Product modeling design based on genetic algorithm and BP neural network

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.