5975 research outputs found
Sort by
Clustering Prediction Techniques in Defining and Predicting Customers Defection: The Case of E-Commerce Context
No full textWith the growth of the e-commerce sector, customers have more choices, a fact which encourages them to divide their purchases amongst several e-commerce sites and compare their competitors’ products, yet this increases high risks of churning. A review of the literature on customer churning models reveals that no prior research had considered both partial and total defection in non-contractual online environments. Instead, they focused either on a total or partial defect. This study proposes a customer churn prediction model in an e-commerce context, wherein a clustering phase is based on the integration of the k-means method and the Length-Recency-Frequency-Monetary (LRFM) model. This phase is employed to define churn followed by a multi-class prediction phase based on three classification techniques: Simple decision tree, Artificial neural networks and Decision tree ensemble, in which the dependent variable classifies a particular customer into a customer continuing loyal buying patterns (Non-churned), a partial defector (Partially-churned), and a total defector (Totally-churned). Macro-averaging measures including average accuracy, macro-average of Precision, Recall, and F-1 are used to evaluate classifiers’ performance on 10-fold cross-validation. Using real data from an online store, the results show the efficiency of decision tree ensemble model over the other models in identifying both future partial and total defection
Design and Analysis of In-Wheel Double Stator Slotted Rotor BLDC Motor for Electric Bicycle Application
No full textThis paper discusses about design and analysis of double stator slotted rotor (DSSR) BLDC motor for electric bicycle application. Usually single stator (SS) BLDC motor is used in an electric bicycle. This type of motor has low performance and need to be charged regularly. The objective of this research is to design and analysis DSSR motor that have high torque. At starts, design specification for the electric bicycle is calculated. Next, design process for DSSR is carried out by using the desired parameter. Lastly, analysis for double stator slotted rotor is simulated using FEM. Result for average back emf, average inductance, inner stator flux density, outer stator flux density, average torque and estimate torque constant is obtained. Result for average torque from FEM archieve the requirement of motor torque for DSSR design where the maximum average torque is 16.2 Nm. This research will give benefit to mankind and society in term of environment protection and energy consumption
Artificial Intelligence Control Applied in Wind Energy Conversion System
No full textThe objective of this paper is to study the dynamic response of the wind energy conversion system (WECS) based on the Doubly Fed Induction Generator (DFIG). The DFIG rotor is connected to the grid via a converter. The active and reactive power control is realized by the DFIG rotor variables control, using the field oriented control (FOC). The vector control of DFIG is applied by the use of tow regulators PI and the neural network regulator (NN). The generator mathematical model is implemented in Matlab/ Simulink software to simulate a DFIG of 1.5 MW in order to show the efficiency of the performances and robustness of the studied control systems. The simulation obtained results shows that the robustness and response time of the neural network regulator is better than those obtained by the PI classical regulator
Performance of Brushless DC Drive with Single Current Sensor Fed from PV with High Voltage-Gain DC-DC Converter
No full textThis paper presents the performance of Brushless DC (BLDC) Motor drive with only one positioning sensor instead of three conventional sensors. The three sensor units are replaced with a single stator current sensor unit in DC bus which further reduces the cost increasing the reliability of the drive system. Using a single sensor in stator requires minimum electronic equipment for the purpose of measurement process. This paper evolves the BLDC motor drive fed from PV system. A high voltage-gain DC-DC converter is presented in this paper to step-up the voltage from PV system. The appropriateness of PV fed BLDC motor drive is verified for variable increamental speed with fixed torque and variable decremental speed with fixed torque operating conditions. BLDC motor drive performance is also performed for variable torque with fixed peed working condition. The proposed system and results are developed using MATLAB/SIMULINK software
Review of DC Offset Compensation Techniques for Grid Connected Inverters
No full textLimitations of DC injection into the AC network is an important operational requirement for grid connected photovoltaic systems. There is one way to ensure that this issue needs a power transformer as a connection to the AC network. However, this solution adds cost, volume, mass, and power losses. Ideally there shouldn't be any DC at the output of the inverter, but practically, a small amount of DC current is present. Therefore, in this paper there are techniques for the DC offset elimination are proposed. Some have drawbacks which was treated by another technique. Also there are best solutions for eliminating DC offset as in section 17, and 18 as it explains how to reduce the DC offset in a transformerless operation with reducing the power losses, mass and the cost effect
