Cascade sliding mode maximum power point tracking controller for photovoltaic systems

Abstract

Introduction. Constant increases in power consumption by both industrial and individual users may cause depletion of fossil fuels and environmental pollution, and hence there is a growing interest in clean and renewable energy resources. Photovoltaic power generation systems are playing an important role as a clean power electricity source in meeting future electricity demands. Problem. All photovoltaic systems have two problems; the first one being the very low electric-power generation efficiency, especially under low-irradiation states; the second resides in the interdependence of the amount of the electric power generated by solar arrays and the ever changing weather conditions. Load mismatch can occur under these weather varying conditions such that maximum power is not extracted and delivered to the load. This issue constitutes the so-called maximum power point tracking problem. Aim. Many methods have been developed to determine the maximum power point under all conditions. There are various methods, in most of them based on the well-known principle of perturb and observe. In this method, the operating point oscillates at a certain amplitude, no matter whether the maximum power point is reached or not. That is, this oscillation remains even in the steady state after reaching the maximum power point, which leads to power loss. This is an essential drawback of the previous method. In this paper, a cascade sliding mode maximum power point tracking control for a photovoltaic system is proposed to overcome above mentioned problems. Methodology. The photovoltaic system is mainly composed of a solar array, DC/DC boost converter, cascade sliding mode controller, and an output load. Two sliding mode control design strategies are joined to construct the proposed controller. The primary sliding mode algorithm is designed for maximum power point searching, i.e., to track the output reference voltage of the solar array. This voltage is used to manipulate the setpoint of the secondary sliding mode controller, which is used via the DC-DC boost converter to achieve maximum power output. Results. This novel approach provides a good transient response, a low tracking error and a very fast reaction against the solar radiation and photovoltaic cell temperature variations. The simulation results demonstrate the effectiveness of the proposed approach in the presence of environmental disturbances.

Electromechanical transient model verification of photovoltaic power generation unit

“…Another method is to combine photovoltaic array with inverter and other components, and simulate by controlled source or equivalent circuit, and establish a photovoltaic power generation system model. The simulation step is millisecond, which belongs to electromechanical transient simulation [6][7][8] .…”

ANFIS based method for faults detection in the photovoltaic system

“…The latter generates a direct electric current which will be converted into alternating current using an inverter. This electricity produced can be immediately used to operate appliances or light [2], [3]. Any solar installation, therefore, requires three elements to ensure the recovery of the rays transmitted by the sun, to then transform them into electricity and distribute them as shown in the Figure 1: i) PV panels, ii) an inverter to convert the electricity obtained into alternating current; This power is controlled by the maximum power point tracker (MPPT) which makes the system always operate at its maximum capacity, and iii) this type of system requires the use of batteries for electricity storage and a charge controller to ensure the durability of the batteries.…”

Maximizing solar photovoltaic system efficiency by multivariate linear regression based maximum power point tracking using machine learning

“…Trial-and-error-based methods like gradient descent calculate the adjacent local MPP using the gradient function. Intelligent prediction algorithms like fuzzy logic control (FLC) and artificial neural networks (ANN) can predict MPP by adjusting the weights of different layers through a training process [4,5]. Optimization methods like ant colony optimization, firefly algorithm, genetic algorithm, and grey wolf optimization attempt to optimize functions or variables to achieve maximum power extraction from the PV generator.…”