Here you can find the paper and the available public pages, files, and source code options for the work
The (photovoltaic) PV models' performance is strongly dependent on their parameters, which are mainly standing on the utilized method and the formulated objective function. Therefore, extracting the PV models' parameters under several environmental conditions is crucial for maximizing its reliability, accuracy and reducing the system's overall cost. According to the scope of this problem, several methodologies have been extensively applied to tackle this problem. Thus, this paper presents an enhanced version of the well-known LSHADE (ELSHADE) method by integrating various contributions in the algorithm itself and the objective function to determine three diode PV models' parameters. In ELSHADE, the population is divided into two phases: a robust mutation scheme performs the first phase, and the chaotic-guided strategy is utilized in the second stage. Moreover, an improved Newton Raphson (INR) method is presented to address the I-V curve equation's chaotic behavior effectively. The results confirm that the proposed ELSHADE-INR can precisely find the global solutions by comparing it with state-of-the-art algorithms and its superiority demonstrated in several statistical criteria under real experimental data. The average values of root mean square error (RMSE), mean bias error, determination coefficient, deviation of RMSE, test statistical, absolute error, and CPU-execution time are 0.0060 and 5.88e-05, 0.9999, 2.54e-05, 0.0538, and 0.0042, 11.39s, respectively. We observed that the proposed ELSHADE is robust and stable, and very promising in its origin to obtain high-quality and accurate parameters.
Various DE-based or Differential Evolution algorithms (DE) have been proposed within the last twenty years for many problems. DE is one of the best-established optimizers ever. After a while, a number of successful history-based adaptive DE versions were developed by researchers that apply linear population size reduction (L-SHADE or LSHADE). LSAHDE is considered one of the most powerful methods and one of the most effective evolutionary methods in optimization. In LSHADE, the population size of DE is repeatedly reduced utilizing a linear function. The LSHADE is well-known method a first ranked approach at the IEEE CEC2014 Competition on Real-Parameter Single Objective Optimization. It originally published at title of “Improving the Search Performance of SHADE Using Linear Population Size Reduction”
Download MATLAB source codes of L-SHADE 1.0.1 algorithm (LSHADE)
Download C++ source codes of L-SHADE 1.0.1 algorithm (LSHADE)
Download Java source codes of L-SHADE 1.0.1 algorithm (LSHADE)
Download MATLAB source codes of SHADE 1.1.1 algorithm (SHADE)
Download C++ source codes of SHADE 1.1.1 algorithm (SHADE)
Download Java source codes of SHADE 1.1.1 algorithm (SHADE)
The LSHADE algorithm is a very promising DE variant that ranked as the best algorithm at the IEEE CEC2014 Competition on Real-Parameter Single Objective Optimization. Originally, authors submitted source code of L-SHADE in C++; however, that initial submitted code had a bug in the archive updating step, which caused minor performance degradation. In this website, the shared codes are corrected versions. In the initial paper, they compared the LSHADE with the state-of-the-art DE algorithms SHADE 1.1, CoDE, EPSDE, SaDE, JADE, and dynNP-jDE on the CEC2014 benchmark cases. The results in the initial work show that the simple LPSR method combined with L-SHADE is also fairly powerful, resulting in significant boosting in the efficacy of the SHADE 1.1.
How to cite?
Ridha, Hussein Mohammed, Hashim Hizam, Chandima Gomes, Ali Asghar Heidari, Huiling Chen, Masoud Ahmadipour, Dhiaa Halboot Muhsen, and Mokhalad Alghrairi. "Parameters Extraction of Three Diode Photovoltaic Models using Boosted LSHADE algorithm and Newton Raphson Method." Energy (2021): 120136. https://doi.org/10.1016/j.energy.2021.120136
" The question of whether a computer can think is no more interesting than the question of whether a submarine can swim."
Edsger W. Dijkstra