Cloud-edge collaborative load library self-learning method

The invention discloses a cloud-edge collaborative load library self-learning method, which comprises the following steps of S1, obtaining a typical load curve of a multi-element load, and constructing a cloud load library; s2, comparing the load recognition model based on the load curve similarity with the load recognition model based on machine learning, and establishing a typical load model; s3, identifying an access load through a typical load model, and establishing a far-end load library self-learning model; according to the method, after the access load is identified, the multi-section measurement data of the access load is acquired, and the cloud load library self-learning model based on the deviation degree is established, so that the self-learning of the cloud load library is realized, and the accuracy of the load model is improved. 本发明公开了一种云边协同的负荷库自学习方法，包括如下步骤：S1：获取多元负荷的典型负荷曲线，构建云端负荷库；S2：对比基于负荷曲线相似度的负荷识别模型和基于机器学习的负荷识别模型，建立典型负荷模型；S3：通过典型负荷模型识别接入负荷，建立远端负荷库自学习模型；本发明通过识别接入负荷后，获取接入负荷的多断面量测数据，建立基于偏离度大小的云