RBF模糊神经网络在舰载C^3I系统效能评估中的应用

Application of RBF fuzzy neural network in effectiveness evaluation of shipboard C^3I system

针对舰载C3I系统指标体系效能评估中存在的影响因素多、因素之间关系不明确和数据量庞大等问题,提出了一种径向基函数神经网络和模糊推理规则相结合的效能评估方法,分析并建立了基于RBF模糊神经网络学习算法的舰载C3I系统作战效能评估模型。利用400组指标权值数据对RBF模糊神经网络进行训练和测试,验证了该模型的有效性和可行性。在此基础之上,对选取的13组作战效能一级指标权值数据进行3次仿真实验。结果表明,该方法能够判定舰载C3I系统作战效能指标权值是否合理分配,为舰载C3I系统的作战效能评估方法提供了一种新的途径,也为决策者根据战场态势进行动态决策提供有效的量化验证。

To solve such problems as various influencing factors, undefined relationships between factors and huge data sizes, the effectiveness evaluation method was proposed, combined with radial basis function （RBF） neural network and fuzzy reasoning rules. Based on the learning algorithm of RBF fuzzy neural net- work, the operational effectiveness evaluation model of shipboard Ca I system was analyzed and estab- lished. The RBF neural network was trained and tested by 400 groups of fuzzy index weight data. The ex- perimental results indicate that this model is effective and feasible. With this understanding, three experi- ments were conducted for simulating 13 groups of the first-order index weights with the trained fuzzy RBF neural network. The results show that this method can judge the reasonability of the operational effective- ness index weights allocation of shipboard C3I system. It has exploited a new evaluation method for the operational effectiveness of shipboard Ca I system. Meanwhile,it may provide effective and quantitative val- idation for the decision makers to make policy according to the battlefield situations.