ABSTRACT Detecting defects and classifying them is a primary requirement for textile industries. Manual methods of defect identification and classification happens in most cases for which the accuracy could not exceed more than 70%. A real time, fast and automated system for defect detection and classification is required for textile industries. This paper addresses this challenge and explores the use of different machine-learning models for the global features such as GLCM and Tamura extracted from checked pattern fabrics. In the training phase, machine-learning models produce an accuracy of > 90% and testing with new fabric images provides a maximum accuracy of 65%. To improve the accuracy of the classification system, an artificial neural network-based machine-learning model with features fusing has been proposed. Using the proposed approach, a testing accuracy of 80% is observed, and an accuracy of 96.1% is achieved in the training phase.

ABSTRACT Today, the use of permanent magnet brushless DC (PMBLDC) motors in vehicles is increasing due to the characteristics of sensorless operation. PMBLDC motor controllers can control the speed and position in the closed-loop feedback system without the need for a position sensor mounted on the shaft. Proportional – Integral – Derivative (PID) controller is one of the most common feedback control algorithms used in PMBLDC motor controllers. Optimizing problems using deterministic methods such as Lagrange and simplex methods requires basic information and complex calculations. Meta-heuristic algorithms are a type of stochastic algorithm that is used to find the optimal solution. Meta-heuristic algorithms are divided into three general categories: evolutionary algorithms, swarm intelligence algorithms, and stochastic algorithms. In this paper, using 14 metaheuristic optimisation algorithms, PID control parameters including settling time, time rise, overshoot, and stability of step response of the mentioned system are optimised. In this paper, 14 meta-heuristic algorithms are simulated and evaluated to optimise the PID coefficients of the controller, including settling time, rising time, excessive increase, and step response stability. The simulation result shows that the genetic algorithm (GA) has the best performance in terms of cost function and biogeography-based optimisation (BBO) in terms of settling time and rising time parameters. Finally, the simulation results are confirmed using experimental results.

ABSTRACT In this paper, an adaptive sliding mode controller (ASMC) based on fuzzy logic is applied to control hybrid smart microgrid power systems under uncertainty. The chattering effects can occur if not adjusted properly due to the sliding mode amplitude control law. The approximation of the ASMC law function is performed by Fuzzy logic, and chattering effects are reduced through a low pass filter. The Lyapunov theory proves the stability of the control system and is also applied to stabilise the system’s control mechanism. The hybrid micro-grid system operation, which consists of distribution generation, loads, PWM IGBT inverter and low pass filter, is simulated using MATLAB 2017b. The reference speed is compared with the hybrid microgrid real speed, which was 1.0 p.u. and gradually controlled through ASMC-fuzzy until the speed desired is achieved for power supply stability. A comparison of results between the simulated ASMC-Fuzzy and PI controller indicated a drastic reduction of THD, which satisfies the required ‘below 5% THD level according to IEEE 1547 standard’ required for operation.

ABSTRACT This paper presents a three-phase induction motor (IM) drive comprising a single-phase Pulse Width Modulated (PWM) converter with an active power decoupling (APD) as an active front-end converter (AFEC) to reduce the size of a DC link capacitor. Consequently, film capacitors can be used in place of bulky electrolytic capacitors that improve the power density, reliability, and cost of AFECs. A control scheme for an APD circuit is developed to mitigate the ripple power, and to control the three-phase IM, a maximum torque per ampere (MTPA) control scheme is implemented. A Simulink model of the proposed system rated for 3.73 kW, 200 V, and 50 Hz three-phase IM drive and its equivalent prototype model are developed in a laboratory and the APD and the MTPA control schemes are implemented with the help of STM32F407VG, a 32-bit microcontroller and dSPACE 1104, respectively. This paper demonstrates the experimental verification of the concepts for APD and MTPA control. Further, a significant reduction in motor current at starting/light load conditions is observed that leads to lower energy requirements. The prospective application of these schemes includes railway traction drive, electric vehicles, and so on, where compactness, reliability, and lower energy requirements are the prime concerns.

ABSTRACT This paper provides a modified single-ended primary inductor converter (SEPIC) with an asymmetrical switched inductor super lift cell (ASISLC) for renewable energy applications. Solar, fuel cells, and so on, provide low output voltage, which is not sufficient for high voltage dc grid or dc loads. Therefore, the existence of a high gain DC-DC converter is mandatory. The proposed converter provides high gain, high efficiency, and less voltage stress at a low duty cycle. The proposed converter incorporates switched inductor together with a super lift cell. The super lift cell has a charge pump capacitor is connected in an asymmetrical fashion. The proposed converter offers continuous current at the input together with the above-mentioned advantages. This paper explains the operation of the ASISLC SEPIC converter in continuous conduction mode (CCM) and discontinuous conduction mode (DCM). The operating principle, theoretical analysis, design process, and comparison of the suggested converters with converters from the literature that use similar operating principles are all discussed in this paper. A MATLAB simulation is done and their results are provided. Finally, a hardware prototype for 200W, 50 kHz switching frequency is built and their results are validated to authenticate the effectiveness of the proposed converter.

ABSTRACT This paper is devoted to the design of a new sinusoidal oscillator using a recently reported configurable analog block. The EXCCII-based circuit uses grounded capacitors, provides independent control over frequency and condition, and the circuit is easy to design. The use of a single active block and four passive components makes the circuit economical in terms of total required components. The detailed non-ideal and parasitic study is given, and design considerations are also discussed. The ratio tuning of oscillating frequency through capacitors is another feature, which is covered and justified through actual simulation results. A modification to the circuit is made for obtaining cascade-able voltage output, which is insensitive to loading effects. The operation using CMOS implementation of EXCCII with the supply voltage of ±1.25 V is verified for different designs. The study provides a further insight to the design of oscillators using current mode elements in general and more specifically by employing a configurable analog block.

ABSTRACT Multilayer-Perceptron Neural Network (MLP-NN)-based sensorless speed control of adaptive Indirect Field-Oriented Control (IFOC) strategy is implemented for online parameter estimation of Induction Motor Drive (IMD) fed from Common mode voltage injection Space vector PWM (CVSVPWM) based Voltage Source Inverter. Harris Hawks Optimization (HHO) is implemented in this work, to train the MLP-NN model by choosing the optimal weight and biases for the estimation of accurate parameters and speed of IMD. The objective of optimal MLP-NN is to improve the IMD reliability and response fast during dynamic operation. The model performances are evaluated by employing statistical metrics of MSE, RMSE, MAE, MAPE, and R for training and testing. These are reported for testing to be 0.000602064, 0.0245, 0.4015, 0.25474, and 0.9997 which indicates the best-fitted prediction model and proves the minimized error. The results reveal that an optimized MLP-NN accomplishes promising performance in estimating the parameters and speed with the least errors such as rs is 3.82%, rr is 4.19%, ls is 0.41%, lr is 0.72%, lm is 0.21%, and strongly tracking of reference speed. In addition, HHO is also employed to evolve the gains of the PI-controller in adaptive-IFOC for generation of reference signals by reducing the computational effort.