Security Risk Identification Model of Power Grid Software Supply Chain Based on Data Mining

With the promotion of digital transformation of power grid enterprises, software supply chain management is gradually developing in the direction of digitalization and intelligence, but at the same time it is also facing risks such as security loopholes, malicious code implantation and supply chain attacks. In this study, an effective mechanism for real-time monitoring, early warning and identification of supply chain security risks is constructed by comprehensively using data mining technologies such as data preprocessing, feature selection, dimension reduction, anomaly detection, association rule mining and prediction model construction. The model identifies potential risk factors by analyzing the data of bidding, contract performance and material supply, and uses machine learning techniques such as support vector machine (SVM) and random forest to classify and predict. In addition, the model also integrates Apache Flink technology for streaming data processing, realizing real-time monitoring and early warning of power grid software supply chain data. The effectiveness of the model in identifying abnormal code submission, software component update risk and time series abnormal risk is verified by actual case analysis. This study not only improves the security of the power grid software supply chain, but also provides new ideas and methods for the security management of other key infrastructure software supply chains.