摘要
针对目前方法构建伪标识判定模型时,由于未能利用独立成分分析方法对Petri网中的数据特征进行提取,导致构建的可达性伪标识判定模型CPU占用率高、网络路径覆盖率低、判定效果差的问题,提出基于约束优化的Petri网可达性伪标识判定模型构建方法。方法首先利用约束优化方法对Petri网中的异常数据进行查找并剔除,依据独立成分分析方法提取Petri网的数据特征;再利用LDA算法对Petri网数据特征进行计算,获取Petri网的边界特征向量;最后基于获取的特征向量,构建Petri网可达性伪标识判定模型。实验结果表明,利用上述方法构建Petri网可达性伪标识判定模型时的CPU占用率低、网络路径覆盖率高、判定效果好。
When constructing the determination model of false identification, the independent component analysis method was often ignored for extracting the data characteristics in Petri net, which leads to a high CPU utilization, low network path coverage and bad determination effect of the constructed model. Therefore, a method for constructing a determination model of false identification for reachability of Petri net based on constrained optimization was presented. Firstly, this method used constrained optimization to find and eliminate the abnormal data in Petri net, and then extracted the data characteristics from Petri net according to the independent component analysis method. Moreover, the method used the LDA algorithm to calculate the data characteristics of Petri net and thus to obtain boundary feature vectors. Finally, based on these feature vectors, a determination model of false identification for reachability of Petri net was built. Experimental results show that the proposed method has low CPU utilization, high network path coverage and good judgment.
作者
王慧英
周恺卿
周辉
WANG Hui-ying;ZHOU Kai-qing;ZHOU Hui(Chongqing Institute of Engineering College of Computer and Internet of Things,Chongqing 400056,China;Jishou University College of Information Science and Engineering,Jishou Hunan 416000,China;Hunan University of Chinese Medicine,Changsha Hunan 410208,China)
出处
《计算机仿真》
北大核心
2022年第6期408-411,416,共5页
Computer Simulation
基金
重庆自然科学基金(201864841314)。
关键词
约束优化
数据去噪
伪标识判定
特征提取
Constrained optimization
Data denoising
Determination of false identification(FIJ)
Feature extraction