Robustness of the Methodology for Harmonic Forecasting in Transmission Networks
摘要
Building on the previous chapter, this chapter delves deeper into the forecasting capabilities of the sequential ANN models. It presents a thorough analysis of the models’ performance in forecasting future harmonic distortions, considering a range of influencing factors such as variations in power injections and network topology. Sensitivity analysis sheds light on the key parameters affecting forecasting accuracy, offering insights into model optimization. The chapter also explores the models’ robustness through various case studies, including scenarios with limited PQ monitoring and increased harmonic orders. The findings demonstrate the models’ effectiveness in providing reliable harmonic forecasting, thereby supporting practical network management and the mitigation of potential harmonic-related issues.