Induction Motor - Science topic

Savitha Pr Follow these procedures to obtain data sets for training a neural network for sensorless speed estimate of an induction motor:

1. Create a test rig: Connect the induction motor and the AC drive as directed by the manufacturer.

2. Run the motor at various speeds: Record the RMS values of the AC drive terminal voltage and phase current, power, and torque while you run the motor at different speeds.

3. Carry out the ABC-Alpha Beta transformation: To conduct the ABC-Alpha Beta transformation in Simulink, use the exported RMS values of the AC drive terminal voltage and phase current.

4. Determine the rotor flux: Calculate the rotor flux using the alpha-beta components of voltage and current.

5. Keep track of the motor speed: Measure and record the motor speed on each run.

6. Repeat the above steps many times: Repeat the preceding procedures to expand the size of your data collection.

7. Preparing the data: Remove any noise, outliers, and discrepancies from the data.

8. Divide the data into two sets: training and validation: Divide the data into training and validation sets for use in training and assessing the neural network's performance.

If you are unable to determine the induction motor resistance, mutual inductance, and leakage inductance values, you may need to utilize additional approaches such as finite element analysis or physical testing.

Add a very high resistor (say 1e12 ohm) in parallel with load.

Ok sir thank you,,,,

My main aim is to get 90% efficiency and ,high torque and high power factor..... for squirrel cage induction motor....

I am using Ansys maxwell Rmxprt software is it sufficient to design?

If rotor of induction have only rotor bars and without laminated iron core, the generated flux in the rotor will be very low.

you may want to check this link

Induction Motor Stator Interturn Short Circuit Fault Detection in Accordance with Line Current Sequence Components Using Artificial Neural Network (hindawi.com)

I would like to share this one with you guys for discussion :

advantages of the fuzzy logic controller over the PID controller

1) Fuzzy logic is a Nonlinear system but PID is a liner system so the performance of fuzzy is better than PID

2)In an uncertainty system, nonlinear control is better than linear control 3)The output of the fuzzy logic controller can be controlled because of the fuzzy set

4)It is well known that fuzzy is more complex compared to PID and more computational but nowadays no issues with complexity. There are advanced microprocessor controllers such as DSpace 1104 and 1103. These controllers are fast with a large capacity of memory so no issues with computational of fuzzy. That is why simplified rules of fuzzy logic controllers are designed to reduce the computational burden of the fuzzy logic controller and the complexity of the system.

Dear Mohamed,

Bravo for your interest on this topic. I simply recommend you explore your analysis on this subject by taking diligent cursory look on the following:

1. What is torque

2. Field oriented machine

3. Investigate how the above 2 parameters apply & effect on the performance of induction motors.

Do above implicitly, and I assure you that you shall shortly surmount your mountain.

I'm available to work with you further on this. Reach me on olalowo@yahoo.co.uk when you are intrinsically sure you've done your research.

Best Regards,

Engr. Dr. Olalowo Samuel OLALEYE

Hi,

Here you have few papers with some of the datasets you are looking for. Sure, as papers, it is not giving the full sets of all the measurement and data, but you could find probably some interesting information.

- H. Henao, C. Demian, and G.-A. Capolino, “A frequency-domain

detection of stator winding faults in induction machines using an

external flux sensor,” IEEE Transactions on Industry Applications,

vol. 39, no. 5, pp. 1272–1279, 2003.

- J. de Pelegrin, U. J. Dreyer, J. P. Bazzo, and J. C. C. da Silva, “Faults

diagnosis in induction motors through thermal mapping produced by

the rdts system,” IEEE Sensors Journal, vol. 21, no. 18, pp. 20 061–

20 068, 2021.

- A. C. Muxiri, F. Bento, D. Fonseca, and A. J. M. Cardoso, “Thermal

analysis of an induction motor subjected to inter-turn short-circuit failures

in the stator windings,” in International Conference on Industrial

Engineering, Applications and Manufacturing (ICIEAM). IEEE, 2019,

pp. 1–5.

- J. Zhao, X. Guan, C. Li, Q. Mou, and Z. Chen, “Comprehensive

evaluation of inter-turn short circuit faults in pmsm used for electric

vehicles,” IEEE Transactions on Intelligent Transportation Systems,

vol. 22, no. 1, pp. 611–621, 2020.

