Abstract
Solar-assisted combined cooling, heating, and power (CCHP) systems offer a sustainable solution to address the escalating energy demand resulting from economic and population growth. This advancement has facilitated a reduction in excessive fuel consumption, thereby mitigating the associated impacts of climate change and global warming. These multi-generation systems undeniably possess the advantage of fuel efficiency, leading to enhanced overall system efficiency compared to individual cooling, heating, and power generation systems. However, optimizing these systems is imperative to realize optimal system design configurations that enhance their thermodynamic performance indicators. Hence, this research study aims to apply multi-objective optimization techniques to CCHP systems.
This work explored the use of three multi-objective optimization approaches to optimize the performance of a solar-assisted combined cooling, heating, and power system. It aimed to obtain optimal configurations comprising decision variables such as compression ratio, pinch point temperature difference, gas turbine inlet temperature, and combustion chamber inlet temperature to maximize the net power and exergy efficiency while minimizing CO2 emission. This study also investigated the effects of each decision variable on the objective function of the combined cooling, heating, and power (CCHP) system via sensitivity analysis, understand the convergence trend of each decision variable, and evaluate its results with that obtained from related works.
This was carried out by first defining and initializing the objective functions, upper and lower boundaries of the decision variables as well as other relevant parameters. The randomly selected candidate solutions by the respective optimization algorithm were iteratively improved using techniques such as leader election, leader-prey, dynamic update, pack-hunting approaches, etc. Furthermore, evaluation processes were carried out by computing the objective values and the roulette wheel was predominantly used to select the new sets of solutions according to their fitness values. The optimization was carried out until the termination criteria which is the maximum number of iterations was attained and the results were analyzed with Pareto fronts presented to provide insight to the decision maker. The grey wolf, Harris Hawks, and Antlion optimization techniques each generated 100 Pareto solutions, representing optimal trade-offs among conflicting objective functions. These solutions
3
offer decision-makers a diverse array of options. Analysis of the Pareto fronts revealed that minimizing CO2 emissions contributes significantly to system efficiency while maximizing net power output does not necessarily result in higher efficiency. Sensitivity and convergence analyses indicated that optimizing a solar-assisted CCHP system involves minimizing compression ratio, pinch point temperature difference, and inlet combustion chamber temperature while maximizing gas turbine inlet temperature. Additionally, the compression ratio most significantly impacts CO2 emissions, gas turbine inlet temperature affects exergy efficiency the most, and inlet combustion chamber temperature has the greatest influence on net power output.
Another set of six optimal solutions was generated by each optimization technique for validation purposes. The results obtained revealed that the Antlion optimization technique obtained an optimal configuration that yielded the best CO2 emission and exergy efficiency of 49.40 gr/MJ and 45.45 % respectively. It also had the best runtime value of 3.02 seconds but produced a relatively lower net power output compared to the other methods. Also, the Harris Hawks had an optimal configuration that yielded the best net power output of 61.91 MW, but it also produced relatively higher CO2 emission and exergy efficiency. The grey wolf technique, on the other hand, attempted to find the right balance between net power and exergy efficiency that had conflicting objectives. This is because, in contrast to the other optimization methods, it simultaneously achieved a relatively high exergy efficiency and net power output. However, it had the highest run-time value of 145.74 seconds. Furthermore, it was observed that the three meta-heuristic optimization techniques employed in this study produced, on average, reduced CO2 emission, and high exergy efficiency, with negligible reduction in the net power (standard deviation of 0.39) when compared to the response surface method used for evaluation purposes. Hence, the optimization approaches presented in this study are all suitable for the multi-objective optimization of a solar-assisted CCHP system.