基于多属性决策的二次系统信息质量评价方法

Information Quality Evaluation Method of Secondary System Based on Multi-Attribute Decision Making

信息质量评价是信息质量管理的重要组成部分,智能变电站二次系统信息质量的评价方法是否合理对评价结果的准确性至关重要。传统的信息质量评价方法存在评价方法复杂、可行性不高、维度单一和覆盖面不全等问题。在此背景下,文章首先构建信息质量评价指标体系;然后将基于改进层次分析法(analytic hierarchy process,AHP)得到的指标主观权重和基于Critic(Criteria importance through intercriteria correlation)法得到的指标客观权重进行耦合,确定指标的综合权重,得到综合评价结果;其次,为了实现信息质量预测,以综合评价值作为先验样本,对基于遗传算法(genetic algorithms,GA)改进的多层前馈神经网络(back propagation neural network,BPNN)进行训练与测试,利用拟合好的网络实现信息质量预测;最后,通过算例分析验证了该方法的准确性和可靠性。

Information quality evaluation is an important part of information quality management.Whether the evaluation method of information quality of smart substation secondary system is reasonable or not is very important to the accuracy of evaluation results.The traditional information quality evaluation methods have some problems,such as complex evaluation methods,low feasibility,single dimension and incomplete coverage.In this context,this paper first constructs an information quality evaluation index system.Then,the subjective weight of the index based on the improved analytic hierarchy process(AHP)and the objective weight of the index based on the Critic(criteria importance through inter criteria correlation)method are combined to determine the comprehensive weight of the index and obtain the comprehensive evaluation results.Secondly,in order to realize the prediction of information quality,taking the comprehensive evaluation value as a priori sample,the back propagation neural network(BPNN)improved by genetic algorithms(GA)is trained and tested,and the fitted network is used to realize the prediction of information quality.Finally,an example is given to verify the accuracy and reliability of the method.