A novel fault diagnosis technique for photovoltaic systems based on artificial neural networks

This paper presents a diagnostic method based on I-V characteristic analysis and Artificial Neural Networks (ANN). A number of attributes are estimated using a simulation model based on a set of working conditions such as solar irradiance and module's temperature. The estimated attributes are then compared with those obtained from the real PV array measurements, which lead to the detection of possible faulty operating conditions. After detecting the presence of possible fault, tow algorithms are developed in order to isolate and identify eight different types of faults. The simulation results show that the proposed method can detect and classify the different faults occurring in a PV array with a high accuracy.

Fault detection method for grid-connected photovoltaic plants

Modular multilevel converters (MMCs) will be extensively used in the high-voltage direct-current transmission networks because of its superior characteristics over line commutated converter. Increasing the reliability of the MMC is directly related to the balancing of the MMC submodule capacitors voltages, which guarantees the proper operation of the converter and lowers the stress on the submodules. This paper presents an adaptive voltage-balancing strategy based on the capacitor voltage estimation, utilizing a hybrid adaptive linear neuron recursive least squares scheme. The proposed strategy eliminates the need of measuring submodules capacitor voltages and associated communication link with the central controller. Furthermore, the estimated capacitor voltages are utilized to detect and localize different types of submodule faults. After isolating the faulty submodules, the proposed fault-tolerant control unit modifies the parameters of the voltage-balancing strategy to overcome the reduction of the active submodules. The dynamic performance of the proposed strategy is investigated, using PSCAD/EMTDC simulations and hardware-in-the-loop real-time simulations, under different normal and faulty operating conditions. The accuracy and the time response of the proposed fault detection and tolerant control units result in stabilizing the operation of the MMC under different types of faults. Consequently, the proposed integrated control strategy improves the reliability of the MMC.

A new switching control strategy is proposed for nonlinear batch processes with different types of actuator faults. First, the nonlinear batch process is transformed into different equivalent hybrid discrete two-dimensional T-S fuzzy models according to different types of faults, and an iterative learning control law is designed. Then, based on the Lyapunov stability theory, sufficient conditions under different types of fault conditions are given to ensure the system stable, and the corresponding switching signals are designed at the same time. Finally, taking a three-capacity water tank as an example, the experiment verifies the effectiveness of the algorithm.

Most of the micro-grids require solar photovoltaic (PV) panels to convert sunlight to electrical energy. The electrical energy converted by the PV panel is DC while most of the applications require AC; an inverter is used for conversion. The PV panels are usually placed on rooftops or at any unattended condition, so the PV panels will get easily damaged by the environmental impacts. The same scenario appears in inverter as well, while different types of faults will occur while connecting various types of loads to the system. This paper presents the different types of faults that occur in an entire solar photovoltaic integrated micro-grid. The novelty of the work undertaken is to provide an extensive coverage of all the faults that occur in a solar PV-assisted micro-grid. In addition to it, the critique also intend towards fault-tolerant and diagnostic techniques, which were not dealt elsewhere for the complete solar PV-assisted micro-grid system. The necessity of this survey is to provide an overview of all the faults that exist in a solar PV-assisted micro-grid and the strategies that require to operate the system even under fault condition to meet the customer need for providing continuous power without any interruption. The various faults considered for the study are at PV modules ,inverter, batteries, and charge controllers. The review first analyzes the different types of faults and then the diagnosis methods are presented. The novel solution to deal with various faults is also recommended and it was compared with the existing solutions to show the effectiveness of the proposed approach.

This paper intends to present an approach to classify different types of faults and to identify the location of the faults in a non-radial power system network using ElectroMagnetic Transients Program(ATP/EMTP)software and Artificial Neural Network(ANN). Firstly, a balanced three-phase system is designed with a RLC load and then different types of faults(single line to ground fault, line to line fault, double line to ground fault and three phase fault) are created at various points of the transmission line. The transmission line model is simulated using the EMTP software. The resulting current waveforms under different fault conditions are observed from the sending end. Fault occurs at the point where peaks appear in the current signal when viewed from the sending end. These current waveforms are analyzed using the wavelet toolbox in MATLAB and the entropy values of the current signals, so obtained, are given as input to the artificial neural network for automatic fault classification and identification of the faulty line. This scheme is also tested under different types of faults with different fault resistances and varying fault locations and the results show that it is able to discriminate the faults and identify the fault locations rapidly and correctly.

