Dynamic Risk Control in Electrical Work: A Real-Time Approach for Accident Prevention

With the rapid development of electrical power industry, the electrical system is becoming more and more complicated. The various time-varying high-risk causes of electrical personal accidents are difficult to discover, bringing management challenges to power enterprises. Electrical personal accidents refer to injuries caused by direct or indirect contact with electrical energy. These incidents can occur in various settings, including homes, workplaces, public areas, and pose significant health risks. Preventing electrical personal accidents requires a combination of education, proper equipment use, adherence to safety standards, and regular maintenance of electrical systems. By implementing these strategies, individuals and organizations can significantly reduce the risks associated with electrical accidents. Moreover, traditional methods relied on subjective experience judgement, but have struggled to dynamically quantify and control risks based on accident investigation reports, leading to the development of this novel approach. To this account, a new study published in Reliability Engineering & System Safety and led by Professor Hua Geng from Tsinghua University and conducted by PhD candidate Hengqi Zhang from Tsinghua University alongside Huarong Zeng and Li Jiang from the Electric Power Research Institute of Guizhou Power Grid Co., Ltd., for the first time the researchers adopted the complex system theory to real-time quantify and control risks of electrical personal accidents, and validated the proposed dynamic risk control method for electrical personal accidents, focusing on its applicability and effectiveness in real-world scenarios.

The researchers collected investigation reports of 155 electrical personal accidents from Tsinghua University and the Electric Power Research Institute of Guizhou Power Grid Co., Ltd. These reports were used to construct two types of accident causation networks (ACNs): training ACNs and test ACNs. These networks were based on the domino theory, where causes and consequences occur in a sequence, leading to accidents. A novel aspect of the authors’ work was the evaluation of similarity between the rankings of causes in the training and test ACNs. They introduced a set-based index to measure this similarity, enabling the researchers to assess how well the training data predicted the test data. This step was important for understanding the dynamic nature of risk factors in electrical work and for validating the proposed method’s predictive capability. Moreover, they proposed two dynamic risk control strategies: one without cause constraint and another with cause constraint. The strategy without cause constraint aimed to calculate the cause control compliance rate under an expected risk transmission decline rate. The strategy with cause constraint focused on calculating the accident prevention rate under a maximum cause control rate. These strategies were quantitatively analyzed using node removal experiments in the ACNs.

The authors’ experiments verified the effectiveness of the proposed dynamic risk control method. By analyzing 155 electrical personal accidents, they also demonstrated that the method could dynamically identify high-risk causes and provide effective risk control suggestions. Moreover, the set-based index revealed a high degree of similarity between the training and test rankings, indicating that the method could accurately predict high-risk causes based on past accident data. Furthermore, the study also discussed the influence of network update delay on the effectiveness of the proposed method. It was found that timely updates of the causation networks are crucial for maintaining the accuracy of risk predictions. Additionally, the quantitative analysis provided insights into the effectiveness of the two proposed risk control strategies and showed that both strategies could significantly reduce the risk of electrical personal accidents, with the specific strategy choice depending on the availability of resources and the desired level of risk reduction. In summary, the innovative work of Professor Hua Geng and colleagues demonstrated the potential of the proposed dynamic risk control method to improve safety in the electrical power industry. By providing a framework for the real-time adjustment of risk control measures based on evolving risk factors, the method offers a promising approach to reducing the frequency and severity of electrical personal accidents. Transforming data sources, this method can be extended to other types of accidents. It is also expected to be applied to analysis and control of many dynamical complex systems, such as power systems.