Optimal economic dispatch for carbon capture power plants using chaos-enhanced cuckoo search optimization algorithm

Abstract

Accelerated global demand for low carbon operation of power systems have stimulated interest in Low Carbon Technologies (LCTs). The increased deployment of LCTs within power systems is fundamental to the emission abatement of power system. Carbon Capture Power Plant (CCPP) technology has a good potential for future low carbon emission. Existing Economic Dispatch (ED) formulations do not consider the flexibly-operated CCPPs. Flexible operation of Carbon Capture and Storage (CCS) units transforms conventional power plants in such a way that emission output and power output could be separately controlled. The resulting CCPPs have to be optimized in order to take advantage of the incentives available in both power and carbon markets. This thesis proposes an improved mathematical modelling for flexible operation of CCPPs. The developed work possesses simple and practical variables to appropriately model the flexible operation control of the CCPPs. Using this proposed model a new emission-oriented ED formulation is developed. With this new formulation, the thesis also proposes the concept of decoupling the emission and economic outputs and then quantifies its significance for power system operations. In addition to that, a new Metaheuristic Optimization Technique (MOT) named as Chaos-Enhanced Cuckoo Search Optimization Algorithm (CECSOA) has been developed to improve global optimum result for ED problem. The algorithm has been tested using standard test systems with varying degrees of complexity. The results proved that the CECSOA is superior to the existing techniques in terms of ability to obtain global optimal points and the stability of the solutions obtained. Simulation results also showed the possibility of $1.09 million of annual operational cost savings based on a practical power system located in the Greek island of Crete by applying this methodology in comparison with conventional techniques such as Genetic Algorithm. Further results showed that for a carbon price of 20 $/tCO2 and a 60% of system capacity utilization, total emission of a power system is reduced by 10.90% as compared to a “business-as-usual” scenario. In terms of optimal ED for CCPPs, results showed that for carbon prices as low as (~ 8 – 10 $/tCO2), it is economically viable to operate a post-combustion CCS unit.