Advances in Time Series Forecasting Development for Power Systems’ Operation with MLOps

Forecast developers predominantly assess residuals and error statistics when tuning the targeted model’s quality. With that, eventual cost or rewards of underlying business application are typically not considered in model development phase. The analysis power system wholesale market allows us to translate a time series forecast method’s quality its respective value. For instance, near real-time capacity procurement takes place market, which is subject complex interrelations operators’ grid activities balancing parties’ scheduling behavior. Such forecasting tasks can hardly be solved with model-driven approaches because large solution space non-convexity optimization problem. Thus, we generate load forecasts by means data-driven based tool ProLoaF, benchmark state-of-the-art baseline models auto-machine learning auto.arima Facebook Prophet.