Non-invasive load identification method based on ABC-SVM algorithm and transient feature

Abstract

The insertion of non-invasive load sensing technology into the power supply entrance is conducive to promoting energy saving construction, power grid actual load prediction, the development of intelligent energy buildings, and the completion of intelligent energy network system construction. Based on this, a load discrimination method based on Artificial Bee Colony algorithm optimized Support Vector Machine (ABC-SVM) was proposed. First, the current signal of the main line is tested through events. After the transient event is detected, the negative charge transient current waveform of the target is separated, and its feature are extracted. Then, the features are input into the pre-trained ABC-SVM model for classification and recognition. In order to improve the performance of the classifier, particle swarm optimization algorithm was used to optimize the parameters of ABC-SVM classifier. Experimental results show that the recognition accuracy of this method is up to 97.69%, and the sample recognition speed is 1.53 μs, which has certain practicability.

Keywords

• Intelligent energy network system
• Load identification
• Transient event
• Artificial bee colony algorithm
• Support vector machine