Support Vector Regression Machine Learning based Maximum Power Point Tracking for Solar Photovoltaic systems

Abstract

Photovoltaic panels use the sun’s radiation on their surface to convert solar energy into electricity. This process is dependent on the temperature of the surface and the intensity of the sun's radiation. To escalate the energy transformation, the solar system must be functioned at its maximum power point (MPP). Every maximum power point tracking (MPPT) technique has a distinct mechanism for tracking maximum power point (MPP). The support vector machine (SVM) regression algorithm is used in this work to develop a novel method for tracking the MPP of a PV panel. The solar panel technical parameters were used to prepare the data for training and testing the SVM model. The SVM algorithm predicts the PV panel's maximum power and relevant voltage for specific irradiation and temperature. The duty cycle of the boost converter corresponding to the maximum power was evaluated using the predicted values. The result of the simulation shows that the proposed control strategy forces the solar panel to work near the predicted MPP. The SVM regression control strategy gives the MPP tracking efficiency of more than 94% for the solar PV system despite variable climatic conditions during its stable state operation. In addition, a comparative analysis of the proposed method was carried out with the existing approaches to confirm the effective tracking of the proposed technique.

Polynomial surface-fitting evaluation of new energy maximum power generation capacity based on random forest association analysis and support vector regression

“…The basic principle of SVR is to calculate the decision function in the feature space for a given training sample set, followed by a regression prediction utilizing the decision function (Venkata et al, 2023). For linearly indistinguishable problems, mapping of the sample data to a high-dimensional feature space by utilizing a kernel function and determination of the linear decision function in the highdimensional feature space outperform other ML algorithms when the number of samples is small.…”

“…Le et al (2023) implemented an overview of modern MPPT algorithms applied to permanent magnet synchronous generators in wind power conversion systems with MPPT methods based on speed convergence, efficiency, self-training, and complexity. Venkata et al (2023) directly adopted solar panel technical parameters and proposed a new method of MPPT for photovoltaic (PV) panels based on an SVR machine. Unlike Venkata et al (2023), which focused on analyzing the unique and complex operating mechanisms of PV panels, this study commences from the perspective of data, which can considerably avoid complex mechanism analysis.…”

“…Venkata et al (2023) directly adopted solar panel technical parameters and proposed a new method of MPPT for photovoltaic (PV) panels based on an SVR machine. Unlike Venkata et al (2023), which focused on analyzing the unique and complex operating mechanisms of PV panels, this study commences from the perspective of data, which can considerably avoid complex mechanism analysis. This model focuses on the maximization method, and studies on upperenvelope analysis are limited.…”

Development of AI-Based Tools for Power Generation Prediction

“…Mahesh et al [33] propose a novel method for evaluating the efficiency of PV panel systems. It combines the Maximum Power Point Tracking (MPPT) technique and SVM Machine Learning to predict the maximal value of the power generated by a PV solar panel.…”

Perfect Tracking Control of Linear Sliders Using Sliding Mode Control with Uncertainty Estimation Mechanism

“…Kernel LMS is a method that can adaptively identify nonlinear functions by approximating them using the kernel method, and it is widely used in the field of signal processing [38][39][40][41]. There are other methods for identifying nonlinear functions, such as Neural Networks (NNs) [42][43][44] and Support Vector Machines (SVMs) [45][46][47], but they have high learning costs and cannot easily achieve adaptive identification. On the other hand, Kernel LMS has the advantage of low learning costs and the ability to achieve adaptive identification.…”