Power Equipment Operation Classification System

Abstract

With the rapid development of technology, the demand for electricity and power equipment has increased significantly. Monitoring the operational status of power equipment is crucial for improving energy management efficiency and ensuring safety. This study focuses on analyzing power consumption data to automate the classification of power usage categories. The method employs a pre-trained neural network model and data normalizers to preprocess and predict new power consumption data. Key features such as voltage, current, and reactive power are extracted and standardized. A neural network classification model is built using PyTorch, trained on a labeled dataset, and evaluated on both training and test sets.

The experimental results demonstrate the effectiveness of the proposed system in distinguishing different operational states of power equipment. By leveraging K-means clustering for label generation and a three-layer fully connected neural network for classification, the model achieves an accuracy of 99.48% on the test set. This high performance highlights the system's potential for practical applications in energy management, offering a reliable and automated solution for power equipment monitoring.

Keywords: power consumption prediction, neural network, data standardization, PyTorch