Direct Torque Control Strategy of PMSM Employing Ultra Sparse Matrix Converter
No full textMatrix converter is a good choice for Permanent Magnet Synchronous Motor (PMSM) drives, because it has high power density and does not require dc-link energy storage. the disadvantages of conventional matrix converter is using 18 active switches, so it becomes expensive and the modulation method becomes more complicated than back to back converter. To minimize this problem, this paper proposes variable speed drive of PMSM using Ultra Sparse Matrix Converter (USMC) based on Direct Torque Control (DTC) methods. This converter uses only 9 active switches, making it cheaper than conventional matrix converter. DTC is designed based on Space Vector Modulation (SVM) to reduce torque and flux ripples due to the hysteresis control in conventional DTC. The simulation results show that DTC based SVM using USMC effectively controls the rotor speed with low torque and flux ripples
Improvement of Power Quality Using Fuzzy Controlled D-STATCOM in Distribution System
No full textThis paper investigates the problems associated with distribution system in terms of delivery of clean power and their solutions. Power quality has become a major issue in the present power system network. The network has mostly inductive nature. This draws more reactive power. This causes harmonics and voltage unbalance problems. So maintain the proper operation of interconnected power system, we are using one of the facts devices such as fuzzy controlled D-statcom. It provides suitable compensation and there by maintain proper power factor and also reduces harmonic contents. The simulation is taken out by MATLAB/SIMULINK and the result shows the effectiveness of GA (Genetic algorithm) simulation. Optimized Fuzzy controlled D-STATCOM for improvement of power quality
FPGA Based Symmetrical Multi Level Inverter with Reduced Gate Driver Circuits
No full textMultilevel converters tender advantages in terms of the output waveform quality due to the increased number of levels used in the output voltage modulation and have been widely accepted for high-power high-voltage applications. This paper introduces topology in multilevel dc link inverter (MLDCLI), which can significantly reduce the switch count and improve the performance. The preferred topology provides a dc voltage with the shape of a staircase approximating the rectified shape of a commanded sinusoidal wave, to the bridge inverter, which in turn gives the required alternating waveform. This topology requires fewer components compared to traditional Multi level Inverters (MLI).Therefore, the overall cost and complexity are significantly reduced particularly for higher output voltage levels. Finally, Matlab/Simulink and XILINX are used as a simulation and compiler architecture of control circuit embedded in FPGA. Simulation and experimental results for fifteen-level inverter are presented for validation
Minimizing the loses of solar power generation by designing an intelligent tracking system implemented on FPGA
No full textThe increasing of using of an electrical power as a power source in a large number of devices can occur a serious problem in our daily life. One of the useful power sources is a solar cell which used to overcome many problems of power generation. In this paper, the solar cell model is proposed to minimize loses of solar power generation by designing of an intelligent tracking system based on FPGA. PSO algorithms are used to train the neural networks to control the speeds and the directions of rotations of two DC motors with the help of FPGA cart. The proposed system was implemented in MATLAB; and for the hardware part, FPGA was used for the implementation of neural networks
Using FPGA Design and HIL Algorithm Simulation to Control Visual Servoing
No full textThis is a novel research paper provides an optimal solution for object tracking using visual servoing control system with programmable gate array technology to realize the visual controller. The controller takes in account the robot dynamics to generate the joint torques directly for performing the tasks related to object tracking using visual servoing. Also, the notion of dynamic perceptibility provides the capability of the designed system to track desired objects employing direct visual servoing technique. This idea is assimilated in the suggested controller and realized in the programmable gate array. Additionally, this paper grants an ideal control framework for direct visual servoing robots that incorporates dynamic perceptibility features. With the aim of evaluating the proposed FPGA based architecture, the control algorithm is applied to Hardware-in-the-loop simulation (HIL) set up of three degrees of freedom rigid robotic manipulator with three links. Furthermore, different investigations are performed to demonstrate the behavior of the proposed system when a trajectory adjacent to a singularity is attained