Best regards

Amr Refky Abd Elwhab Thank you for the detailed elaboration. I might not express my self clealy. What I want to say is to represent each phase with their own equivalent circuit—i.e., per phase equivalent circuit A with A parameters to represent phase A, per phase equivalent circuit B with B parameters to represent phase B, per phase equivalent circuit C with C parameters to represent phase C. The input voltage is the corresponding stator phase voltage. That's like with one phase staor voltage and current, we can determine this phase corresponding parameters. Actually I verify the validity of this when the motor is under normal situation, but it doesn't perform well when it's under fault state—i.e, ITSC fault.

That's way I want to ask if it's okay to use equivalent circuit under asymmetric winding(one equivalent circuit for one phase with their own unique parameters).

After these day's research I had a conclusion but I'm not so sure. We can use equivalent circuit under asymmetric windings, but the input voltage should be the dostorted voltage under the configuration of 3 wires without neutral line, if it's 4 wires, the stator phase voltage is the same as supply voltage, which is what I use in simulation, and this might be the issue.

Mohamed-Mourad Lafifi Thanks, it's a detailed induction motor modeling paer, but it does't cover the phase voltage measuring. The input voltage in the paper won't be affected by winding asymmetry.

Jose Gregorio Elorreaga Ramos These links might help:

1. https://www.mathworks.com/help/physmod/sps/ref/simplifiedinductionmotor.html

2. https://www.mathworks.com/help/physmod/sps/powersys/ug/simulating-an-ac-motor-drive.html

Ray Wang Welcomе sir:

Most likely something is not correct regarding the compensation (feedforward terms) for the reference command

It might be noted that a BLDC is basically a PMSM just designed for trapezoidal back-emf which can be nicely used for low-cost, position-sensorless control, e.g., using hall effect sensors for generating switching pulses. In this regard, the difference between BLDC and PMSM is only minor and given the same amount of design effort their power and torque density can be pushed to nearly the same level. Against this background, the terminology ‘brushless DC motor’ is also largely incorrect from my personal perspective, since its current, voltage and flux waveforms are everything but DC – it is just a special three-phase AC motor variant.

However, another main difference is typically on the control and sensor level. Automotive PMSM drives typically incorporates positions and current sensors together with sophisticated control and functional safety to ensure proper drive function. These features are typically not part of a BLDC drive, however, there is not technical issue to integrate them.

Hi, sir, did you find the fault dataset?

Research on classical field-oriented control w.r.t. to linear feedback control is largely finished. Current research focus is on model predictive control (cf. https://www.wiley.com/en-gb/Model+Predictive+Control+of+High+Power+Converters+and+Industrial+Drives-p-9781119010906) and machine learning-based control (e.g., using reinforcement learning cf. https://ieeexplore.ieee.org/abstract/document/9376968). Both can (but not must) be integrated within a field-oriented coordinate system.

Personally, I do not see the point why to use fuzzy control for motor drives at all. Electrical drive systems can be considered to be fully deterministic (except measurement noise) while their dynamics are well understood and easy to model. Hence, there is no fuzziness / stochastic impact within a drive system which would justify fuzzy logic based control actions.

If you want to program it in Matlab, you may have a view in the Matlab Global Optimization Toolbox which actually covers the mentioned optimization algorithms.

You may also find the following papers interesting which actually performed the discussed task on a real-world drive system:

Were you able to resolve this?

Dear Prof. Messaoudi,

I am interested to work with you since Renewable energy is one of my main and favorite research field. I have also completed a research in past based on renewable energy sources and their effectiveness.

Bennadji Hocine

These articles might be an asset, have a look:

1. http://journals.iaumajlesi.ac.ir/ms/index/index.php/ms/article/view/393

2. https://ieeexplore.ieee.org/document/8613614

3. https://www.asjp.cerist.dz/en/article/67325

4.

5. https://pp.bme.hu/eecs/article/download/14245/8516

1. Instead of calculating transfer function or transfer function matrix better is to go with the state-space model........

With this type of modeling, then you can use the Kalman filter for identification purposes....

2. If you want to stick with the transfer function model...... it is better to first add a decoupling block to decouple the d-axis and q-axis circuits.....

Then depending on the type of control i.e FOC in your case.....you can easily relate rotor speed with q-xis voltage......for simplification you can make TL=0.....Also, as FOC is implemented so id=constat, meaning d-axis dynamics (or d-axis circuit dynamics) are totally ignored.....meaning as if there is no d-axis circuit.......but mind, it is valid only while decoupling block is used.......

