Worst CVaR based energy management for generalized energy storage enabled building-integrated energy systems

Abstract

Generalized energy storage (GES) can reduce the renewable energy curtailment, carbon emission and operational cost risk in the building-integrated energy system (BIES) energy management when faced with uncertainties of multi-energy loads and PV. In this context, GES integrated with the differentiated building thermal load-based virtual energy storage (VES) is proposed in the energy management model of BIES. Specifically, the differentiated building thermal load VES is introduced by differentiated personal factors (e.g., metabolic rate and clothing insulation) based on the predicted mean vote. Furthermore, a worst conditional value at risk (WCVaR) energy management method is developed to evaluate the operational cost risk of BIES. Meanwhile, possible probability distribution uncertainties of PV, electric and thermal loads are all fully considered by a boxlike uncertainty set. Then, the proposed WCVaR model is transformed into a linearized model by the Lagrange duality theory. Finally, the whole energy management model is solved by the CPLEX solver. Case studies and comparative analysis are performed based on a representative BIES. Numerical results show that the proposed model can reduce 33.55% of operational cost, 74.25% of carbon trading cost, and 16.96% of PV curtailment cost for BIES compared with that without GES. In addition, the conditional value at risk (CVaR) with the proposed WCVaR method ($5700.22) is between that with the traditional CVaR method considering multiple operational scenarios ($5684.68) and that with the traditional CVaR method only including the worst operational scenario ($6043.48), which indicates that the proposed WCVaR method can avoid overly-high and overly-low operational cost risks.