A mathematical model of optimal array reconfiguration (OAR) is proposed for maximizing the power output and minimizing the power deviation between the power output and the frequency regulation (FR) signal of a photovoltaic power plant. The multi-objective bonobo optimizer (MOBO) is designed to rapidly obtain an optimal Pareto front because of its high optimization efficiency. The decision-making method called VIseKriterijumska Optimizacija I Kompromisno Resenje (VIKOR) is employed to determine the best compromise solution from the obtained Pareto front. To verify the effectiveness of the proposed multi-objective optimization, two case studies with constant and varying FR signal are carried out on a 10×10 Total-Cross-Tied (TCT) PV array under partial shading conditions (PSC). The simulation results show that the proposed method can significantly reduce the power deviation of the two objective functions compared to that without optimization. Both power deviations obtained by the proposed method under varying FR signal are only 51.10% and 64.45% of that without optimization. This work is supported by the National Natural Science Foundation of China (No. 61963020).