Thus, you will finally have a plant with q-axis voltage as input and q-axis current as output with a cascade block whose input is q-axis current and output is torque developed with cascade block whose input is torque developed and output is rotor speed.....Finally, you can easily close two loops q-axis current and speed loops with two PI controllers to tune......here identification can easily be done by a Neural Network block......

Good luck and thank you so much

Dear Utkarsh Singh ,

In LV distribution networks, reactive power is not efficient in modifying electrical parameters due to X/R ratio of conductors. Although works discuss the effects on voltage in LV, IM tends to generate voltage problems like undervoltage.

Hi. Apply for PhD and continue your education. Then after that everything will be easier for you.

Hi , Zarzisur Rahman Rony

Yes, I have already worked on improving the power quality of three-phase inverters by comparing 3 methods (SPWM, SVPWM and SHE PWM).

the best method was the SHE PWM which eliminates harmonics and gives a very low THD

Concerning your question, in another work, I made a comparative study between the vector and sinusoidal PWM, I found almost the same value of distortion (THD) for different switching frequency, but in the SVPWM, the switching losses decreases which increases the value of the fondalmental (I made a simulation under MATLAB and I validated the results experimentally by the FPGA tool (RT-Lab).

Kind regards

It is not clearly mentioned the type of Invertor.

For domestic use DC to AC inverter DC link capacitor is nor neccessary but for inverter connecting to High AC voltage like Grid , DC to AC invertor needs DC link capacitor to reduce high frequency ripples at input side of invertor.

set the rotational reference frame, Vd ref = 0 and Vq ref = fixed value

the fixed value of the Vq ref depends on how much modulation index you want to apply for the inverter

for more information regarding the modelling of the IPMSM for FOC please read the following paper:

L. Zhong, M. F. Rahman, W. Y. Hu and K. W. Lim, "Analysis of direct torque control in permanent magnet synchronous motor drives," in IEEE Transactions on Power Electronics, vol. 12, no. 3, pp. 528-536, May 1997, doi: 10.1109/63.575680.

Thanks Anil for your suggestion. Yes it is very useful book. However, I am talking about a special machine type with special structure. The PMIM has a permanent-magnet (PM) rotor that is free to rotate inside a squirrel-cage rotor. The internal self-excitation from this PM rotor allows the PMIM to operate at a higher power factor and this is an advantage of this type. Unfortunately there is no plenty of research work about the control and analysis of this machine type

1 Assume we are dealing with transformed variables from

actual stationary 3 phase variables of machine viz synchronous or Induction motor to rotating ref frame ie d q and 0 sequence.

2. In the case of SM the rotating ref franme can be fixed to the daxis .Recall this d axis carries the field which is already excited via slip rings or pm so we dont need vd to excite a rotating field , we need only vq to produce the orthogonal torque component of flux.

3. In case of IM it is alittle more complex as the rotor carries no

current say at synch speed , the rotating flux has to be developed by both vd and vq (depending on ref frame fixed to stator flux or rotor flux both id and iq will have to be manipulated). In other words for an intuitive idea consider approx

a simple single phase IM circuit with just xm and r2/s . The stator current will consist of im and ir . ir is determined by both wsynch and slip at a particular mech wr . If we wish to simulate this as a dc motor in the rotating ref frame then we neeed to manipulate both vd and vq .

Cheers

Good day, where you able to get the answer for this?

IF you did, please kindly share

Modulation index ma is the ratio of DC-link voltage and output voltage amplitude.

Dear Mr. Widagdo Purbowaskito,

We have done a few experiments in past years on BLDC motor. Based on MCSA, 3^rd harmonic components of motor currents are analyzed to localize and isolate different faults. We did not do "fault thresholding" since it will be very specific for a system and it will differ based on the loads connected to the motor. However, for a constant load and operating environment, the magnitude of the 3^rd harmonic components increases as the fault severity rises.

To my knowledge, there is no specific study that establishes a "severity index" based on MCSA. I suggest you look for different frequency domain features i.e. harmonic magnitudes, root variance frequency, peak frequency, SNR, entropy, THD, etc. Not all features will be suitable for all systems, you can find the most suitable features based on your problem definition. I can suggest you the following papers and you may look into the references cited in these papers for similar studies.

