Research on the location of space debris impact spacecraft based on genetic neural network

The existence of space debris will pose a serious threat to the safety of spacecraft in orbit and cause serious impact damage to spacecraft. Therefore, the timely and accurate determination of the impact location of space debris is very important to improve the safety and reliability of the orbiting spacecraft in the long run. This paper proposes an intelligent back-propagation artificial neural network-based algorithm for locating the impact source of space debris. The method of building and training the algorithm is described, based on the moment of arrival of the stress waves detected by the PVDF (Polyvinylidene Fluoride) sensor array and the sensor coordinates. Based on the ABAQUS finite element simulation data, the neural network is trained. The result shows that compared with the traditional TOA (Time of Arrival) localization algorithm, the localization accuracy of the BP (Back Propagation) neural network localization algorithm is improved by 34.7%. The optimal objective is to minimize the error between the space debris impact point and the localization point. The parameters of the PVDF sensor array (diameter d of the circular PVDF thin film sensor, radius R of the PVDF sensor array) were optimized by genetic algorithm, and the optimal sensor array parameters were obtained based on the optimization algorithm.