A Self-adaptive Digital Twin with Broad Learning System: An Example of Heat Pump

This paper introduces a novel self-adaptive digital twin (DT) based on broad learning system (BLS), which has potential to be evolved in the power and energy sectors. Traditional data-driven DT approaches in these sectors struggle with the requirement for extensive historical data and flexibility in adapting to changes in operating conditions. By integrating BLS, our method notably decreases the volume of initial training data required and improves the system's ability to adjust to new conditions uncovered in initial training data. As an example, the proposed method is applied on a 5 kW air-source heat pump system. Finally, the effectiveness of the proposed method is demonstrated through comparison with a benchmark model calibrated with experimental data.