Liu Fei-Fei, Chu Shu-Chuan, Hu Chia-Cheng, Watada Junzo, Pan Jeng-Shyang
College of Computer Science and Engineering, Shandong University of Science and Technology, Qingdao, 266590, China.
College of Science and Engineering, Flinders University, 1284 South Road, Tonsley, SA 5042, Australia.
Heliyon. 2023 Jun 23;9(6):e16468. doi: 10.1016/j.heliyon.2023.e16468. eCollection 2023 Jun.
The traditional parameter estimation methods for photovoltaic (PV) module are strictly limited by the reference standards. On the basis of the double diode model (DDM), this paper proposes a modified PV module that is independent of the reference conditions and can be used for the transformation and reconfiguration of PV module. With respect to the issue of the slow convergence precision and the tendency to trap in the local extremum of the QUATRE algorithm, this research incorporates the QUATRE algorithm with recombination mechanism (RQUATRE) to tackle the problem of parameter estimation for the improved PV modules described above. Simulation data show that the RQUATRE wins 29, 29, 21, 17 and 15 times with the FMO, PIO, QUATRE, PSO and GWO algorithms on the CEC2017 test suite. In addition, in a modified PV module for the parameter extraction problem, the final experimental results achieved a value of 2.99 × 10 at RMSE, all better than the accuracy values of the compared algorithms. In the fitting process of IAE, the final values are also all less than 10%, which can satisfy the fitting needs.
传统的光伏（PV）模块参数估计方法受到参考标准的严格限制。基于双二极管模型（DDM），本文提出了一种改进的光伏模块，该模块独立于参考条件，可用于光伏模块的变换和重构。针对QUATRE算法收敛精度慢和易陷入局部极值的问题，本研究将QUATRE算法与重组机制（RQUATRE）相结合，以解决上述改进光伏模块的参数估计问题。仿真数据表明，在CEC2017测试套件上，RQUATRE算法在与FMO、PIO、QUATRE、PSO和GWO算法对比时，分别有29次、29次、21次、17次和15次获胜。此外，在改进的光伏模块参数提取问题中，最终实验结果的均方根误差（RMSE）值达到2.99×10，均优于对比算法的精度值。在积分绝对误差（IAE）的拟合过程中，最终值也均小于10%，能够满足拟合需求。
