Improving Photovoltaic System Performance with Artificial Neural Network Control

This paper proposes a new technique for energy managing in hybrid power systems. The proposed management system is designed to manage the power flow between the hybrid power system in order to satisfy the load requirements based on artificial neural network (ANN) and fuzzy logic controllers. The neural network controller is employed to achieve the maximum power point tracking (MPPT) for different types of sources like Photovoltaic(PV), wind energy and Fuel cell at DC-link.The developed fuzzy logic controller (FLC)-based MPPT method and ANN based method was assessed using a hybrid system comprised PV panels, wind turbine (WT) and Fuel Cell with DC–DC converters. The dynamic behaviour of the proposed model is examined under different operating conditions. The analysis of the proposed hybrid system provides higher power as compared to PV, WT and FC systems at different loads. This research is utilized for the stand-alone system as well as a grid. ANN gives a better performs compared to the Fuzzy in the MPPT method for the PV panels, wind turbine (WT) and Fuel Cell with DC–DC converters for the DC loads. The result analysis of the Fuzzy and the ANN are simulated in the MATLAB/Simulink.

The use of solar panels has some advantages over other conventional electrical generating methods, such as there is no sound pollution in collecting solar energy via solar panels, and it has a minimum need for maintenance. In addition, it helps in the greenhouse effect which does not contribute to any CO <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf> pollution. Using photovoltaic (PV) systems that are connected to a load will require a Maximum Power Point Tracker (MPPT) to maintain the highest possible efficiency of power generated. Due to the difference in PV and load resistance, the MPPT will control the duty cycle of the Insulated Gate Bipolar Transistor (IGBT) in the DC-DC converter to match the PV and load resistance. However, the use of MPPT will require a controller to seek the maximum power generated in the PV system. In this paper, we design and implement the Artificial Neural Network (ANN) based MPPT method to improve the efficiency of the power observation for the PV system. Mainly, we compare three sets of data to train the ANN controller, 104 samples, 10001 samples, and 2000001 samples, as ANN uses prediction in its calculations to find the best output efficiency.

This paper details an maximum power point tracking (MPPT) approach based on artificial neural network (ANN) to track the maximum power produced by a PV panel. This approach is rapid and accurate for following the maximum power point (MPP) during changes in weather conditions such as solar irradiation and temperature. A PV system structure including an MPPT controller is studied, designed, and simulated in this work. The aim of this paper is to use the artificial neural network (ANN) technique to develop a MPPT controller for PV applications. To increase the performance of the ANN-MPPT controller, a proportional integral (PI) controller is also included. In addition, the performance of an ANN-based MPPT controller is also compared to the conventional perturb and observe (P&amp;amp;O) method. To analyze the results, simulations are performed by using MATLAB software.

Inverter's controller directly determines the grid-integration quality of power conversion systems. To address complex operating conditions, e.g. grid voltage distortion, parameter variations and weak grid scenarios, this paper proposes a lightweight artificial neural network (ANN) controller with strong adaptability. In implementation, the ANN controller generates control signals while simultaneously optimizing its weights in real time using the gradient descent algorithm. Distinctively, the weight gradients are directly calculated using the loss function and weights, which compared to the error backpropagation method, significantly reduces computational complexity, and therefore achieves the computational lightweight feature. In addition, the ANN phase-locked loop (ANN-PLL) is constructed to provide phase alignment for current reference while enabling fully ANNbased inverter control architecture. In principle, the proposed ANN controller relies neither on an offline training dataset nor on the system model, and achieves adaptive weights adjustment in real time with minimal computational effort. Through physical experiments, the proposed lightweight ANN controller was compared with the sliding mode controller and the model-based ANN controller, verifying its superior performance under complex operating conditions, such as grid voltage distortion, input voltage variation, current reference variation, filter parameter variation, and extremely weak grid (short circuit ratio=1.09).

In this paper, Standalone PV system is designed to operate without the need of electric utility grid which are commonly designed to supply specific DC electric loads. These type of PV system are generally powered using photovoltaic array and consist of solar charging modules, controllers and regulators. The objective of this work is to implement the Radial Basis Function Neural Network (RBFNN) based Maximum Power Point Tracking (MPPT) technique to attain constant DC link voltage of the ZETA converter. This method utilizes a ZETA converter to achieve better voltage gain and lower ripple in the output current and voltage. Using an Artificial Neural Network (ANN) controller to operate a high frequency converter, which improves power conversion efficiency. To ensure safety and prevent electrical faults, an isolation transformer provides galvanic isolation between the high frequency converter and the PV system. Galvanic isolation avoids electrical faults from being transmitted and safeguards connected loads. The rectifier is used to supply the Direct Current (DC) load with the output from the isolation transformer. The rectifier changes the transformer’s Alternating Current (AC) output into DC current to power DC loads. An intelligent MPPT is employed to stabilize the converter’s output voltage and address the intermittent characteristics of PV, and also assists in sustaining the DC link voltage without alterations and ripples from the converter. The efficiency of this work is authenticated via MATLAB simulation 2021a.

