Shahril Irwan Sulaiman

Sizing grid-connected photovoltaic system using genetic algorithm

This paper presents an intelligent-based algorithm for sizing grid-connected photovoltaic (GCPV) system using Genetic Algorithm (GA). GA had been used to determine the optimal PV module and inverter from pre-developed PV module and inverter databases such that the expected technical performance of the design could be optimized. In addition, the technical sizing outputs such as the number of photovoltaic (PV) modules, PV array configuration and inverter-to-PV array sizing factor were computed. The GA had outperformed the Evolutionary Strategies (ES) during the sizing process in terms of producing better optimum results. Low error had also been produced by GA when compared to a benchmark sizing algorithm using iterative sizing approach.

A hybrid artificial neural network for grid-connected photovoltaic system output prediction

This paper presents a hybrid Particle Swarm Optimization-Artificial Neural Network (PSO-ANN) for predicting the kWh output from a grid-connected photovoltaic (GCPV) system. In this study, the ANN-based prediction utilized solar irradiance (SI), ambient temperature (AT) and module temperature (MT) as the inputs and kWh energy from the GCPV system as the sole output. Besides that, Particle Swarm Optimization (PSO) was used to optimize the number of neurons in the hidden layer during the ANN training process such that the Root Mean Square Error (RMSE) of the prediction was minimized. After the training process, testing was performed to validate the ANN training. The results showed that the proposed hybrid PSO-ANN had outperformed the hybrid Fast Evolutionary Programming-Artificial Neural Network (FEP-ANN) in producing lower RMSE. In addition, the optimal learning algorithm and population size in PSO were also investigated in this study.

Partial Evolutionary ANN for Output Prediction of a Grid-Connected Photovoltaic System

This paper presents the Evolutionary Programming (EP) based technique to optimize the architecture and training parameters of a one-hidden layer backpropagation Artificial Neural Network (ANN) model for the prediction of total AC power output from a grid connected photovoltaic system. A partial Evolutionary Programming-ANN (EPANN) model has been developed for the prediction. It utilizes solar radiation, wind speed and ambient temperature as its inputs while the output is the total AC power produced from the grid connected PV system. EP is used to optimize the regression performance of the ANN model by determining the optimum values for the number of nodes in the hidden layer as well as the optimal momentum rate and learning rate for training. The results obtained from the EPANN have been compared with the results from a classical ANN with similar input and output settings. It is observed that the prediction of total AC power output from a grid connected PV system could be accelerated and simplified using the partial evolutionary ANN model. Index Terms—Artificial neural network (ANN), Correlation coefficient (R), Evolutionary programming-ANN (EPANN), and Photovoltaic (PV).

Optimizing three-layer neural network model for grid-connected photovoltaic system output prediction

This paper presents the Evolutionary Neural Network (ENN) model for the prediction of output from a grid-connected photovoltaic system installed at Malaysian Energy Centre (PTM), Bangi, Malaysia. The ENN model had been developed using Evolutionary Programming (EP) through the optimization of the number of nodes in the hidden layer, the learning rate and the momentum rate. The ENN model employs solar irradiance and ambient temperature as its inputs while the sole output is the kilowatt-hour energy output produced from the grid connected PV system. On the other hand, the objective function of the ENN is to maximize the correlation coefficient, R of the prediction task. In this study, the optimal pool population size in the ENN algorithm was investigated. Apart from that, the maximum average correlation coefficient obtained for the ENN training is 0.9942. Besides that, the testing process produced sufficiently high correlation coefficient value of 0.9922.

Power prediction for grid-connected photovoltaic system in Malaysia

This paper presents the prediction performance and analysis of the power output from grid-connected photovoltaic (GCPV) system using power model. The power model used is presently applied in the testing, commissioning and acceptance test of grid-connected systems procedure in Malaysia. This study involved outdoor testing and validation of model through mathematical formula, root mean square error (RMSE), Pearson correlation coefficient (R2) and standard deviation (STD) determination. Results showed that most of the time, the actual power output was lower than the modelled power output. This implies that more derating factors should be considered in the existing power model such as aging and installation criteria.

