BP neural network ammunition storage reliability evaluation method based on improved ant colony optimization

Abstract

The invention relates to a BP neural network ammunition storage reliability evaluation method based on improved ant colony optimization, which mainly improves the accuracy and robustness of ammunitionstorage reliability assessment, weakens the problem that an ant colony algorithm is easy to fall into a local optimal solution, and makes up the defects that the BP neural network is easy to fall into a local minimum and the algorithm result is unstable. The method comprises the following planning steps of establishing a neural network prediction model by utilizing a change rule of ammunition storage reliability data; introducing a progeny ant pheromone contribution factor into the ant colony algorithm, and reasonably adjusting the pheromone concentration contributed by the progeny ant colony; and optimizing the weight and the threshold of the BP neural network by utilizing the global search capability of the improved ant colony algorithm, optimizing network structure parameters through an optimization process of ant colony foraging and exploring the shortest route, and carrying out ammunition storage reliability evaluation. The pheromone is updated through the improved method, the problem of a local optimal solution is effectively avoided, and the problem that the BP neural network is likely to fall into a local minimum is solved through intelligent optimization of the weight andthe threshold of the BP neural network.

Description

technical field [0001] The invention relates to the technical field of ammunition storage reliability, in particular to an evaluation method for ammunition storage reliability based on an improved ant colony algorithm to optimize BP neural network. Background technique [0002] Ammunition has the characteristics of "long-term storage and one-time use", which makes the reliability of ammunition storage an important technical index to measure the quality of ammunition. Scientific and reasonable evaluation of the reliability of ammunition in storage is an important guarantee for the combat effectiveness of the army. Military powers such as the United States and the Soviet Union have long realized the importance of the reliability of ammunition storage, and have formulated a series of ammunition reliability standards. In recent years, scholars at home and abroad have conducted a lot of research on the evaluation methods of ammunition storage reliability, including model-based ev...

AI Technical Summary

Problems solved by technology

[0005] Since the ACO algorithm has not been developed for a long time, the theoretical basis and application promotion still need to be further researched. There are very few related literatures at home and abroad that combine the ACO algorithm and BP neural network for the evaluation of storage reliability; especially the improvement of the ACO algorithm Optimal BP neural network is used in this special field of ammunition storage reliability evaluation, and there is no relevant literature at home and abroad.

Images