• Energy Research

IntroductionPhotovoltaic systems offer immense potential as a future energy source, yet maximizing their efficiency presents challenges, notably in achieving optimal voltage due to their nonlinear behavior. Operating current and voltage fluctuations, driven by temperature and radiation changes, significantly impact power output. Traditional Maximum Power Point Tracking (MPPT) methods struggle to adapt accurately to these dynamic environmental conditions. While Artificial Intelligence (AI) and optimization techniques show promise, their implementation complexity and longer attainment times for Maximum Power Point (MPP) hinder widespread adoption.MethodThis paper proposes a hybrid MPPT technique that integrates the Pelican Optimization algorithm (POA) with the Perturb and Observe algorithm (P&O) for a grid-connected photovoltaic system (PV). The proposed technique consists of two loops: PO as the reference point setting loop (inner loop) and POA as a fine-tuning (outer)loop. The combination of inner and outer loops minimizes oscillations by adjusting the perturbation direction and enhancing the convergence speed of the MPPT.Results and DiscussionTo validate the efficacy of the proposed MPPT technique for different environmental conditions, a comprehensive comparison is conducted between the proposed hybrid pelican and perturb and observe (HPPO) technique and other MPPT algorithms. The proposed technique has optimized PV and grid outputs with an MPPT efficiency of 99%, best tracking speed, and total harmonic distortion (THD) for all conditions below 5% agree with IEEE 519 standards.

IntroductionPhotovoltaic systems offer immense potential as a future energy source, yet maximizing their efficiency presents challenges, notably in achieving optimal voltage due to their nonlinear behavior. Operating current and voltage fluctuations, driven by temperature and radiation changes, significantly impact power output. Traditional Maximum Power Point Tracking (MPPT) methods struggle to adapt accurately to these dynamic environmental conditions. While Artificial Intelligence (AI) and optimization techniques show promise, their implementation complexity and longer attainment times for Maximum Power Point (MPP) hinder widespread adoption.MethodThis paper proposes a hybrid MPPT technique that integrates the Pelican Optimization algorithm (POA) with the Perturb and Observe algorithm (P&O) for a grid-connected photovoltaic system (PV). The proposed technique consists of two loops: PO as the reference point setting loop (inner loop) and POA as a fine-tuning (outer)loop. The combination of inner and outer loops minimizes oscillations by adjusting the perturbation direction and enhancing the convergence speed of the MPPT.Results and DiscussionTo validate the efficacy of the proposed MPPT technique for different environmental conditions, a comprehensive comparison is conducted between the proposed hybrid pelican and perturb and observe (HPPO) technique and other MPPT algorithms. The proposed technique has optimized PV and grid outputs with an MPPT efficiency of 99%, best tracking speed, and total harmonic distortion (THD) for all conditions below 5% agree with IEEE 519 standards.

When tracking the peak power point in PV systems, incremental conductance is the most common technique used. This approach preserves the first trap in the local peak power point, but it is unable to quickly keep up with the ever-changing peak power point under varying irradiance and temperature conditions. In this paper, the authors propose a hybrid algorithm, combining an artificial ecosystem optimizer and an incremental-conductance-based MPPT to solve these issues of traditional MPPT under varying irradiance and temperature conditions. The proposed hybrid algorithm has been applied to three scenarios, namely the constant irradiance condition, the varying irradiance condition, and the varying temperature condition. Under the constant irradiance condition, the PV array is maintained at a temperature of 25 °C and an irradiance of 1000 W/m2. The voltage of the DC link of the neutral-pointed-clamped inverter is maintained at 1000 V. Under the varying irradiance condition, the irradiance of the PV array is increased from 400 W/m2 to 1000 W/m2with a step size of 0.2 s. The same step size is maintained while decreasing the irradiance level from 1000 W/m2 to 400 W/m2, with a step change of 0.2 s. However, the temperature is maintained at 25 °C. Under the varying temperature condition, the temperature of the PV array varies from 35 °C, 25 °C, 15 °C, 10 °C, 15 °C, 25 °C, and 35 °C with a step size of 0.2 s, and the irradiance is maintained at 1000 W/m2. The DC link voltage in all three conditions is maintained at 1000 V, which confirms that the hybrid algorithm has been able to vary the duty cycle of the pulse wave modulation generator in such a manner that the variable DC voltage produced by the PV array has been changed by the flyback converter into a stable DC voltage. The simulation results show that the total harmonic distortion (THD) under all the simulated scenarios is within 5%, which agrees with IEEE standards. In the future, this algorithm may be compared with other types of available MPPTs under partial shading.

