International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 12 Issue: 04 | Apr 2025
p-ISSN: 2395-0072
www.irjet.net
Performance Analysis of Hybrid MPPT Controller for PV Boost Converter Priya Manoria Electrical engineering Department Government Polytechnic College Katni M. P. ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - This study examines the Perturb and Observe
(P&O) and Incremental and Conductance (I&C) MPPTs, both of which use neural network (NN)-based artificial intelligence techniques to improve their output. The purpose of the DC-DC converter is to control the PV module's output voltage in buck or boost mode as necessary. The DC-converter is designed using a single switch unidirectional architecture, with gate pulses regulated by an MPPT algorithm based on neural networks. The P&O and I&C algorithms with and without the NN approach are compared. To validate the suggested topology, the results are acquired using MATLAB program for Simulink.
voltage and current at MPP. The DC load is powered by the DC-DC converter, specifically the boost. In order to precisely convalesce the MPP under a variety of operating situations, the output from MPPT is used to create the converter's proper duty cycle. When the hybrid MPPT algorithm is used, the duty cycles provide PWM for the converter's switches and achieve smooth fluctuations with regard to varied irradiations. A compact DC system with 250 W of power and input-to-output voltage fluctuations between 30 and 80 V is built. In order to determine which strategies converge quickly to determine the maximum voltage and current at the moment of irradiation variance, results are compared under varying irradiance and performance is examined.
Key Words: Maximum Power Point Tracking (MPPT), incremental and conductance (IC), constant voltage, perturb and observe (PO), neural networks (NN), DC-DC converter.
2. NEURAL NETWORK BASED HYBRID MPPT A highly effective technique for improving PV system performance and guaranteeing smooth operation in varying weather conditions is MPPT. In order to build the duty-cycle of the DC-DC converter to follow the MPP, MPPT is used to track the MPP [9].
1.INTRODUCTION During the past two decades, solar has astonishingly gotten deep into the main stream power system. Solar irradiation, which fluctuates on an hourly, daily, monthly, and annual basis, is the source of solar energy. Therefore, it is not possible to generate power continuously from sunlight. However, by monitoring the highest generation at the instantaneous irradiance, the output from the sola-cell may be maximized at any given time. Maximum Power Point Tracking (MPPT) is used to accomplish this [1]. When PV production peaks at a location known as the Maximum Power Point (MPP), which is constantly shifting in relation to temperature and solar radiation, MPPT aids in tracking the output. MPP is a point on the PV/VI curve that represents a particular PV module's peak voltage, current, and power. There are many different topologies for building MPPT controllers in the literature; the most widely used ones include hill climbing, incremental and conductance (IC), constant voltage, perturb and observe (PO), and others [2– 5]. All of these topologies are widely used and customary. By using any clever techniques, the efficiency of the traditional topologies may be multiplied by many. Intelligent methods, such as neural or fuzzy, aid in the quick and precise monitoring of MPP [6–8].
Using the NN-algorithm to adjust the fluctuation in MPP voltage and current may significantly improve the MPPT's convergence speed and quick adaptation. In order to supply the gating of the switches of the PV-boost converter, this study develops a hybrid NN-based IC and PO method [10]. ANN is used because of its high degree of convergence flexibility and dependability. The NN determines the voltage and current at MPP for each operating point with a specified temperature and irradiance. The PO algorithm determines the proper duty cycle for the DC-converter's switches based on the estimated parameters. In a variety of climatic conditions, this will perfectly recover the maximum power and be in line with the MPP. The conventional PO and IC approaches are then compared with the proposed approach. A proportional-integral (PI) controller is used to ensure that the capacitor voltage balance is maintained. Figure 1 displays the full schematic design of the suggested architecture. NN is often used for MPPT control since it doesn't require a physical model or intricate mathematical computations. It can also manage the large nonlinearities in the P/V characteristics of the PV panel [11-13].
The intelligent MPPT hybrid approach based on neural networks (NN) is presented in this paper. With the aid of an algorithm that facilitates quick convergence of the MPP at the specified irradiance, NN is intended to calculate the
© 2025, IRJET
|
Impact Factor value: 8.315
|
ISO 9001:2008 Certified Journal
|
Page 952