Figures
Abstract
This paper presents a control method for a system composed of a photovoltaic (PV) array, five-phase impedance source inverter, five-phase induction motor and centrifugal pump. This method is based on controlling the motor speed to control the pump power as the insolation level or temperature change to attain the maximum power extraction from the PV-array. The motor speed is controlled by using artificial neural network (ANN) which is trained to provide the desired inverter frequency and modulation index at any insolation level and temperature to attain the maximum PV operating power. The data of the neural network are based on the operation of the induction motor at constant air gap flux and perturb and observe method for maximum power point tracking. Simulation results are obtained using MATLAB Simulink to verify the proposed control method.
Citation: Hassan AM, Hassan AE-W, Elbarbary ZMS, Al-Gahtani SF, Omar AI, Metwally ME (2024) MPPT control of a solar pumping system based five-phase impedance source inverter fed induction motor. PLoS ONE 19(1): e0295365. https://doi.org/10.1371/journal.pone.0295365
Editor: Praveen Kumar Balachandran, Vardhaman College of Engineering, INDIA
Received: October 18, 2023; Accepted: November 21, 2023; Published: January 18, 2024
Copyright: © 2024 Hassan et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting information files.
Funding: The authors extend their appreciation to the Deanship of Scientific Research at King Khalid University under for funding this work through General Research Project under Grant number (RGP.1/373/44).
Competing interests: The authors have declared that no competing interests exist.
Introduction
The use of a photovoltaic (PV) array as a source of electrical energy has many benefits. It is both noiseless and dependable, while also being environmentally friendly and simple to install. Although it requires little maintenance, its initial installation costs can be high, and its power conversion efficiency can be low. Nevertheless, photovoltaic pumping is an encouraging application of photovoltaic energy. By storing solar energy as potential energy in a water reservoir, energy can be consumed on demand. This eliminates the need for large, expensive, and heavy banks of lead acid batteries, which have a significantly shorter lifespan than photovoltaic panels. However, it is crucial to maintain an efficient power conversion chain from panels to mechanical pump, despite the lack of batteries [1]. Solar-powered pumping systems have become increasingly popular due to their numerous benefits over traditional diesel pumps. These advantages include lower operational and maintenance costs, as well as improved reliability. Since the demand for water is highest during the day, solar pumping systems are well-suited to meet this need. However, they are also subject to weather and environmental conditions that can affect their performance. This means that intermittent power shortages may occur, particularly on cloudy days. To reduce costs and improve efficiency, it is important to operate the PV source at its maximum power point. This will prevent the need for oversized PV panels and improve the reliability of the entire system [2].
There have been several studies that have dealt with the solar pumping systems [1–14]. A field-oriented control method for a battery less solar pumping system composed of a push-pull converter and three-phase induction motor was presented in [1]. In [2], a field-oriented control method for a standalone solar powered water pumping system employing permanent magnet synchronous motor was presented. The effect of voltage and load variation on a solar pumping system employing three-phase induction motor was investigated via experimental results in [4]. A cost effective solar pumping system employing brushless direct current motor was presented in [5]. Control methods for a solar pumping system employing an open-end winding induction motor and dual inverter were presented in [6,7]. In [8], a control method was introduced for a single stage solar pumping system utilizing brushless direct current motor and a common voltage source inverter. Design and control of a solar pumping system employing switched reluctance motor and dual supply buck-boost converter was presented in [9]. In [10], a CUK-SEPIC converter for solar water pumping system was presented. This system was operated at maximum power point using a hybrid gravitational search algorithm and particle swarm optimization. The motor used in this system was switched reluctance motor. In [11], a hybrid photovoltaic-wind water pumping system was presented. This system employed a CUK converter, battery bank, permanent magnet synchronous generator, rectifier, three-phase inverter, and three-phase induction motor. The system was controlled using a modified artificial bee colony based on maximum power point tracking algorithm. In [12], a modified hybrid grey wolf optimization fuzzy logic controller based maximum power point tracking for a solar pumping system employing switched reluctance motor was presented. An optimized hybrid firefly based ant colony optimization as maximum power point tracking for a solar pumping system utilizing permanent magnet synchronous motor was presented in [13]. In [14], a hybrid firefly algorithm-particle swarm optimization based maximum power point tracking for a solar pumping system employing three-phase induction motor was presented. However, there is not any research that has dealt with solar pumping systems based on five-phase impedance source inverters and five-phase induction motors. The impedance source inverter has many advantages, such as superior efficiency and compact structure. The impedance source inverter is preferred over conventional two-stage inverters because of its fewer power devices [15,16].