Optimizing one-hidden layer neural network design using Evolutionary Programming

This paper presents the optimization of one-hidden layer Artificial Neural Network (ANN) design using Evolutionary Programming (EP) for predicting the energy output of a grid-connected photovoltaic system installed at Malaysian Energy Centre (PTM), Bangi, Malaysia. In this study, the architecture and training parameters of the multi-layer feedforward back-propagation ANN model had been optimized while the prediction performance of the ANN was maximized. The proposed Evolutionary Programming-ANN (EPANN) model employs solar radiation and ambient temperature as its inputs while the sole output is the kilowatt-hour energy output produced from the grid connected PV system. The prediction performance was quantified using the average correlation coefficient and it was maximized by determining the optimum values for the number of nodes in the hidden layer, momentum rate and learning rate during an evolutionary training. Besides searching for the optimal number of nodes and optimal training parameters for each model, the highest correlation coefficient for the prediction required for the EPANN was investigated. It was found that the maximum average correlation coefficient obtained for the EPANN training is 0.9962. Besides that, the testing process produced sufficiently high correlation coefficient value of 0.9976.

A hybrid bat Algorithm Artificial Neural Network for grid-connected photovoltaic system output prediction

This research have been conducted to predict the output power of grid-connected photovoltaic (GCPV) system using hybrid Bat Algorithm-Artificial Neural Network (BA-ANN) in this paper. In this project, ANN utilized data from GCPV database includes Solar Irradiance (SI), Ambient Temperature (AT) and Module Temperature (MT) as the inputs and apply output power as a single output. More importantly, bat algorithm optimization was apply to minimize Root Mean Square Error (RMSE) by optimized the number of neurons in the hidden layer, learning rate and momentum rate. After training steps, testing will take a part for affirm the ANN training. The results obtained have been compared with the results from Evolutionary Programming-Artificial Neural Network (EP-ANN) with the similar input and output configurations. It is observed that result for BA-ANN had performed more than EP-ANN in term of producing lower RMSE. Besides that, optimal learning algorithm, time taken, and population were also take part in this research.

Harmony search-based optimization of artificial neural network for predicting AC power from a photovoltaic system

Grid-Connected Photovoltaic (GCPV) system is a type of photovoltaic (PV) systems which has been widely used as a renewable-based electricity generation. Nevertheless, the intermittency and fluctuation in weather conditions have caused inconsistent and varying output performance of a GCPV system. This paper presents a Multi-Layer Feedforward Neural Network (MLFNN) model for predicting the AC power from a GCPV system. Harmony Search (HS) was also employed to optimize several MLFNN parameters such that the prediction error could be minimized. The AC Watt-output of a GCPV system was predicted using MLFNN with solar irradiance, ambient temperature and operating PV module temperature as its inputs. These data were collected from a GCPV system located at Green Energy Research Centre (GERC), Universiti Teknologi MARA, Malaysia. In optimizing the MLFNN, HS was introduced to determine the optimal number of neurons in hidden layer, the learning rate and the momentum rate during training. After the training, testing process was conducted to validate the training process. In both training and testing, the prediction performance was quantified using Root Mean Square Error (RMSE). The performance of the HS-MLFNN was later compared with the performance of an Evolutionary Programming (EP)-MLFNN in predicting the AC power. The results showed that the hybrid HS-MLFNN had outperformed the hybrid EP-MLFNN by producing lower RMSE during both training and testing.

Artificial neural network versus linear regression for predicting Grid-Connected Photovoltaic system output

This paper presents a classically trained Multi-Layer Feedforward Neural Network (MLFNN) technique for predicting the output from a Grid-Connected Photovoltaic (GCPV) system. In the proposed MLFNN, the selection of the training parameters was conducted using a series of prescribed steps. The MLFNN utilized solar irradiance (SI) and module temperature (MT) as its inputs and AC kWh energy as its output. When compared with the linear regression method, the MLFNN offered superior performance by producing lower prediction error.

Artificial immune system for sizing grid-connected photovoltaic system

This paper presents an intelligent-based sizing algorithm for the design of grid-connected photovoltaic (GCPV) system using artificial immune system (AIS). AIS had been used to select the optimal PV module and inverter from pre-developed PV module and inverter database such that the expected technical performance of the design could be optimized. The AIS had outperformed the genetic algorithm (GA) and particle swarm optimization (PSO) during the sizing process in terms of producing better optimum results despite having the largest computation time.