Aspects of effectiveness and significance: The use of machine learning methods to study CuIn1-xGaxSe2 solar cells

The goal of this work is to enhance the efficiency of CuIn1-xGaxSe2 (CIGS) thin film solar cells by investigating the critical factors affecting their device performance and the correlations between them. To achieve this goal, machine learning algorithms are employed to uncover the primary parameters and correlations affecting CIGS solar cell device performance. The experimental data is used to develop the data sets for machine learning analysis. The correlation studies allow for the investigation of the key factors governing device performance. The algorithms used in the study include linear regression (LR), random forest (RF), extreme gradient boosting (XG), decision tree (DT), support vector machine regressor (SVM), stochastic gradient descent regressor (SGD) and Bayesian ridge (Bayesian). The results showed that decision trees provide the most accurate predictions of CIGS solar cell efficiency, with a root mean square error of 0.11 and 1.83 and a Pearson coefficient of 0.9 and 0.88 for the training and test data sets, respectively. Additionally, this research provides important insight into the necessary components and ideal device dimensions, offering helpful guidelines for subsequent experimenting optimization endeavours.

Automated summary

The aim of this research is to improve the efficiency of CuIn1-xGaxSe2 (CIGS) thin film solar cells by investigating the factors affecting their device performance and their correlations. Machine learning algorithms are used to identify the primary parameters and correlations influencing CIGS solar cell performance. Various algorithms, such as linear regression, random forest, and decision tree, are employed in the study. The results show that decision trees provide the most accurate predictions of CIGS solar cell efficiency. The research also provides valuable insights into the necessary components and ideal device dimensions, offering useful guidelines for future optimization efforts.