Energy sources are very important for the development of a country. Due to global warming and other environmental effects there is an urgent need for clean energy. The study and research on solar photovoltaic systems is increasing to get the electricity on the electric power grid as well as on local domestic load level. This technology development nowadays focuses on the improvement related to the enhancement in the performance of these solar PV modules with factors dependent on the conditions at the installation sites. The present work is based on the experimentation that is conducted on a laboratory set made as per the hot and dry climate zone of India. Its experimental set up consists of two solar cell array PV modules with similar electrical and mechanical parameters under the experimental sets. Work is focused on the analysis of the real time performance records and measurements by high quality standard instruments at the time zone of different months in the year without and with the presence of effect of cooling due to artificial wind. The experimental observation describes that due to increase in the module temperature because of heating by solar irradiance degrades the performance of the solar PV module in terms of net energy output but by the inclusion of the controlled artificial wind based cooling mechanism helps in supporting the process of bringing down the solar cell PV module temperature as a result the gain of a net energy is increased for similar time constraints and irradiance. Performance measure ratio is also consequently observed to be improved. Finally the experimental and simulated energy by the artificial neural network (ANN) is observed for both of the wind cooled module and without the wind cooled module experimental and simulated energy. ANN based simulated model related estimated values of energy is observed to be closer to experimental values for both modules. ANN is helpful in finding the accurate estimation of the performance ration of solar as that of experimental results.

A methodology for evaluating and visualising the response of power system distance protection units most commonly used is presented. A general case, with different topologies, of a double-circuit transmission line is considered and different types of inter-circuit faults are treated. Single line and simple faults are simplified cases of the general one. Faults are treated by means of the adequate connection between the sequence networks, where different types of faults are represented by a resistance star associated to both lines and ground. Different combinations of these resistances result in different types of faults. Fault conditions are introduced to sequence networks of the rest of the system through phase current controlled current sources, and the resulting sequence voltages at the fault point are applied to fault resistances through sequence voltage controlled voltage sources. To calculate faults the mesh analysis method has been modified to take into account the dependencies fixed by the controlled sources. Real previous state of the power system (load flow, sequence source impedances at both ends and sequence impedance between ends) is taken into account. A useful and reliable way of representation is presented to obtain the expected response of different types of distance units in multi-circuit lines in presence of inter-circuit simultaneous faults.

An implementation of an intelligent photovoltaic module on reconfigurable Field Programmable Gate Array (FPGA) is described in this paper. An experimental database of meteorological data (irradiation and temperature) and output electrical generation data of a Photovoltaic (PV) module (current and voltage) under variable climate condition is used in this study. Initially, an Artificial Neural Network (ANN) is developed under Matlab/Similuk, environment for modeling the PV module. The inputs of the ANN–PV module are the global solar irradiation and temperature while the outputs are the current and voltage generated from the PV‐module. Subsequently, the optimal configuration of the ANN model (ANN–PV module) is written and simulated under the Very High Description Language (VHDL) and ModelSim. The synthesized architecture by ModelSim is then implemented on an FPGA device. The designed MLP‐photovoltaic module permits the evaluation of performance of the PV module using only environmental parameters and involves less computational effort. The device can also be used for predicting the output electrical energy from the PV module and for a real time simulation in specific climatic conditions. Copyright © 2010 John Wiley & Sons, Ltd.

The concept of PV panel characteristics under partially shaded conditions, individual PV panel Maximum Power Point Tracking (MPPT) for better utilization of power output of each panel and artificial neural networks (ANN) controller for tracking actual MPP under partially shaded conditions are presented in this paper. By this setup we extract maximum obtainable solar power from a PV module and use the energy for a DC application. Proposed work presents a two-stage maximum power point tracking (MPPT) controller for a photovoltaic (PV) source using Artificial Neural Network (ANN), under varying weather conditions of solar intensity, solar irradiation and module temperature. At the first-stage, the Artificial Neural Network controller is trained with the data of voltage, irradiance, intensity and temperature of a solar panel which gives the range of duty ratio as an input to the MPPT controller. In the second stage, a simple MPPT controller searches for MPP in the range given by the ANN controller thus quickens the response of the system and by changing the duty cycle of a DC-DC boost converter accordingly tracks the MPP. The MPP of each individual pv panel are tracked and cascaded at the end through a dc link. In this method, experimental data collection of a solar panel is used for development of the ANN architecture which is developed in Weka-3.9 and MATLAB. The whole system is simulated in MATLAB Simulink.