When tracking the peak power point in PV systems, incremental conductance is the most common technique used. This approach preserves the first trap in the local peak power point, but it is unable to quickly keep up with the ever-changing peak power point under varying irradiance and temperature conditions. In this paper, the authors propose a hybrid algorithm, combining an artificial ecosystem optimizer and an incremental-conductance-based MPPT to solve these issues of traditional MPPT under varying irradiance and temperature conditions. The proposed hybrid algorithm has been applied to three scenarios, namely the constant irradiance condition, the varying irradiance condition, and the varying temperature condition. Under the constant irradiance condition, the PV array is maintained at a temperature of 25 °C and an irradiance of 1000 W/m2. The voltage of the DC link of the neutral-pointed-clamped inverter is maintained at 1000 V. Under the varying irradiance condition, the irradiance of the PV array is increased from 400 W/m2 to 1000 W/m2with a step size of 0.2 s. The same step size is maintained while decreasing the irradiance level from 1000 W/m2 to 400 W/m2, with a step change of 0.2 s. However, the temperature is maintained at 25 °C. Under the varying temperature condition, the temperature of the PV array varies from 35 °C, 25 °C, 15 °C, 10 °C, 15 °C, 25 °C, and 35 °C with a step size of 0.2 s, and the irradiance is maintained at 1000 W/m2. The DC link voltage in all three conditions is maintained at 1000 V, which confirms that the hybrid algorithm has been able to vary the duty cycle of the pulse wave modulation generator in such a manner that the variable DC voltage produced by the PV array has been changed by the flyback converter into a stable DC voltage. The simulation results show that the total harmonic distortion (THD) under all the simulated scenarios is within 5%, which agrees with IEEE standards. In the future, this algorithm may be compared with other types of available MPPTs under partial shading.

Effective machine learning regression models are useful toolsets for managing and planning energy in PV grid-connected systems. Machine learning regression models, however, have been crucial in the analysis, forecasting, and prediction of numerous parameters that support the efficient management of the production and distribution of green energy. This article proposes multiple regression models for power prediction using the Sharda University PV dataset (2022 Edition). The proposed regression model is inspired by a unique data pre-processing technique for forecasting PV power generation. Performance metrics, namely mean absolute error (MAE), mean squared error (MSE), root mean squared error (RMSE), R2-score, and predicted vs. actual value plots, have been used to compare the performance of the different regression. Simulation results show that the multilayer perceptron regressor outperforms the other algorithms, with an RMSE of 17.870 and an R2 score of 0.9377. Feature importance analysis has been performed to determine the most significant features that influence PV power generation.

Effective machine learning regression models are useful toolsets for managing and planning energy in PV grid-connected systems. Machine learning regression models, however, have been crucial in the analysis, forecasting, and prediction of numerous parameters that support the efficient management of the production and distribution of green energy. This article proposes multiple regression models for power prediction using the Sharda University PV dataset (2022 Edition). The proposed regression model is inspired by a unique data pre-processing technique for forecasting PV power generation. Performance metrics, namely mean absolute error (MAE), mean squared error (MSE), root mean squared error (RMSE), R2-score, and predicted vs. actual value plots, have been used to compare the performance of the different regression. Simulation results show that the multilayer perceptron regressor outperforms the other algorithms, with an RMSE of 17.870 and an R2 score of 0.9377. Feature importance analysis has been performed to determine the most significant features that influence PV power generation.