently in production, so it puts forward higher requirements on the safety and reliability of crane. Once the crane is caused by structural failure, it will have very serious consequences, and the loss is immeasurable. As the basic unit of mechanical equipment, metal structure constitutes the support platform of mechanical equipment. Metal structure is similar to the skeleton of the crane in the human body, and it is mainly subjected to crane load, the quality of its design will have an impact on the technical, safety, service life, etc.As the object of this paper, the structure of crane is studied by different methods, and the structure of crane can be researched from different aspects. Structure design method used in this paper include: allowable stress method, limit state design method, allowable stress – stress ratio method, limit state-stress amplitude method, fruit fly optimization algorithm with learning, specular reflection algorithm, robust design, reliability design and non-parametric design, etc. The research contents of this paper as follows:1) The classical and commonly used allowable stress method and a new structure design theory limit state method are adopted in engineering design. Taking the structure of bridge crane and truck crane as the research object, the static test of two crane structures is carried out by using the allowable stress method and the limit state method, and the failure strength of the two structures is verified by allowable stress – stress ratio method and the limit state- stress amplitude method. The calculation results are compared and analyzed, and the advantages and disadvantages of the four design methods are analyzed and evaluated from the redundancy. The calculation results show that the limit state method is a relatively lightweight design method, and the limit state method has obvious advantages when the structure is subjected to a nonlinear relationship between the load and the internal force.2) Two optimization algorithms are studied and proposed. First, an improved learning algorithm is proposed by the author's understanding and research on the fruit fly optimization algorithm. The algorithm overcomes the shortcomings of the traditional fruit fly optimization algorithm, and has the ability to solve practical problems and the broad application prospects of the project. The second is propose a new global optimization method – specular reflection algorithm which is inspired by the function of mirror. The specular reflection algorithm adopts a new simulation object, which is different from the general search methods and iterative methods. The traditional mathematical theory is used to analyze the global convergence of the specular reflection algorithm, it is proved that the specular reflection algorithm has the ability to converge to the global extremum. And the numerical results show that the calculation speed and speed of specular reflection algorithm is significantly higher than that of the traditional intelligent optimization algorithm and learning algorithm. The light weight design of the crane structure is carried out by using the specular reflection method, which proves that the method has the ability to deal with the discrete optimization problem.3) A robust design model based on sensitivity is proposed, and the model is applied to the robust optimization design of crane metal structure. On the basis of the structural optimization design model, the specular reflection algorithm is combined with the design method of sensitivity, and the tolerance of the uncertain parameters of the structure design is considered. Through analyzing the engineering example, the robust optimization model proposed in this paper can effectively deal with the structural problems, which is an advanced design method to improve the reliability of structure.4) The theoretical derivation of reliability index is carried out, and a method for calculating the approximate solution of the structural reliability is obtained, combining this method with the specular reflection algorithm, and the model of structure reliability optimization is established. Based on the lightweight design model of structure, the reliability can meet the design requirement at the same time by adding structure reliability constraint, that is, the manufacturing cost of the structure can be reduced to a maximum extent under the premise of the safety of structure.5) The concept of non-parametric design is put forward, and the non-parameter design method is applied to the structure design model. Take the structure of general-purpose bridge crane and gantry crane as example, the method of non-parameter optimization design, non-parameter robust optimization design and non-parameter reliability optimization design is applied, the non-parameter design method can be proved to be feasible through a numerical example, and the three structure design model which is proposed by this paper are proved to be accurate once again.