Abstract: Aiming at the problem that the quality of cable harness wiring quality is difficult to control and judge, the paper puts forward a comprehensive evaluation method of cable harness wiring quality of complex mechatronic products based on artificial neural network, which is complicated and difficult to regulate and unify the complicated mechanical and electrical products. Firstly, the influencing factors of the cable harness wiring process are analyzed, and the quality index system of cable harness wiring is established. Secondly, on the basis of the index system, AHP is used to qualitatively analyze the factors. Finally, the multi-index comprehensive evaluation method based on artificial neural network and AHP method is used to simulate the quality of cable harness wiring. The evaluation results are compared with the expert evaluation results, and the learning ability of the network is analyzed. Using the analysis of a complex mechatronic products, the results not only verify the effectiveness and feasibility of the method, but also show the superiority of artificial neural network in comprehensive evaluation.

Key words: artificial neural network, cable harness wiring quality, quality assessment, complex mechatronic products

摘要： 针对复杂机电产品线缆敷设过程复杂、难以规范和统一而导致的线缆敷设质量难以控制与评判的问题，提出了一种基于人工神经网络的复杂机电产品线缆敷设质量综合评判方法。在对线缆敷设过程中的影响因素分析后建立了线缆敷设质量等级指标体系；在该指标体系的基础上，运用层次分析法（Analytic Hierarchy Process，AHP）对各因素进行定性定量分析，结合专家打分法获得量化的决策结果；采用人工神经网络与AHP法混合的多指标综合评判方法模拟专家对线缆敷设质量进行评价，并将敷设质量的模型评估结果与专家的评估结果进行比较，分析人工神经网络的学习能力。利用某飞机中的线缆敷设数据进行实例分析，验证了方法的有效性和可行性，展现了人工神经网络在综合评价方面的优越性。

关键词: 人工神经网络, 线缆敷设质量, 质量评估, 复杂机电产品