Search

This paper studies reliability analysis of structures with multiple failure modes. The analysis consists of three parts. First, the memetic animal migration optimization algorithm, applied for multiple optimum points detection in a multimodal optimization problem, is used to search for the most probable points of the limit state functions of main failure modes. Next, the multi-point first-order reliability method is used to estimate the failure probability of structures with multiple failure modes. Last, the saddlepoint approximation based second-order reliability method is further extended to system reliability analysis, aiming to improve the accuracy of system reliability prediction. [ABSTRACT FROM AUTHOR]

Tool changing is on the basis of tool wear failure and breakage failure. Utilization rate of the tool is low, which cannot guarantee the reliability of the whole machining system. In machining process, as the number of parts produced increases, tool wear is constantly increasing, which will contribute to the reduction of the reliability of the cutting tool. That account for processing of substandard products. Combining the moment estimation with the maximum likelihood estimation with the dynamic reliability analysis method, the study builds a mathematical model of the dynamic reliability for machining process under the premise of regarding cutting parameters as random variables. The reliability of overall machining process is lower than a given target; the proposed model can identify the tool which has the biggest failure rate quickly and accurately. The failure rate formula of each tool involved in each operation is deduced as well. Based on failure rate, an algorithm for defining the critical tool and its corresponding tool change time is proposed. Beyond that, for maximizing the utilization of every tool, the given model can pick up the cutting parameter which has the largest sensitive degree to the reliability via sensitivity analysis method. Then, the selection of relevant stock removal should be changed so as to improve the reliability of cutting tool and the whole process system, as well as enabling the cutter continue to work and delaying tool change time finally. Ultimately, the manufacturing process can maximize the cutters' potential and thus reduce the number of tool changes, as well as the production costs. [ABSTRACT FROM AUTHOR]

The parameters of a system have the randomness generally in the process of milling, which influences the stability of the milling. This paper uses the neural network to get a comprehensive analysis of the influences of random factors in milling and proposes a method for reliability analysis of the regenerative chatter stability in milling. Dynamic model of milling regenerative chatter is established, and stability lobe diagram is obtained by the full-discretization method (FDM). The neural network is applied to approximate the functional relationship of the limit axial cutting depth; then the reliability is computed with the Monte Carlo simulation method (MCSM) and the moment method (MM), respectively. Finally, the results of an example are used to demonstrate the efficiency and accuracy of the proposed method. [ABSTRACT FROM AUTHOR]