Error Propagation Model Using Jacobian-Torsor Model Weighting for Assembly Quality Analysis on Complex Product

Abstract

The assembly quality of a complex product is the result of the combined effects of multiple manufacturing stages, including design, machining and assembly, and it is influenced by associated elements with complex coupling mechanisms. These elements generate and transmit assembly quality deviations during the assembly process which are difficult to analyze and express effectively. Current studies have focused on the analysis and optimization of the assembly surface errors of single or few components, while lacking attention to the impact of errors on the whole product. Therefore, in order to solve the above problem, an assembly quality deviation analysis (AQDA) model is constructed in this paper to analyze the deviation transfer process in the assembly process of complex products and to obtain the key features to optimize. Firstly, the assembly process information is extracted and the assembly quality network model is established on the basis of complex networks. Second, the Jacobian-Torsor (J-T) model is introduced to form a network edge weighting method suitable for the assembly process to objectively express the error propagation among product part features. Third, an error propagation model (EPM) is designed to simulate the error propagation and diffusion processes in the assembly network. Finally, the assembly process of an aero-engine fan rotor is used as an example for modeling and analysis. The results show that the proposed method can effectively identify the key assembly features in the assembly process of complex products and determine the key quality optimization points and monitoring points of the products, which can provide a decision basis for product quality optimization and control.