Assessment of Power Quality Events with Arduino Technology and Machine Learning

With the employment of electronic devices that are harmonic sources, power quality (PQ) disruptions are increasing day by day in the automation era. These PQ disruptions have the greatest impact on both consumers and suppliers. Monitoring and analyzing the electrical network’s PQ is critical for the power system’s stability and reliability. It also aids in the reduction of significant financial losses to consumers and serious system equipment failures. Existing high-end PQ analyzers are both expensive and inconvenient to use. It is not practical to utilize such large and expensive analyzers for PQ monitoring in our daily lives or for research purposes. The use of Arduino technology to conduct PQ analysis of varied loads is proposed in this research. To read the voltage and current from sensors, we use the Arduino platform as a data-collecting system. Furthermore, the Arduino fast Fourier transform (FFT) library on Arduino allows for the implementation of floating-point FFT operations. The harmonics and total harmonic distortion (THD) are calculated using FFT analysis. To identify and detect the various PQ events from the acquired data, machine learning (ML) techniques such as artificial neural network (ANN) and support vector machine (SVM) are employed. As a result, the proposed technique is employed to construct a user-friendly, simple, and smart PQ analyzer. The proposed methodology was evaluated on a hardware prototype, and different PQ characteristics such as power factor, THD, and classification of disturbances such as sag and swell were calculated. Furthermore, the results are compared and confirmed using MATLAB and Simulink software, as well as an error analysis of the data obtained to determine accuracy. The novelty behind this work is that PQ analyzer is developed in software platform and analyzed with ANN and SVM. The same Simulink model is developed in hardware with Proteus circuit design and is verified. It is found that the proposed design of PQ analyzer perfectly detects the PQ events.