Electric machines utilizing multiphase inverter technology are currently preferred over conventional three-phase machines and inverters. This is due to several advantages such as the reduction of torque pulsation amplitude and increase of frequency of these pulsations, higher torque density, decreased per phase rotor harmonic current distortion without the need for an increase in voltage per phase, lower dc-link current ripples, better noise characteristics, and higher reliability with increased fault tolerance [17–21].
In this paper, a control method is proposed for a water pumping system composed of PV array, five-phase impedance source inverter (ZSI), five-phase induction motor and centrifugal pump. This method is based on controlling the motor speed to control the pump power as the insolation level or temperature change to attain the maximum power point. The motor speed is controlled by using an artificial neural network which is trained to provide the desired inverter frequency and modulation index at any insolation level and temperature to attain the maximum PV operating power. The data of the neural network are based on the operation of the induction motor at constant air gap flux and P&O method for MPPT. The main contribution of this work can be summed up as follows:
- A new standalone solar pumping system is proposed. This system is composed of PV array, five-phase ZSI, five-phase induction motor and centrifugal pump.
- A new control method for the standalone solar pumping system is proposed. This control method achieved maximum utilization of solar energy.
- The artificial neural network is employed in the control methodology to improve the system performance.
- The system is simulated using MATLAB SIMULINK to prove the validity of the proposed control methodology for the proposed solar pumping system.
System modelling
In this section, each part of the system is modelled in the following subsections.
PV array model
The PV array is composed of several modules and each module consists of a number of solar cells [22]. The electrical equivalent circuit of a solar cell is shown in Fig 1 [23,24].
The expression relating the PV cell output voltage, V, and output current, I, is given by [23]:
(1)
Where
, Iph is the cell photo current, Io is the cell reverse saturation current, q is the electron charge, A is an ideality factor, k is Boltzmann’s constant and Tcell is the absolute cell temperature [25]. When the PV array is formed from several parallel strings, Np, and each string consists of number of series connected modules, Nm, and each module consists of number of series connected cells, Ns, the expression relating the PV array output voltage, Vg, to its output current, Ig, can be obtained from:
(2)
where Nsg = Nm.Ns, Iphg = Iph.Np, Iog = Io.Np,
,
,
, Iphg is the PV array photocurrent, Iog is the PV array reverse saturation current, Rsg is the PV array series resistance, Rshg is the PV array shunt resistance and Λg is the factor of the PV array [26,27].
In order to use Eq (2) at any insolation level, the per unit insolation level, G, is included into the equation as follows:
(3)
where in Eq (4), Iphg is the array photocurrent at an insolation level of 100% (1000 W/m2).
Five-phase impedance source inverter model
The five-phase impedance source inverter includes impedance network that connects the main converter circuit with a DC source, PV array [28]. This network consists of two inductors, with equal inductance (L), and two capacitors, with equal capacitance (C), Fig 2. The diode in Fig 2 prevents the reverse current flow. The most important feature of this inverter is that its output voltage can be controlled from zero to infinity, i.e., the impedance source inverter is a buck-boost inverter. Either the conventional voltage source inverter or current source inverter cannot attain this feature [29].
The five-phase impedance source inverter shown in Fig 2 has three groups of switching states. The first group is called active states where the load is connected directly to the supply. The second group is called null states where either all the upper or all the lower switches are conducted simultaneously. The third group is called shoot-through state in which the dc-link terminals are shorted by gating both the upper and lower switches of at least one inverter leg. During the shoot through state the capacitor voltage is boosted by receiving the energy from the inductor. The total number of shoot through states is 31 states [30].
The impedance-source network voltage of the impedance source five-phase should verify equal inductors voltage, vL, and equal capacitors voltages, vc [29]. In a sampling period T, the inverter bridge enters the shoot-through state for a duration To. During this time, the diode (D) in the power circuit (Fig 2) becomes reverse-biased, and the PV source is isolated from the Z-network. The inductors are charged by the Z-source capacitors in this mode, which is included in the conventional zero periods (TZ) of every switching cycle (T). The shoot-through interval or its duty cycle is determined by the necessary voltage boost. However, it’s worth noting that the shoot-through interval is only a small fraction of the switching cycle, requiring a relatively small capacitor to suppress the voltage [29]. The voltage equation of the Z-network can be expressed as follows [29].
