为解决毁伤类电磁目标威胁评估存在专家主观性不易避免、目标类型庞杂、影响因素复杂、时间约束性强等客观问题,建立了毁伤类电磁目标威胁评估指标体系,并提出了与之相适应的云极限学习机(CloudELM)威胁评估方法,该方法借助云模型方法对传统的极限学习机进行改进,有效解决了极限学习机在知识表示方面的缺陷。案例仿真验证了指标体系与评估方法在脱离专家条件下自主评估的有效性,同时利用云支持向量机(CloudSVM)进行对照检验,说明了方法的高精确度与低时间复杂度优势。

In order to solve the objective problem of the threat assessment of the damage electromagnetic target, such as expert subjectivity was not easy to avoid, the target type and the influence factors were complex, the time constraint was strong, the threat assessment index system of damage electromagnetic target was established, and the corresponding threat assessment method of Cloud ELM was proposed. With the help of the cloud model method, the traditional ELM was improved, which could effectively solve the defect in knowledge representation of ELM. The simulation results verified that the effectiveness of independent evaluation of the index system and the evaluation method under the condition of separation from the experts, at the same time, the Cloud SVM was used for contrast test, and the high accuracy and low time complexity of the method were illustrated.