Z-Source Inverter Fed Induction Motor Drive control using Particle Swarm Optimization Recurrent Neural Network

Abstract

The suggestion is prepared for Particle Swarm Optimization (PSO) Recurrent Neural Network (RNN) based Z-Source Inverter Fed Induction Motor Drive in this document. The proposed method is employed to develop the presentation of the induction motor while decreasing the Total Harmonic Distortion (THD), eliminating the oscillation period of the stator current, torque and speed. Currently, as the input parameters, the PSO technique uses the induction motor speed and reference speed. It optimizes the raise of the PI controller and produces the reference quadrature axis current from the input parameters. By employing the RNN the reference three phase current for accurate control pulses of the voltage source inverter is predicted. The RNN is trained by the input motor actual quadrature axis current and the reference quadrature axis current with the associated target reference three phase current. The training process used the supervised learning process. Next the proposed technique is implemented in the MATLAB/simulink platform and the competence is scrutinized by comparing with the other techniques such as PSO-Radial Biased Neural Network (RBNN) and PSO-Artificial Neural Network (ANN). The comparison results show the superiority of the proposed approach and confirm its potential to effort out the problem. 8 9 10 11 12 13 14 15 16 17 18 leg is restricted as it short circuits the DC source [3]. In 30 conventional voltage-source inverter, the two switches 31 of the same-phase leg can in no way be gated on at the 32 same time as doing so would cause a short circuit (shoot 33 through) to occur, which would destroy the inverter [2]. 34 Z source inverter increases above the difficulties in the 35 Traditional VSI and CSI [6]. 36 The Z-source inverter is a dc voltage source which 37 prolonged by a relatively large capacitor supplies the 38 most important circuit of the inverter bridge [5]. Z-39 source inverter can improve dc input voltage with no 40 necessity of dc-dc boost converter or step up trans-41 former, hence prevailing over output voltage limitation 42 of traditional voltage source inverter with lower its 43 cost [1]. To take the isolated load, it is applied to 44 change variable magnitude, variable frequency voltage 45 into dependable constant voltage and steady frequency 46 1064-1246/15/$35.00 © 2015 -IOS Press and the authors. All rights reserved U n c o r r e c t e d A u t h o r P r o o f 2 R.S.S. Kumar and S.M. Girirajkumar / Z-Source Inverter Fed Induction Motor Drive control supply [11]. For both voltage-boosting and inversion 47 with an attractive inverter topology it presents a differ-48 ent choice [12]. The Z-source inverters can be planned 49 with their maximum modulation ratio set to the prevail-50 ing nominal case. Any flow in energy demand is next 51 supervised by fluctuating the inverter shoot-through 52 time duration, which in result is a third state introduced 53 for gaining voltage boosting in Z-source inverter [14]. 54 The induction motor drive system fed by z-source 55 inverter may have the benefits ...

RETRACTED: An efficient grasshopper optimization with recurrent neural network controller-based induction motor to replace flywheel of the process machine

“…The input parameters, for example, rotor speed and three phases current required for the proposed system, are accumulated from the induction motor. In this section, the adequacy of the proposed strategy is dissected by contrasting the implementation results and the distinctive existing strategies; for example, salp swarm optimization algorithm (Faris et al, 2018; Mirjalili et al, 2017b) and grasshopper optimization algorithm (Aljarah et al, 2018; Mirjalili et al, 2017a), binary salp swarm optimization algorithm (Faris et al, 2018), support vector machine (Aljarah et al, 2018), particle swarm optimization with recurrent neural network (Selva Santhose Kumar et al, 2015). The setpoint speed w r * and the actual speed w r are given to the input of the grasshopper optimization algorithm.…”

“…The recurrent neural network is worked based on the human brain and includes some artificial neurons that are artificial training and testing algorithms. Normally, the recurrent neural network structure involves three layers named as the input layer, hidden layer, and output layer (Selva Santhose Kumar and Girirajkumar, 2015). The actual quadrature axis current I Q and the setpoint quadrature axis current I Q * exist in the input layer.…”

RETRACTED: Low-voltage ride-through capability enhancement of wind energy conversion system using an ant-lion recurrent neural network controller

“…Utilizing distinctive forms of WECS, the fault protection schemes were examined and the upgrading of LVRT ability within the line during fault condition was pointed out by Howlader and Senjyu. 21 To upgrade the LVRT ability of doubly fed induction generator (DFIG), a functioning crowbar assurance (ACB_P) framework was depicted by Swain and Ray, 22 and it additionally enhances the PQ of the system. Not at all like the conventional crowbar (CB) was the protection scheme outlined with the resistor in series of capacitor and it having resistors as it were.…”

Robust energy management of micro‐grid system using an efficient RNN‐MDA approach

“…In general, input layer, output layer, hidden layer are three essential layers of the RNN. For demonstrating complex time‐dependent phenomena, a few kinds of RNN structures have been proposed 39 . Much advancement has been made with feed‐forward systems and consideration has as of late swung to creating algorithms for systems with recurrent connectionist which have critical capacities not found in feedforward systems including attractor elements and the capacity to store data for later utilization specifically noteworthy is their capacity to manage time‐shifting information or yield through their own natural temporal operation 40 .…”