In this paper, a combined approach of Principal Component Analysis (PCA)-based Extreme Learning Machine (ELM) for boiler output forecasting in a thermal power plant is presented. The input used for this prediction model is taken from the boiler unit of the Yermarus Thermal Power Station (YTPS), India. Calculation of the accurate electrical output of a boiler in an operating system requires the knowledge of hundreds of operating parameters. The dimensionality of the input dataset is reduced by applying principal component analysis using IBM@SPSS Software. In the process of principal component analysis, a dataset of 232 parameters is standardized into 16 principal components. The total dataset collected is divided into training and testing datasets. The extreme learning machine is designed for various activation functions and the number of neurons. Sigmoid and hyperbolic tangent activation functions are studied here. Its generalization performance is examined in terms of the Mean Square Error (MSE), Mean Absolute Error (MAE), Root Mean Square (RMSE), and Mean Absolute Percentage Error (MAPE). ELM and PCA–ELM are compared. In both the ELM and PCA–ELM models, when the extreme learning machine was designed with a sigmoid activation function with 100 nodes in the hidden layer, RMSE was 5.026 and 4.730, respectively. Therefore, the developed combined approach of PCA–ELM proved as a promising technique in forecasting with reduced errors and reduced time.
Electric vehicles (EV) are promising alternate fuel technologies to curtail vehicular emissions. A modeling framework in a hybrid electric vehicle system with a joint analysis of EV in powering and regenerative braking mode is introduced. Bidirectional DC–DC converters (BDC) are important for widespread voltage matching and effective for recovery of feedback energy. BDC connects the first voltage source (FVS) and second voltage source (SVS), and a DC-bus voltage at various levels is implemented. The main objectives of this work are coordinated control of the DC energy sources of various voltage levels, independent power flow between both the energy sources, and regulation of current flow from the DC-bus to the voltage sources. Optimization of the feedback control in the converter circuit of HEV is designed using an artificial neural network (ANN). Applicability of the EV in bidirectional power flow management is demonstrated. Furthermore, the dual-source low-voltage buck/boost mode enables independent power flow management between the two sources—FVS and SVS. In both modes of operation of the converter, drive performance with an ANN is compared with a conventional proportional–integral control. Simulations executed in MATLAB/Simulink demonstrate low steady-state error, peak overshoot, and settling time with the ANN controller.
1 May 2016 in Indonesian Journal of Electrical Engineering and Computer Science
This paper deals with the design and simulation of an efficient solar photovoltaic system with a maximum power point tracking system (MPPT). Maximum power point (MPP) is obtained by using Perturb and Observe (P&O) algorithm. The output from solar panel is fed to the DC-DC (Boost) converter which steps up the output voltage. It is then fed to a 3-phase inverter. The inverter used is a 3-phase two-level inverter implemented with a Model Predictive Control strategy. Model of the system is considered in order to predict the control variables. Optimum switching state is selected by minimizing the cost function for each sampling period. This is achieved through modelling and MATLAB simulation of various stages that constitute the overall system.
1 December 2016 in Indonesian Journal of Electrical Engineering and Computer Science
This paper presents a Model Predictive Direct Power Control (MPDPC) strategy for a grid-connected inverter used in a photovoltaic system, as found in many distributed generating installations. The controller uses a system model to predict the system behavior at each sampling instant. Using a cost function, the voltage vector with least power ripple is generated. The resultant voltage vector is applied during the next sampling period which gives flexible power regulation. The effectiveness of the proposed MPDPC strategy is verified using MATLAB/ SIMULINK.