When the output voltage of the PV array, denoted as VPV, matches the DC link voltage (vdc), the two voltages are equal.
Eq (5) provides a way to represent the maximum DC link voltage present on the inverter bridge.
(6)
Where B represents the boost factor, while D0 is the shoot-through duty ratio and is defined as the ratio of the time during which both switching devices in a converter are on simultaneously to the total switching period and is equal to
.
The relationship between the boost factor (B) and the shoot-through duty ratio is given by the following equation:
(7)
In the steady state of non-boost mode operation, the voltage across the Z-source capacitor equals the voltage across the DC-link of the ZSI. To put it differently,
(8)
The maximum power point voltage VPV*of the PV panel determines the AC output voltage control in the inverter. The modulation index variation enables control over the peak value of the AC output voltage, which can be mathematically represented.
(9)
where m is the modulation index.
There are several types of modulation techniques used to control the inverter. The most conventional modulation techniques are simple boost control, maximum boost control, maximum boost control with third harmonic injection, and modified space vector pulse width modulation control [31]. The simple boost control modulation technique is adopted in this study.
Simple boost control modulation technique
This technique employs two values Vp and Vn as shoot-through modulating signals. These signals have positive and negative peak values of five-phase sinusoidal reference signals as shown in Fig 3 [30]. For a unity carrier signal height, Vp corresponds to value of m [32]. Shoot through occurs only when carrier signal is greater than Vp or lower than Vn, otherwise it operates just as traditional sinusoidal pulse width modulation (SPWM). From Fig 3, shoot-through duty ratio Do is related to modulation index m as follows:
(10)
Substituting Eq (10) into Eq (7), we can obtain the boost factor B as follows:
(11)
Substituting Eq (11) into Eq (9), the peak value can be obtained from:
(12)
Five-phase induction motor model
The five-phase induction motor is modeled in terms of the DQ model in the stationary reference frame. The Voltage ABC to DQ Transformation can be obtained from [33]:
(13)
Where α = 2π/5.
The current DQ to ABC Transformation can be obtained from [33]:
(14)
The following equation represents the DQ motor voltage equations [33]:
(15)
and
(16)
The above motor differential equation can be rewritten in the following form in which the matrix inversion is avoided:
(17)
where
The developed torque can be obtained as follows:
(18)
When the variation of the motor speed is taken into consideration during the transient period, Eq (13) becomes nonlinear and numerical solution to obtain the motor currents is required. For this purpose, the motor mechanical equation is used. The induction motor mechanical equation is given by:
(19)
where ωm is the motor speed in rad/sec, J is the inertia and Tl(ωm) is given by:
(20)
Where TL is the load torque, Tfw is the friction and windage torque and Te is obtained from Eq (18).
Control methodology
The main purpose of controlling the system under consideration is to extract the maximum power from the PV array as the insolation level and temperature change [35]. This is achieved by controlling the motor speed as the insolation level changes or temperature [36]. When either the insolation level or temperature changes, the motor frequency and voltage, constant air-gap flux, will be controlled to change motor speed and thus, the pump power, Eq (20), extract the PV array maximum power. The motor frequency is controlled by controlling the inverter reference frequency [37]. The motor voltage is controlled by controlling the modulation index to maintain constant air-gap flux. The PV array maximum power is tracked by perturb and observe method. This method is achieved by changing the modulation index and the inverter frequency until the maximum power is extracted from the PV array. The inverter frequency is obtained from a lookup table whose data are obtained from steady-state analysis for the induction motor which is controlled by operating the motor at constant air-gap flux. The result of the analysis gives data for the motor input power, which is assumed to be equal to PV-array output power, and the corresponding data for the inverter frequency. The algorithm is shown in Fig 4.
The Artificial Neural Network (ANN) is a widely utilized technology in various fields of research and application [38]. It proves beneficial in solving complex problems that cannot be effectively addressed through traditional mathematical approaches [39]. Numerous studies have proposed the application of ANN technique in selecting voltage inverter switches for IM power supply [40,41]. The concept involves replacing the conventional switching table, which determines the inverter states, with a neural selector capable of managing control signals. This architecture incorporates a multilayer neural network that replaces hysteresis comparators and the selection table.