Photovoltaic (PV) module temperature is an important parameter in PV system operation as the system output power decreases as the module temperature increases. Therefore, the modeling of operating PV module temperature is crucial to understand the climatic factors which contribute to the variation of the PV module temperature. This paper presents the modeling of operating PV module temperature from a Grid-Connected Photovoltaic (GCPV) system located at Green Energy Research Centre (GERC), Universiti Teknologi MARA, Malaysia. An Artificial Neural Network (ANN) was developed to model the operating PV module temperature with solar irradiance and ambient temperature set as the ANN inputs. In addition, Cuckoo Search (CS) was introduced to search for the optimal number of neurons of ANN hidden layer, learning rate and momentum rate such that the Mean Absolute Percentage Error (MAPE) of the modeling process could be minimized. The results showed that CS had outperformed an Artificial Bee Colony (ABC) algorithm for the ANN training optimization by producing lower MAPE.KeywordsOperating photovoltaic module temperaturemodelingArtificial Neural NetworkCuckoo SearchMean Absolute Percentage Error

Fault diagnosis in photovoltaic (PV) installations is a fundamental task to protect the components of PV systems (modules, strings and inverters), from damage and to eliminate possible fire risks. In this paper, an intelligent fault detection and diagnosis method has been presented and implemented into a reconfigurable Field Programmed Gate Array (FPGA). Only faults that can be appeared in PV arrays are examined in this work. The designed method consists of two parts: the first one is based on signal threshold approach, and the second one is based on an artificial neural network (ANN). The whole parts of the system have been implemented into FPGA board (named ZYNQ XC7Z010-1CLG400C). Xilinx System Generator (XSG) and VIVADO tools have been used to simulate and implement the algorithm. Results demonstrate with success the possibility implementation of the designed diagnosis method.

The climate changes affect photovoltaic (PV) module temperature significantly. The module temperature is one of the most important factors that influence the PV module efficiency and a deep analysis of PV module temperature will aid in better understanding of the environmental influences on the PV module performance. The module temperature depends on many parameters such as solar radiation, ambient temperature, air humidity, speed and direction of the wind, PV module orientation, dust and sand deposition on PV module, and PV module materials. An experimental research was conducted to investigate the effect of these factors on the PV module temperature in the Renewable Energy Laboratory of the Graduate University of Advanced Technology in Iran. The results of this study highlighted that the deposited dust over the PV module surface increases the module temperature and this consequently decreases the PV module power. It was also revealed that a combination of the temperature increase and the incident solar radiation decrease due to the dust deposition over the PV module enhances significantly the module power reduction.

<p>This study aims to design a new architecture of the artificial neural networks (ANNs) using the Xilinx system generator (XSG) and its hardware co-simulation equivalent model using field programmable gate array (FPGA) to predict the behavior of Chua’s chaotic system and use it in hiding information. The work proposed consists of two main sections. In the first section, MATLAB R2016a was used to build a 3×4×3 feed forward neural network (FFNN). The training results demonstrate that FFNN training in the Bayesian regulation algorithm is sufficiently accurate to directly implement. The second section demonstrates the hardware implementation of the network with the XSG on the Xilinx artix7 xc7a100t-1csg324 chip. Finally, the message was first encrypted using a dynamic Chua system and then decrypted using ANN’s chaotic dynamics. ANN models were developed to implement hardware in the FPGA system using the IEEE 754 Single precision floating-point format. The ANN design method illustrated can be extended to other chaotic systems in general.</p>

Current signature analysis of induction motor becomes very much popular for its Condition monitoring purpose. Most of these studies in recent years are focused on hardware. Though, to build an experimental set-up for fault emulating machines is not an easy task. It involves too much cost and risk of damage. Moreover, it is not possible to create different types of faults in the same machine. In this paper an accurate but simple model of 3-phase Induction motor is being proposed which describes the machine at low frequency as well as at high frequency for different types of fault situations. The model, based on DM, CM characteristics of the machine, is well suited for power frequency (50 Hz) and high frequency applications like adjustable speed drives (ASDs). A mechanical model of a three-phase induction machine is incorporated to this electrical model to make a universal model of dynamic squirrel cage induction motor for overall simulation. The proposed model has the feature to incorporate different types of electrical fault like, stator inter-turn fault, phase-to-phase and phase-to ground insulation fault etc. The premature conditions of the above faults can also be emulated.