Product modeling design based on genetic algorithm and BP neural network

Jia Xuan Han, Min Yuan Ma, Kun Wang

研究成果: Article同行評審

47 引文 斯高帕斯(Scopus)


At present, the rapid development of industrial products still lacks reliable theoretical support in terms of styling design. In order to provide a set of effective reference basis for designing a better product appearance plan, this paper takes the shape design of drones as an example. The optimization feature of genetic algorithm optimizes the BP neural network to construct a hybrid GA–BP model, so as to efficiently evaluate and screen out scientific design schemes. By adding 13 of the 16 selected product design schemes to the hybrid GA–BP evaluation system, we perform training to obtain simulated and actual values, and finally, the remaining three design schemes are used for verification. Our results show that the relative errors of the two sets of data verification are 3.4%, 1.9% and 3.1%, respectively. In theory, such accuracy is very high, which basically reflects that the evaluation system of hybrid GA–BP product modeling design enables the design plan to be evaluated quickly, conveniently, effectively and scientifically.

頁(從 - 到)4111-4117
期刊Neural Computing and Applications
出版狀態Published - 2021 5月

All Science Journal Classification (ASJC) codes

  • 軟體
  • 人工智慧


深入研究「Product modeling design based on genetic algorithm and BP neural network」主題。共同形成了獨特的指紋。