The data obtained from the above-mentioned method are used in an artificial neural network to generate the reference modulation index and reference frequency at a given insolation level and temperature as shown in Fig 5. Training data are generated from Simulink of PV system model obtained from Simulink MATLAB® simulation model designed. In this work the neural network used to emulate the look up table based the desired insolation level and reference frequency.
The structure of the proposed neural network has 3 input nodes, 8 neurons in the hidden layer, and 4 neurons in the output layer. The type of activation function used is a sigmoid function, where the sigmoid functions commonly used in back-propagation networks, in part because it is differentiable. The input data to the proposed neural network are insolation level, temperature, actual generated power. The outputs of the neural network are reference modulation index and reference frequency.
Simulation results
Dynamic results are obtained for the system under consideration for different values of insolation levels, 100%, 80%, 40% and 60% at 25 C° temperature. The inverter switching frequency is taken to be 2 kHz. The five-phase induction motor parameters and the PV-array data are given in appendix. These results are given in Figs 6 to 11 to prove the validity of the proposed method of control.
From Fig 6(a), we find that when the insolation level decreases to 80% and 40%, the frequency of the voltage coming out of the inverter will decrease, as in Fig 6(b), thus reducing the speed of the motor. By applying Eq (20), we find that the power of the motor decreases and thus the pump power decreases. We also find that when the insolation level rises to 60%, the frequency of the voltage coming out of the inverter will increase, and thus the speed of the motor will increase, the power of the motor will increase, and thus the power of the pump will increase.
It can be seen from Fig 7 that when the insolation level decreases, the current of the solar cell decreases and the voltage increases, and thus the power output from it changes. Thus, we find that the photovoltaic array operates at the maximum power point for the characteristics of the photovoltaic array, as shown in Fig 8. It can be seen from Fig 9 that the input voltage and inverter current change with the change in insolation level and contain ripples. It can also be noted from Fig (10) that the motor line current is almost sinusoidal, because of the inductance in the motor windings. It can also be noted from Fig 11 that as the insolation level changes, the operating speed changes in response to reaching the maximum power point.
Another set of dynamic results are obtained for the system under consideration for different values of temperature, 25°C, 35°C, 45°C, at 100% insolation level. The PV array is operated at maximum power point at the different values of temperature. These results are given in Figs 12 to 17 to again prove the validity of the proposed method of control.
It can be noticed from Fig 12 that the effect of changing the temperature on the values of modulation index and the reference frequency is small. This is expected because the effect of temperature on the PV-array maximum power point is small as shown in Fig 13. Therefore, the effect of temperature variation on the other quantities is very small as shown in Figs 14–17, which proves the validity of the proposed method.
Conclusions
A control method is proposed for solar pumping system based five-phase impedance source inverter and five-phase induction motor. This method is based on controlling the motor speed to control the pump power as the insolation level or temperature change to attain the maximum power extraction from the PV-array. The motor speed is controlled by using an artificial neural network which is trained to provide the desired inverter frequency and modulation index at any insolation level and temperature to attain the maximum PV operating power. The data of the neural network are based on the operation of the induction motor at constant air gap flux and perturb and observe method for maximum power point tracking. Dynamic simulation results are obtained using MATLAB Simulink that proved the validity of the proposed control method.
References
- 1. Vitorino MA, De Rossiter Corrêa MB, Jacobina CB, Lima AMN. An effective induction motor control for photovoltaic pumping. IEEE Trans Ind Electron. 2011;58: 1162–1170.
- 2. Antonello R, Carraro M, Costabeber A, Tinazzi F, Zigliotto M. Energy-efficient autonomous solar water-pumping system for permanent-magnet synchronous motors. IEEE Trans Ind Electron. 2017;64: 43–51.
- 3.
Barlow R, McNelis B, Derrick A. Solar pumping: an introduction and update on the technology, performance, costs, and economics. World Bank Tech Pap. United States; 1993. https://www.osti.gov/biblio/6675026.
- 4. Kini PG, Bansal RC. Effect of voltage and load variations on efficiencies of a motor-pump system. IEEE Trans Energy Convers. 2010;25: 287–292.
- 5. Youssef MZ. Design and Performance of a Cost-Effective BLDC Drive for Water Pump Application. IEEE Trans Ind Electron. 2015;62: 3277–3284.
- 6. Jain S, Thopukara AK, Karampuri R, Somasekhar VT. A single-stage photovoltaic system for a dual-inverter-fed open-end winding induction motor drive for pumping applications. IEEE Trans Power Electron. 2015;30: 4809–4818.