This paper includes an application of layered artificial neural network controller to study the load-frequency control problem in a power system. The control scheme guarantees that steady state error of frequencies and inadvertent interchange of tie-lines are maintained in a given tolerance limitation. The proposed control has been designed for a three-area interconnected power system such that two areas include steam turbines and the other area includes a hydro turbine. Only one artificial neural network (ANN) controller, which controls the inputs of each area in the power system together, is considered. In the study, a back propagation-through-time algorithm is used as a neural network learning rule. The performance of the power system is simulated by using a conventional integral controller and ANN controller, separately. By comparing the results for both cases, the performance of an ANN controller is better than conventional controllers.

This paper proposes a capacitor switching logic for a multi input Single Ended Primary Inductance Converter (SEPIC) with efficient Maximum Power Point Tracking (MPPT) scheme for multiple non aligned solar panels. The proposed scheme is an improvement of time division multiplexing control of multi input converters for low power solar energy harvesters. In the proposed design, the capacitor switching logic gives the flexibility of selecting panels on demand providing additional advantage of isolation. For better accuracy an efficient algorithm has been implemented for determining MPPT. The MPPT is practically determined in each cycle when the corresponding capacitor is connected to converter. Using SEPIC converter at the output provides a constant 15V for charging the battery for solar panel voltage range from 8V to 25V. The design improvements will bring increased efficiency for medium power solar systems.

A hybrid maximum power point tracking for partially shaded photovoltaic systems in the tropics

This paper aims at improving the excitation control system for the doubly fed induction generator-based wind farm (WF). The proposed decentralized coordinated neural control is consisted of two controllers, i.e., a multiple model predictive controller (MPC) and an artificial neural network (ANN) controller. The MPC is based on the interaction measurement model and designed to guarantee the performance of system-wide optimization around the chosen operating points, while the ANN controller is used to calculate the weights of MPC controllers and trained to guarantee the control performance over the full operating region. The proposed control strategy allows to reduce the computational cost, and at the same time to take advantage of the distributed nature of the power system. Moreover, a simple, generic hybrid power system is used to demonstrate transient stability contributions. The problems of large-scale WF integration and its impact on transient stabilities of power systems are discussed through the eigenvalue analysis and time domain simulation. The results show that system damping and transient stability of hybrid power system are considerably improved.

In this paper, an artificial neural network (ANN) controller based position sensorless control of permanent magnet synchronous motor (PMSM) using Space vector modulation-Direct torque control (SVM-DTC) for variable speed drive has been presented. The SVM-DTC require the initial position of the rotor during the starting of the PMSM drive. The installation of the shaft-mounted position sensor requires additional space, assembly, wiring circuit, and is fragile component. The speed sensor-less control of PMSM enhances the performance of drives in harsh environments and reduces the overall cost of the drive and improve mechanical reliability. The speed estimation requires the knowledge of drive parameters, the model-based speed control technique is suitable for low and medium-speed motor drive applications without knowing the exact parameter of the PMSM drive. The Rotor Flux based Model Reference adaptive system (RF-MRAS) is used for a wide speed operation and estimates rotor angle in dynamic conditions. The presence of an integrator in the voltage model of RF-MRAS affects the low speed performance of the drive, therefore to improve the speed response at low speed, the ANN controller is used to replace the Proportional-Integral (PI) controller, which is employed in the adaptive model of the speed observer. The performance of the control scheme is simulated at variable speed and load conditions with the help of the OPAL-RT 4500 simulation platform.

This paper presents the five-level direct power control (DPC) using artificial neural networks (ANNs) controller of a doubly fed induction generator (DFIG) based wind power generation systems (WPGSs). The validity of the proposed strategies is verified by simulation tests of an induction generator. The harmonic distortion of stator current, reactive power, and stator active power are determined and compared to the above strategies. The five-level DPC strategy with ANN controller is shown to be able to minimize the reactive power, harmonic distortion of stator current, and stator active power ripples and to improve performance DPC control scheme. Â