(1) T. A. Shifat and J. W. Hur, "An Effective Stator Fault Diagnosis Framework of BLDC Motor Based on Vibration and Current Signals," in IEEE Access, vol. 8, pp. 106968-106981, 2020, doi: 10.1109/ACCESS.2020.3000856.

(2) T. A. Shifat and H. Jang-Wook, "Remaining Useful Life Estimation of BLDC Motor Considering Voltage Degradation and Attention-Based Neural Network," in IEEE Access, vol. 8, pp. 168414-168428, 2020, doi: 10.1109/ACCESS.2020.3023335.

Thanks all

please send me space state model of dfig

Dear Muhammad Mubashir Hussain

I found these attached documents helpful for induction motor simulation in presence of stator faults.

If You divide by square od rotor flux value please limit this to 0.01 from down.

Hello Krunal Shah ,

There are many ways to estimate the rotor speed. You can use one of the well-known observers like Extended Kalman filter, Extended Luenberger observer (I say the extended version of Kalman and Luenberger observer because the model of the Induction motor is nonlinear and rotor speed-dependent), High gain observer, sliding mode observer, model reference adaptive system (MRAS), fuzzy logic observer, neural network observer,...

In your situation, I recommend the use of MRAS because you have already estimated the motor current and flux

As I know, even though there are a lot of works, where the sensorless control works nicely in no-load condition, the control often goes wrong when the load torque is applied. Also, the control accuracy at low speed remains an actual issue.

Dear Michael,

I suggest you to see links and attached files on topic.

Best regards

in you tube

Dear Mohammed .H Khudhur Al-Huseiny ,

I'm not too sure about this statement, as power mechanically is the product of rotational speed and torque, electrically it contains voltage and current - no frequency.

Even at varying frequency, the power extracted can remain constant if the torque is reduced the same was as the speed (or frequency) increases.

Therefore, there is no inherent relation between power delivered by the motor and the frequency it is operated from.

A most basic example:

Running at no load, the output power for any given frequency is zero.

Cheers - Martin

I think that adaptive control is suitable for this case.

There are at least two ways to do it and they are not so simple.

The first one is to perform decomposition of the supply voltages into forward and backward symmetrical voltage sytem and make calculations for both of them.

You have to calculate input powers for both and output mechanical power also for both of them. The forward one is driving and the backward torque is breaking.

The other way is to comput the performance of the whole system by use of a dynamical mathematical model (including convereter if neccerary), wait for a steady-state (if it is going to be steady) and calculate input and output power.

In that case You can aso calculate power consumed by a converter.

Wish You success!

Piotr Wach, Opole, Poland

Dear Ahmed,

Also i suggest you to see links and attached file on subject.

Best regards

@Wail Ali Ali Saleh,

What comes to my mind is in this situation there could be two sources for the current THD. Either you have a some sort of switching device between your multi-level inverter and the induction motor or the motor core is saturated.

The former is less likely because a switching device would have produced high order harmonic distortions but in the FFT image that you posted the main contributor is the 3rd order harmonic, which narrows down my guess to the latter option: "saturated motor core".

Dear Menno,

Reluctance increases with the airgap while reactance decreases. Are you not confusing them?

Anyways, the magnetizing current of the motor depends on the specific magnetic loading and the air gap of the machine. A large air-gap leads to higher magnetizing current (and poor power factor). Indeed, the more you increase the air gap, the less "usefull magnetic interaction" you have between the rotor and the stator. Thus, to be able to start turning the same rotor (so same torque), you need to apply more current in the further away stator in order to compensate for the bad "magnetic interaction" between the stator and the rotor.

Now, to "accuretaly predict" but "without the use of programming" doesn't seem very compatible. Though, you can find quite easy-to-use software that are open-source (understand free of cost, if that is an issue):

Finite elements: http://www.femm.info/wiki/HomePage

Analytics: https://github.com/Eomys/pyleecan

And... many others

Best,

Guillaume

Hello! The experience I have is based on thermal circuits calculating the elevation of temperatures in transformers. Regarding the calculation of thermal resistances, it is necessary to consider the three types of heat exchange:

- Conduction: Rcd=l/(k*S), where l is the length of the path that the heat will travel [m], k is the thermal conductivity of the material [W/(m*K)] and S is the section cross section available for heat to flow [m²];

- Convection: Rcv=1/(h*S), where h is the convection heat transfer coefficient [W/(m²*K)] and S is the cross section available for heat to flow [m²];

- Radiation: Rrd=1/(hr*S), where hr is the radiation heat transfer coefficient [W/(m²*K)] which is temperature dependent and it makes the circuit solution iterative and Sr is the radiant surface [m²].

