基于混合遗传BP神经网络的城市系统作战能力评估
发布时间：2018-08-19 08:41
【摘要】：针对现有城市系统作战能力评估方法较少的问题,利用反向传播(back propagation,BP)神经网络在能力评估方面所具有的自适应、自学习、强容错性和泛化映射等优势,建立了评估指标体系并给出了指标的隶属函数。通过模拟退火遗传算法(simulated annealing and genetic algorithm,SAGA)优化BP神经网络的连接权重和阀值,弱化了指标评价中的人为因素,提高了评价结果的准确性、客观性和权威性,有效解决了传统遗传算法和BP神经网络易陷入局部极小值、收敛速度慢和抗干扰能力差等问题。仿真实例验证了该方法对城市系统作战能力评估的可行性和有效性。
[Abstract]:In order to solve the problem that there are few methods for evaluating the operational capability of urban systems, the advantages of adaptive, self-learning, strong fault-tolerance and generalization mapping of back propagations-BP neural network are used in this paper. The evaluation index system is established and the membership function of the index is given. The connection weight and threshold value of BP neural network are optimized by simulated annealing genetic algorithm (simulated annealing and genetic algorithm saga), which weakens the human factors in index evaluation and improves the accuracy, objectivity and authority of evaluation results. The traditional genetic algorithm and BP neural network are easy to fall into local minima, slow convergence speed and poor anti-interference ability. A simulation example is given to verify the feasibility and effectiveness of this method in evaluating the operational capability of urban systems.
【作者单位】： 火箭军工程大学初级指挥学院;
【基金】：国家自然科学基金(61372167,61379104)资助课题
【分类号】：TP18
，
本文编号：2191142
[Abstract]:In order to solve the problem that there are few methods for evaluating the operational capability of urban systems, the advantages of adaptive, self-learning, strong fault-tolerance and generalization mapping of back propagations-BP neural network are used in this paper. The evaluation index system is established and the membership function of the index is given. The connection weight and threshold value of BP neural network are optimized by simulated annealing genetic algorithm (simulated annealing and genetic algorithm saga), which weakens the human factors in index evaluation and improves the accuracy, objectivity and authority of evaluation results. The traditional genetic algorithm and BP neural network are easy to fall into local minima, slow convergence speed and poor anti-interference ability. A simulation example is given to verify the feasibility and effectiveness of this method in evaluating the operational capability of urban systems.
【作者单位】： 火箭军工程大学初级指挥学院;
【基金】：国家自然科学基金(61372167,61379104)资助课题
【分类号】：TP18
，
本文编号：2191142
本文链接：https://www.wllwen.com/kejilunwen/zidonghuakongzhilunwen/2191142.html
