基于两级序优化模型的火控攻击配置方案寻优  

Attack allocation plan optimization of fire control based on two-stage ordinal optimization model

作　　者：周德云[1] 潘潜[1] 李枭扬 张堃[1] 

机构地区：[1]西北工业大学电子信息学院,陕西西安710129

出　　处：《系统工程与电子技术》2017年第1期86-92,共7页Systems Engineering and Electronics

基　　金：国家自然科学基金(61401363);航空科学基金(20155153034)资助课题

摘　　要：火控攻击配置方案寻优是多输入参数的仿真优化问题,针对传统方法在解决此类优化问题时存在的计算量大、耗时长等问题,提出一种基于两级序优化模型的火控攻击配置方案寻优方法。以空空导弹成功发射概率作为评价指标,采用分级寻优策略将寻优过程分为粗糙评估寻优与精细评估寻优。在粗糙评估寻优中,建立基于粒子群-遗传规划的粗糙评估模型,对样本空间所有配置方案快速评估得到优质解集。在精细评估寻优中,采用迭代寻优的方法,逐步提升粗糙评估模型的精度,并缩小优质解集的规模。最后利用仿真模型评估得到满足效能指标要求的最佳配置方案。算例仿真分析表明了算法的有效性与可行性。Attack allocation plan optimization of fire control is a simulation optimization problem with multi- input parameters. For traditional optimization algorithms involve massive calculation and are time consuming, a two-stage ordinal optimization model is proposed. The launching success probability of air-to-air missile is used as the criterion to evaluate the plan, The optimization strategy includes crude and fine evaluation optimization. First, a crude evaluation model based on the particle swarm optimization and genetic programming （PSO-GP） algorithm is established. And all the allocation plans are rapidly evaluated by the crude model to get the high quality solution set. Second, a more refined crude model is established to improve the accuracy of the high-quali- ty solution set step by step. Then the optimal allocation plans are obtained by evaluating the final high-quality solution set with the simulation model. Numerical simulation analysis shows the effectiveness and feasibility of the proposed model.

关 键 词：火控攻击配置方案 仿真优化 两级序优化模型 粒子群-遗传规划算法 

分 类 号：TP202.7[自动化与计算机技术—检测技术与自动化装置]