故障率预测模型的训练、故障率预测方法及相关装置

• Format: Patent
• Language: Chinese
• Published: 31.10.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC COMMUNICATION TECHNIQUE; ELECTRIC DIGITAL DATA PROCESSING; ELECTRICITY; PHYSICS; TRANSMISSION

本发明实施例提供了故障率预测模型的训练、故障率预测方法及相关装置，该方法包括：收集引发输电线路故障的环境因素，作为原始统计数据；按照过滤式筛选、迭代寻优的方式，从原始统计数据中提取有关输电线路故障的特征数据，作为原始特征数据；以部分原始特征数据作为训练的样本，训练故障率预测模型；以部分原始特征数据作为测试的样本，对故障率预测模型进行测试。对原始统计数据进行过滤和寻优，可获取与输电线路故障信息关联最为密切的原始特征数据，使用原始特征数据训练故障率预测模型，采用该模型对输电线路的实时故障概率进行计算，可以提高故障率预测的效率，缩减故障率预测的周期，降低故障率预测的成本。 The embodiment of the invention provides a fault rate prediction model training method, a fault rate prediction method and related devices, and the method comprises the steps: collecting environmentalfactors causing a fault of a power transmission line, and enabling the environmental factors to serve as original statistical data; according to a filtering type screening and iterative optimizationmode, extracting feature data related to the power transmission line fault from the original statistical data to serve as original feature data; training a failure rate prediction model by taking partof the original feature data as a training sample; and taking part of the original characteristic data as a test sample, and testing the fa