The escalating demands and increasing awareness for the environment, resulted in deployment of Photovoltaic (PV) system as a viable option. PV system are widely installed for numerous applications. However, the challenges in tracking the maximum power with intermittent atmospheric condition and varying load is significant. Maximum Power Point Tracking (MPPT) algorithms are employed and based on their convergence speed, control of external variations and oscillation, the output power efficiency, and other significant factors viz. the algorithm complexity and implementation cost, novel MPPT approach are preferable than the conventional approach. This paper presents an artificial intelligence-based optimization controller for MPPT in a PV system under varying load and irradiance conditions. Comparative analysis of Artificial Neural Network (ANN) and Particle Swarm Optimization (PSO) based MPPT is simulated and analysed. The PV system consisting of PV array and boost converter with MPPT controller feeds the DC load. The power conversion and panel efficiency are the significant factors to determine the effectiveness of tracking maximum power point. The simulation results show the performance of these controllers on the PV panel output power and the load side output power under changing loads and irradiance. In addition, the comparison of PV panel efficiency of ANN and PSO based MPPT techniques w.r.t changing loads is carried out. Based on the above analysis, PSO based MPPT algorithm marginally outperforms the ANN based MPPT algorithm. Further, the implementation of hybrid MPPT (ANN &amp;PSO) for higher accuracy and tracking capability can be carried out as future work.

Harsh outdoor operating circumstances of photovoltaic (PV) array leads to non-uniform aging phenomenon among PV panels which in turn reduces the life span of the PV modules and the energy efficiency of the whole array. Under non-uniform aging conditions for series-parallel (SP) connected PV modules, the characteristics of the entire PV array are deformed resulting in multiple peaks where one of them is the global peak. The appearance of such multiple peaks can mislead traditional maximum power point tracking (MPPT) techniques to get trapped at local peaks, which leads to misleading power losses. This paper proposes a current collector optimizer (CCO) topology to extract maximum power from a non-uniform aged PV array. By using CCO topology, the PV array characteristics have a unique maximum power point (MPP) which easy to follow by a simple MPPT algorithm. Consequently, the proposed topology does not suffer from misleading power losses. Besides, the CCO topology requires offline rearrangement of aged modules to obtain the optimal maximum power. A comparative study is conducted to demonstrate the efficacy of the proposed topology. Ultimately, the paper presents an offline algorithm for rearranging erratic aged CCO modules to extract the optimum power from large-scale PV arrays.

When subjected to partial shading (PS), photovoltaic (PV) arrays suffer from the significantly reduced output. Although the incorporation of bypass diodes at the output alleviates the effect of PS, such modification results in multiple peaks of output power. Conventional algorithms—such as perturb and observe (P&O) and hill-climbing (HC)—are not suitable to be employed to track the optimal peak due to their convergence to local maxima. To address this issue, various artificial intelligence (AI) based algorithms—such as an artificial neural network (ANN) and fuzzy logic control (FLC)—have been employed to track the maximum power point (MPP). Although these algorithms provide satisfactory results under PS conditions, a very large amount of data is required for their training process, thereby imposing an excessive burden on processor memory. Consequently, this paper proposes a novel optimization algorithm based on stochastic search (random exploration of search space), known as the adaptive jaya (Ajaya) algorithm in which two adaptive coefficients are incorporated for maximum power point tracking (MPPT) with a rapid convergence rate, fewer power fluctuations and high stability. The algorithm successfully eliminates the issues associated with existing conventional and AI-based algorithms. Moreover, the proposed algorithm outperforms other state-of-the-art stochastic search-based techniques in terms of fewer fluctuations, robustness, simplicity, and faster convergence to the optima. Extensive analysis of results obtained from MATLAB® is done to prove the above performance parameters under static insolation conditions (using a three, four and a five-module series-connected PV system), under dynamically varying insolation (using a four-module series connected system), by changing the PV module rating (using a four-module series connected system) and using an IEC standard.

Solar photovoltaic (PV) system contributes a significant role in meeting power requirements and helps the decentralization of power generation and distribution. The output of the individual photovoltaic module depends on local temperature and irradiance. Since the variation of these parameters depends on the environmental factors, it is difficult to obtain maximum power in all instances. The maximum power point tracking (MPPT) technique is used for tracking the maximum power available at a particular instant. This paper presents the maximum power extraction from a photovoltaic system using Artificial Intelligence (AI) techniques. AI based methods provide a fast dynamic response with high accuracy and mitigate the effects of partial shading. A comparative study is performed for various AI based MPPT techniques such as Fuzzy, Artificial Neural Network (ANN), and Adaptive Neuro Fuzzy Inference System (ANFIS). PV module output is interfaced with the resistive load through the Z-Source inverter (ZSI). The ZSI boost the input voltage and provides an AC output voltage. The AI based MPPT techniques have been investigated with conventional Z-source Inverter (ZSI) and Switched Inductor Z-Source inverter (SL-ZSI). A simple boost control technique is employed to control, both conventional and SL Z-source inverter. The performance of the system is verified using MATLAB Simulink.