Prediction of Harmonic Distortion in Sparsely Monitored Transmission Networks with Renewable Generation

As electrical power networks transition towards net-zero emissions, accurate prediction of harmonics is crucial in ensuring the stability and reliability of the system. This paper presents a novel approach for probabilistic harmonic prediction in transmission networks with limited monitoring, utilizing sequential artificial neural networks (ANNs). The approach proposed a predicting process that incorporates both individual-order and total harmonic distortions (THD). The primary stage involves the development of an ANN model that predicts the individual-order harmonic distortions at buses connected with renewable generators and non-linear loads, leveraging their net power injections. In the subsequent stage, a sequential ANN model is constructed to estimate the probabilistic harmonic distortions at unmonitored buses by utilizing past harmonic measurements obtained from installed power quality monitors. To enhance the performance of the model, a sensitivity analysis was conducted to identify the key parameters that significantly affect accuracy, robustness, and reliability. The results affirm the effectiveness of the proposed methodology in accurately predicting harmonic propagations within power systems with limited monitoring. This capability proves valuable in identifying and mitigating potential harmonic issues, thereby aiding the development of future power networks.