Power quality monitoring of a low voltage distribution network
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
A rising population creates a high demand for electricity which leads to network failures, the Low Voltage (LV) network to be precise since is directly impacted by the consumer loads. As the usage of LV networks continues to grow, LV monitoring needs to become more widespread and detailed so that network operators can see accurate and real-time information. Research has been done over the years to find efficient ways of monitoring and assessing the power quality of LV networks. This project explored the use of software development tools for power quality monitoring and analysis. Models were designed to create the disturbances of interest in power systems. Simulated power systems were decomposed using the wavelet transform toolbox in MATLAB to attain their energy densities. The obtained energy densities were input into a trained neural network in MATLAB to detect and classify fault types with respect to the parameter deviation in the power quality. An accuracy of 92.3% correct classification was achieved for the neural network training process. In MATLAB Server Pages (MSP), once all processes were completed, the results of the analysis were sent to a database. The results were then made accessible via the internet by creating a login system for users.