基于混合神经网络的野战仓库伪装效能评估

Camouflage Effect Evaluation of Field Depot based on Hybrid Neural Network

随着现代高技术的发展及其在军事领域的广泛应用,战场透明度明显增加。在未来战场上,发现即意味着被摧毁。野战仓库的生存也面临前所未有的严重威胁,伪装就成为其提高自身生存能力的有效手段。伪装效果的好坏将直接影响其战场生存力。采用遗传算法改进BP神经网络模型,分析了其基本思路以及具体实施步骤,并运用其对野战仓库伪装效能进行评估。实例表明,该模型能够较好地克服人为因素和模糊随机性的影响,评估结果更为科学可信。

Camouflage is an effective measure to improve survivability of the field depot. Genetic algorithm is used to improve BP （Back Propagation） neural network model, the paper analyses the basic thought and practical approach and applies this model to evaluate the effect of the field depot camouflage.The example shows that this model can overcome the influences which are caused by human factor and fuzzy randomness. The evaluation result is more scientific and credible.