الفهرس 
Chapter (1) INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 161 |  |  |
| | | from | 161 | | |
Abstract
Machine learning has become an important tool in many areas, including the prediction and optimization of solar energy systems, energy storage devices, mobile data, and business data. When it comes to energy storage devices, the last decades have witnessed the rapid spread of usage of carbon-based electrodes for electric double-layer capacitors (EDLCs) due to their low cost, high porosity, and large surface area.
It is crucial to develop an accurate and efficient forecasting model for electrochemical performance to reduce the time needed for making suitable designs and choosing testing electrode materials. Designing high-performance materials for supercapacitors by examining the interplay between the characteristics of a material and structural features is now seen as a major issue. For this reason, predicting capacitance is crucial for evaluating a material’s suitability for supercapacitor-electrode applications.
As a result, the use of machine learning (ML) approaches in creating a predicting model for the capacitance of carbon-based supercapacitors look critical and provide the electrode characteristics’ relative relevance. We seek establish a novel approach for the supported design of outstanding performance supercapacitor materials by utilising machine learning (ML). Which used to examine the interaction between porous carbon materials’ (PCMs) structural characteristics and capacitance. Used data is extracted from nearly a hundred published experimental research papers to select supercapacitors with certain electrode morphologies such as mesoporous, nanoporous, microporous, and hierarchical porous carbon electrode. The data was examined using machine learning techniques to predict the supercapacitor’s specific capacitance (F/g). Electrode material structural qualities and various physicochemical test features such as electrolyte material, pore volume, and specific surface area. Electrochemical test features acquired via galvanostatic charge-discharge (GCD) and electrochemical impedance spectroscopy (EIS) test investigations for the same purpose include: charge-transfer resistance (RCT), cell configuration, current density, applied potential window, and equivalent series resistance (ESR) were used as input features to predict the corresponding capacitance performance.
In the present study, Linear Regression (LR), Lasso, Support Vector Machine Regression (SVMR), Regression Tree (RT), Artificial Neural Networks (ANN) with different structures, and Adaptive Neuro-Fuzzy Inference System (ANFIS) were employed to calculate the capacitance of the supercapacitor. The efficiency of the ML models was assessed in terms of the root mean square error (RMSE), mean absolute error (MAE), and The correlation between anticipated yield and actual provided yield (R).
The developed ANFIS model, with RMSE, MAE, and R values of 22.8, 39.7647, and 0.90004, respectively, the constructed ANFIS model compares favorably to other models created for this purpose in terms of prediction performance. According to the analysis of the input features done using the SHAP (SHapley Additive exPlanations) framework, the specific surface area had the biggest impact on the ANN model.