- 7. Jain S, Karampuri R, Somasekhar VT. An Integrated Control Algorithm for a Single-Stage PV Pumping System Using an Open-End Winding Induction Motor. IEEE Trans Ind Electron. 2016;63: 956–965.
- 8. Kumar R, Singh B. Single Stage Solar PV Fed Brushless DC Motor Driven Water Pump. IEEE J Emerg Sel Top Power Electron. 2017;5: 1377–1385.
- 9.
Mishra AK, Singh B. Solar photovoltaic array dependent dual output converter based water pumping using Switched Reluctance Motor drive. 2016 IEEE 6th International Conference on Power Systems, ICPS 2016. 2016. pp. 1–6.245.
- 10. Priyadarshi N, Bhaskar MS, Padmanaban S, Blaabjerg F, Azam F. New CUK-SEPIC converter based photovoltaic power system with hybrid GSA-PSO algorithm employing MPPT for water pumping applications. IET Power Electron. 2020;13: 2824–2830.
- 11. Priyadarshi N, Padmanaban S, Bhaskar MS, Blaabjerg F, Holm-Nielsen JB. An improved hybrid PV-wind power system with MPPT for water pumping applications. Int Trans Electr Energy Syst. 2020;30: e12210.
- 12. Priyadarshi N, Padmanaban S, Bhaskar MS, Azam F, Khan B, Hussien MG. A novel hybrid grey wolf optimized fuzzy logic control based photovoltaic water pumping system. IET Renew Power Gener. 2022;n/a.
- 13.
Priyadarshi N, Bhaskar MS, Modak P, Kumar N. A Novel Optimized Orthogonal Hybrid Firefly Based Ant Colony Method Occupying MPPT For Single Stage PV Water Pumping Applications. 2022 IEEE 2nd International Conference on Sustainable Energy and Future Electric Transportation, SeFeT 2022. 2022. pp. 1–5.
- 14.
Priyadarshi N, Bhaskar MS, Modak P, Kumar N. Hybrid Firefly-PSO MPPT Based Single Stage Induction Motor for PV Water Pumping with Deep Fuzzy-Neural Network Learning. 10th IEEE International Conference on Power Electronics, Drives and Energy Systems, PEDES 2022. 2022. pp. 1–5.
- 15.
Zhang J, Wang H, Wang C, Sun R. Variable universe fuzzy controlled incremental conductance MPPT for Z-source PV inverter. Atlantis Press; 2015. pp. 183–187.
- 16. Thangaprakash S. Unified MPPT control strategy for Z-source inverter based photovoltaic power conversion systems. J Power Electron. 2012;12: 172–180.
- 17. Sharma PG, Rangari S. Simulation of Inverter Fed Five Phase Induction Motor. International Journal of Science and Research. 2013. pp. 2319–7064. www.ijsr.net.
- 18.
Guzmán H, Durán MJ, Barrero F. A comprehensive fault analysis of a five-phase induction motor drive with an open phase. 15th International Power Electronics and Motion Control Conference and Exposition, EPE-PEMC 2012 ECCE Europe. 2012. p. LS5b.3-1-LS5b.3-6.
- 19.
Dattu NA, Rashmi MR. Modeling of five phase induction motor drive. Proceedings of 2017 IEEE International Conference on Technological Advancements in Power and Energy: Exploring Energy Solutions for an Intelligent Power Grid, TAP Energy 2017. 2018. pp. 1–5.
- 20. Diyoke GC, Eya CU. Modeling and Simulation of Five-Phase Induction Motor Fed With Pulse Width Modulated Five-Phase Multilevel Voltage Source Inverter Topologies. J Adv Electr Devices. 2019;4.
- 21. Kyslan K, Lacko M, Ferková Ž, Petro V, Padmanaban S, Perduková D. Current Limitation Method for V/f Control of Five-Phase Induction Machines. Errami Y, editor. Int Trans Electr Energy Syst. 2022;2022: 5165666.
- 22.
Roaf S, Gupta R. Solar Power: Using Energy from the Sun in Buildings. In: Elliott D, editor. Sustainable Energy. London: Palgrave Macmillan UK; 2007. pp. 84–107. https://doi.org/10.1057/9780230378384_5
- 23. De Soto W, Klein SA, Beckman WA. Improvement and validation of a model for photovoltaic array performance. Sol Energy. 2006;80: 78–88.
