摘要
采用层次分析法,从重大报错风险、审计人员风险、审计程序风险3个维度构建了电力企业审计风险评估指标,对BP神经网络初始权值、阈值进行优化,提出了电力企业审计风险评估的BSA-BP模型。将提出的模型和专家评估结果进行对比,2种方法保持了一致性。采用BAS-BP神经网络找出了电力企业审计工作开展过程中的薄弱环节,采取针对性的措施进行改进。2次审计评估结果表明,采取针对性的措施改进后,企业的审计风险明显降低。
Using the analytic hierarchy process,this paper constructed the audit risk assessment index of power enterprises from three dimensions of major error risk,auditor risk and audit procedure risk,optimized the initial weights and thresholds of BP neural network,and proposed the BSA-BP model for the audit risk assessment of power enterprises.Comparing with the proposed model with the expert evaluation results,the two methods were consistent.In addition,BAS-BP neural network was used to find out the weak links in the audit work of power enterprises,and targeted measures were taken to improve it.The results of the second audit evaluation showed that after taking targeted measures to improve,the audit risk of enterprises was significantly reduced.
作者
王鑫根
WANG Xingen(Guangzhou power trading center of China Southern Power Grid Co.,Ltd,Guangzhou 510800,China)
出处
《粘接》
CAS
2023年第4期187-191,共5页
Adhesion
关键词
BP神经网络
天牛须搜索算法
审计风险评估
电力企业
BP neural network
beetle antennae search algorithm
audit risk assessment
electric power enterprises