As for the question regarding the number of resistances required, it is a question of discretization, the more resistances, the more precise, but the greater the effort of discretization and computation. I consider it a good approach to use the smallest number that one can intuitively be deduced with the circuit nodes arranged at the points where it is intended to know the temperature rise.

Hi

Please check the following link. I believe it will help you.

https://www.mathworks.com/matlabcentral/answers/517583-tms320f28335-eqep-block-in-simulink

Regards

Abbas Miry

Thank you sir...

sir, is it get converged ? and give proper result?

Saptarshi Pal Chaudhuri i have the same problem. do you solve this problem?

can you help me about that

Saptarshi Pal Chaudhuri

Thank you for your answer, Dr. Rahul.

Use the below mentioned link for the said purpose.

You are welcome

Hi, torque and speed are related to the current and frequency supplied by the 3p inverter, respectively. There is a linear relationship between the d current and the torque of the machine for a given constant rotor flux operation in the linear region of operation (max torque region). Consequently, it is the change in the torque how is translated into a disturbance to the DC-link controller. You have two options: first, to model the whole system (inverter + motor) and derive the transfer function which relates the torque to the DC-link or, second, finds an empirical transfer function Gvdc/id. Thus, using this transfer function evaluates the closed-loop system stability which will depend on the controller you are using too.

Sir I have implemented field oriented control of six phase induction motor using hysteresis controller but now i am trying to use SVPWM for the same but i cant get accurate results Nikhil Krishna Bajjuri

Hello Dr. Ghada,

We are a company specialized in the design and manufacture of electric motors, electric generators and magnetic systems http://www.erneo.fr.

Based on your specific need, ERNEO is able to develop induction or reluctance machines with mutiphase winding (double star, polyphase, etc.).

For more information, you can contact us. Find below our contact information.

Best regards.

Dr. Nassim BEKKA Electromagnetic Designer ERNEO SAS 505b rue Louis Lépine 34000 MONTPELLIER FRANCE Mail : nassim.bekka@erneo.fr Tel : +33(0)4.11.28.17.21 Web : www.erneo.fr/

Problem has been solved. and it related to tuning of PI Controller. now its working fine in all four quadrants of operation. @ Kris Scicluna

Dear Lalit,

You can control the speed of the induction motor by changing the motor voltage while V while keeping V/F consatnt where F is the frequency of the voltage applied on the motor. This condition is required to keep the flux phi constant.

So, yes you can use DC/DC boost converter to control the voltage and then you can use DC/AC inverter to generate the AC for the motor. You have to change the frequency of the inverter by changing the switching speed of the inverter switches. You have to generate as sinusoidal waveform as possible.

This can be accomplished by using PWM and filtering.

Best wishes

If the supply frequency is varied the machine parameters , losses and efficiency will be varied.

You need system state space. You should put inport before system and outport after the system output. I think your system also discrete time. for such system you should use dlinmod.

https://www.mathworks.com/help/simulink/slref/dlinmod.html

Use VFD (variable frequency drive) starter

It depends on your application and what you mean by "electric vehicles".

Normally for a city vehicle at least a 100 kW electric motor is needed.

As you know, electrical active power is (1.73) VL IL Cos(phi) . If you consider the cos(phi) =1 and a Delta connection and 48 volts system, so:

IL=100kW / (48) (1.73) = 1200A for each phase. Which is really a great amount of currents, and needs a great deal of copper to allow it to pass in order to minimize the ohmic loss and heat.

Instead, if you use 400 volts system :

IL=100kW / (400) (1.73) = 140 A

It is completely tolerable.

Less copper, lower weight, lower price, less heat

Were you able to simulate the inter turn short circuit fault using the suggested models? Did u use Simulink for simulations?

@ O. Peter, following

Hi I'm Professor Soheil Sayyed Hosseini To solve the problem see my article on this issue of my control, motor which is published by the University of Rochester. If it is acceptable, say I wish you well.

If you study the differences in phase splitter and phase inverters, the idea belongs to audio amplifier (