- 24. Ghoneim AA. Design optimization of photovoltaic powered water pumping systems. Energy Convers Manag. 2006;47: 1449–1463.
- 25. Winston DP, Kumar BP, Christabel SC, Chamkha AJ, Sathyamurthy R. Maximum power extraction in solar renewable power system—a bypass diode scanning approach. Comput Electr Eng. 2018;70: 122–136.
- 26. Elmetwaly AH, Younis RA, Abdelsalam AA, Omar AI, Mahmoud MM, Alsaif F, et al. Modeling, Simulation, and Experimental Validation of a Novel MPPT for Hybrid Renewable Sources Integrated with UPQC: An Application of Jellyfish Search Optimizer. Sustainability. 2023. p. 5209.
- 27. Said A, Abd-Allah MA, Mohsen M, I.omar A. Alleviation of the transients induced in large photovoltaic power plants by direct lightning stroke. Ain Shams Eng J. 2022;14: 101880.
- 28. Ramachandran N, Murugan S, Vinayagam C, Balachandran PK. Real-Time Implementation of a Seven-Level Multilevel DC Link Inverter for Solar PV System during Partial Shading. Electr Power Components Syst. 2023;51: 2125–2134.
- 29. Samanbakhsh R, Koohi P, Martin Ibanez F, Martin F, Terzija V. A Z-source inverter with switched network in the grid-connected applications. Int J Electr Power Energy Syst. 2023;147: 108819.
- 30. Bakar MS, Rahim NA, Daniyal H, Ghazali KH. Experimental study of SBPWM for Z-source inverter five phase. Int J Power Electron Drive Syst. 2015;6: 45–55.
- 31. DebBarman S, Roy T. Different Types of PWM Techniques analysis for Z-Source Inverter. IOSR J Electr Electron Eng. 2014;9: 09–17.
- 32. Aljafari B, Balachandran PK, Samithas D, Thanikanti SB. Solar photovoltaic converter controller using opposition-based reinforcement learning with butterfly optimization algorithm under partial shading conditions. Environ Sci Pollut Res. 2023;30: 72617–72640. pmid:37173605
- 33. Ward EE, Härer H. Preliminary investigation of an invertor-fed 5-phase induction motor. Proc Inst Electr Eng. 1969;116: 980.
- 34.
Diab AAZ, Al-Sayed A-HM, Abbas Mohammed HH, Mohammed YS. Development and Stabilization of Adaptive State Observers for Induction Machines. In: Diab AAZ, Al-Sayed A-HM, Abbas Mohammed HH, Mohammed YS, editors. Singapore: Springer Singapore; 2020. pp. 19–31. https://doi.org/10.1007/978-981-15-2298-7_3
- 35. Meenakshi Sundaram B, Manikandan B V., Praveen Kumar B, Prince Winston D. Combination of Novel Converter Topology and Improved MPPT Algorithm for Harnessing Maximum Power from Grid Connected Solar PV Systems. J Electr Eng Technol. 2019;14: 733–746.
- 36. Oliver JS, David PW, Balachandran PK, Mihet-Popa L. Analysis of Grid-Interactive PV-Fed BLDC Pump Using Optimized MPPT in DC–DC Converters. Sustain. 2022;14: 7205.
- 37. Hysa G, Meşe E. On the Behavior of Five-Phase Induction Motor Drive Under Normal and Faulty Conditions. Int J Innov Technol Interdiscip Sci wwwIJITIS.org. 2021;4: 754–763. Available:
- 38. Sankar RRS, Kumar JSV, Rao MG. Adaptive fuzzy PI current control of grid interact PV inverte. Int J Electr Comput Eng. 2018;8: 472–482.
- 39. Mondal SK, Pinto JOP, Bose BK. A neural-network-based space-vector PWM controller for a three-level voltage-fed inverter induction motor drive. IEEE Trans Ind Appl. 2002;38: 660–669.
- 40. Kumar R, Gupta RA, Bhangale S V., Gothwal H. Artificial neural network based direct torque control of induction motor drives. IET Seminar Digest. 2007. pp. 361–367.
- 41. Zemmit A, Messalti S, Harrag A. Innovative improved direct torque control of doubly fed induction machine (DFIM) using artificial neural network (ANN-DTC). Int J Appl Eng Res. 2016;11: 9099–9105.