Optimal bidding strategies of advanced adiabatic compressed air energy storage based energy hub in electricity and heating markets

Integration of power and heating systems can not only improve energy efficiency but also reduce the peak generation capacity by narrowing the gap between peak and valley demands. Advanced adiabatic compressed air energy storage (AA-CAES) is a large-scale and environmental-friendly storage technology that can supply heat and power. It can be adopted as an energy hub that integrates electricity and heating systems. This paper studies the strategic behavior of an AA-CAES based energy hub in deregulated electricity and heat markets. Firstly, a heat market clearing model with elastic thermal demands is developed, and the impact of demand elasticity on the stability of the heat market is analyzed. Secondly, based on the proposed heat market model, a bi-level optimization model is established to explore the strategic bidding and scheduling of the merchant AA-CAES based energy hub in both markets. Through exploiting the weakly coupling structure of the bi-level problem, a special decomposition method is proposed to compute the market equilibrium. The optimal response of the lower-level power/heat market clearing problem is analytically expressed by multi-parametric programming techniques, where the bidding strategies are parameters. With the lower-level parameterized solution, the original complex joint optimization problem is transformed into a small-scale mixed integer quadratic program with only a few binary variables, and thus can be solved efficiently. Case studies verify the proposed model and method. Results show that the AA-CAES based energy hub/heat market is conducive to an 8.6%/22.5% reduction in the operation cost of the heat market/energy hub.

Automated summary

This paper studies the strategic behavior of an AA-CAES based energy hub in deregulated electricity and heat markets, and proposes a special decomposition method to compute the market equilibrium. Case studies verify the proposed model and method.