Debnath, Dipanwita2026-10-132026-10-132022-11-04http://nits.ndl.gov.in/handle/123456789/229In the modern decade, the intensive developments of technology and changing lifestyle of the current generation of people have made energy an essential and inseparable commodity, but at the same time growing shortage of finite natural resources and degradation of the environmental condition has focused attention on adding the renewable development principle with future technologies. Among all the possible green renewable resources, energy derived from solar power getting continuous significant importance due to countless clear reasons. It is fully clean and inexhaustible in nature and holds out the promise of providing the desired amount of electrical power in need. However solar energy has some drawbacks like lower operational efficiency and high investment cost which present room to expand the research problems. For a reliable solar unit, solar cell accurate modeling is necessary before actual implementation to improve the system performance. The modeling of PV cells involves the extraction of optimized parameters of a selected PV electrical diode model. Since the current vs. voltage (I-V) relationship of PV cell is nonlinear, accordingly model parameter estimation process is viewed as a complicated nonlinear optimization problem, and to denominate this problem numerous dynamic research is going on. So, this dissertation starts with the modeling of different photovoltaic solar cell models based on single diode model (SDM), double diode model (DDM), and a PV module to extract the most optimized parameters of selected models. Further, to maximize the energy of solar PV units, a maximum power point tracker with a potential tracking algorithm is needed for trapping the highest power continuously from a varying source. Under the situation of varying environmental conditions, the output power of the PV module fluctuates. The classical methods (like P&O and incremental conductance method) are unable to give an optimal response in that situation. Many MPPT techniques have been carried out by different researchers in the past to increase the operational efficiency of a PV system, but a lot of research probability still lies in developing an efficient MPPT algorithm. Adding more, to increase solar-based PV yields energy it is important to remember that the temperature of PV cells should not exit its expected module rated value. With increasing surface operating temperature, the operational efficiency of PV units decreases and for that reason to keep the surface temperature close to the surrounding ambient temperature a suitable smart cooling system is necessary. The parameter estimation represents a procedure where measured current-voltage data are used to calculate the parameters of a solar PV cell. Here a very modern and potential optimization technique, named sine-cosine algorithm (SCA) is utilized to accurately estimate the parameters of PV cells/module using experimentally measured I-V data for single and double diode models along with KC200GT PV module. Being SCA technique is a global optimization technique, it does not suffer from singularity problem and provide considerably the most optimized PV panel parameters with better system depiction. The comparison response of SCA outperforms the several published works in terms of root mean square of error (RMSE) and other statistical analysis means. After having success in estimating accurate parameters of solar cells, this thesis subsequently proceeds for developing a novel hybrid global MPPT algorithm to gain maximum power from PV module under the situation of partial shading conditions. Here, improved grey wolf optimization (IGWO) technique, a fusion of parallel grey wolf optimization (PGWO), and opposition-based learning is employed to design the MPPT control structure; for this study, the PV system along with the suggested MPPT algorithm with its components and subsystems is developed through simulation and experimental studies. Firstly, the potentiality of the recommended algorithm to handle the situations: (i) partial shading conditions (ii) gradual solar irradiance and level of temperature changes (iii) rapid irradiance and level of temperature changes have been tested through MATLAB/SIMULINK and subsequently, a prototype setup using ESP32 microcontroller is developed for its practical implementation under partial shading situation. The performance of IGWO has been benchmarked with two classical MPPT methods PSO and P&O. The simulated result of IGWO based MPPT shown here is more effective concerning tracking efficiency as compared to widely used PSO and P&O methods for gradual and rapid change in solar irradiance and temperature levels. The experimental outcomes also show an acceptable performance for the proposed technique. The proposed controller also offers stable sets of responses means there exists no oscillatory behavior around optimal value. The IGWO controller has also succeeded in tracking global maxima efficiently under partial shading. The third work recited in this thesis involves an experimental work equipped with Internet of Things (IoT) platform for enhancing the PV panel performance using the front panel water cooling method for the climate of Bhubaneswar, India. To achieve higher operational efficiency from PV panels, various cooling techniques have been proposed and implemented, but all those methods need continuous observation. A solar panel with no cooling arrangement is used as a base model and compares the test results for both with and without cooling conditions to get the consolidated outcomes. The main objective of using IoT is to give the cooling system a smart touch which ensured the underlying approach successfully records the real-time data for PV system. The result section of this part shows acceptable performance after considering the cooling circuit tooled in IoT system as compared to the non-cooled one.en