矿区架空线路运行环境恶劣，易受到外界因素影响而导致线路损坏，需要准确评估矿区架空线路外破风险水平。然而现有的定性评估方法存在主观性强、评价结果可比性差等缺点，定量评估方法虽然有着较高的客观性，但其准确评估的基础是大量高质量数据。为均衡评估结果的客观性和评估数据的获取难度，采用半定量评估方法中的作业条件危险性分析（LEC）法，并基于矿区线路实际运行环境对其进行改进，提出了一种基于改进LEC法的矿区架空线路外破风险评估方法。首先，通过分析矿区架空线路的实际运行环境，明确主要的外破风险要素，构建矿区架空线路外破风险评估指标体系。其次，借助基于YOLOv5的图像识别策略来辨识线路环境中的外破风险源，实现线路外破风险数据的实时获取，克服了传统LEC法人工获取的数据实时性较差、数据量不足的缺点。然后，改进了LEC法的要素赋值规则，基于图像辨识结果对要素进行赋值，实现线路外破风险的实时评估，提高了评估结果的客观性，解决了传统LEC法的要素赋值依赖评价者个人经验的弊病。最后，为衡量各类风险带来的叠加影响，利用层次分析法确定各风险评价指标权重，最终实现矿区架空线路外破风险的综合评估。结合露天煤矿实际运行过程中某场景进行案例分析，结果表明该方法能对具体场景中的架空线路外破风险等级进行有效评估。

The operating environment of overhead lines in mining areas is harsh. The lines are easily affected by external factors, leading to line breakage. It is necessary to accurately evaluate the risk level of external breakage of overhead lines in mining areas. However, existing qualitative evaluation methods have shortcomings such as strong subjectivity and poor comparability of evaluation results. Although quantitative evaluation methods have high objectivity, the accurate evaluation is based on a large amount of high-quality data. In order to balance the objectivity of the evaluation results and the difficulty of obtaining evaluation data, the likelihood exposure consequence (LEC) method in semi quantitative evaluation method is adopted. Based on the actual operating environment of mining area lines, an improved LEC method is proposed for the risk assessment of external breakage of overhead lines in mining areas. Firstly, by analyzing the actual operating environment of overhead lines in mining areas, the main risk factors of external breakage are identified. The risk assessment index system for external breakage of overhead lines in mining areas is constructed. Secondly, using the YOLOv5 based image recognition strategy to identify the sources of external breakage risk in the line environment, real-time acquisition of external breakage risk data of the line is achieved. It overcomes the shortcomings of poor real-time performance and insufficient data volume obtained manually by traditional LEC methods. Thirdly, the element assignment rules of the LEC method are improved. The elements are assigned based on image recognition results to achieve real-time evaluation of the risk of external breakage to the line. It improves the objectivity of the evaluation results and solves the problem of the traditional LEC method's element assignment relying on the personal experience of the evaluator. Finally, in order to measure the superimposed impact of various risks, the analytic hierarchy process is used to determine the weight of each risk evaluation index. Ultimately, the comprehensive assessment of the risk of external breakage to overhead lines in mining areas is achieved. A case study is conducted during the actual operation of an open-pit coal mine. The results show that this method can effectively evaluate the risk level of external breakage of overhead lines in specific scenarios.