Optimal Bidding and Operation of a Power Plant with Solvent-BasedCarbon Capture under a CO2 Allowance Market： A Solution with aReinforcement Learning-Based Sarsa Temporal-Difference Algorithm

摘要

In this paper, a reinforcement learning （RL）-based Sarsa temporal-difference （TD） algorithm is applied tosearch for a unified bidding and operation strategy for a coal-fired power plant with monoethanolamine（MEA）-based post-combustion carbon capture under different carbon dioxide （CO2） allowance market con-ditions. The objective of the decision maker for the power plant is to maximize the discounted cumulativeprofit during the power plant lifetime. Two constraints are considered for the objective formulation. Firstly,the tradeoff between the energy-intensive carbon capture and the electricity generation should be made un-der presumed fixed fuel consumption. Secondly, the CO2 allowances purchased from the CO2 allowance mar-ket should be approximately equal to the quantity of COs emission from power generation. Three case stud-ies are demonstrated thereafter. In the first case, we show the convergence of the Sarsa TD algorithm andfind a deterministic optimal bidding and operation strategy. In the second case, compared with the inde-pendently designed operation and bidding strategies discussed in most of the relevant literature, the SarsaTD-based unified bidding and operation strategy with time-varying flexible market-oriented CO2 capturelevels is demonstrated to help the power plant decision maker gain a higher discounted cumulative profit.In the third case, a competitor operating another power plant identical to the preceding plant is consideredunder the same CO2 allowance market. The competitor also has carbon capture facilities but applies a differ-ent strategy to earn profits. The discounted cumulative profits of the two power plants are then compared,thus exhibiting the competitiveness of the power plant that is using the unified bidding and operation strat-egy explored by the Sarsa TD algorithm.