-
INDIRECT VECTOR CONTROL OF INDUCTIONMOTOR WITH RANDOM LOADING
USING ANN
By M.DINESH Reddy (R.No.: 645403)Internal Guide: Sri.T. Reddy
ME, (Ph.D)
-
OVER VIEWIntroduction Vector Control Scalar Control Set
backs
Induction motor Modelling
Neural Controller Conventional Controller Set backs
Simulations
Results Conclusion Bibiliography
-
The simulation of Vector controlled Induction motor is done
indirectly by Matlab Simulink.
The Torque - Speed characteristics are similar to dc drive with
high starting torque and variable speed.
Vector control decouples Torque and Flux control of IM.
The Indirect Vector control scheme is preferred in the industry
because of its simplicity.
The IM is modeled using the d-q transformation as it adequately
simulates the transient performance.
The dynamic behavior can be made to match with that of
Separately excited dc motor.
INTRODUCTIONOverview
-
Scalar ControlControls only Magnitude of control variables
System gives Sluggish response due to the coupling effect.
Less effective
High Current transients
System easily prone to instability.
Overview
-
Vector ControlBoth the Magnitude and Phase alignment of vector
variables are controlled.
Decouples the Torque and Flux control of InductionMotor. Vector
control gives fast dynamic response without high current
transients.
Vector control operation leads to Low energy Consumption,Low
operating costs and High efficiency.
Induction motor can be controlled like a separately exited dc
motor.
Overview
-
Fig (a) Separately Excited DC motorFig (b) Vector- Controlled
Induction motor.Overview
-
Principle of Vector Control:
.Frequency as well as the phase are controlled indirectly with
the help of the unit vector.
Overview
-
Direct Vector Control:Direct relies on Flux feedback.
Complex to implement.
High cost.
Measurement is not accurate.Indirect Vector Control:Indirect
relies on Speed feedback
Easily to implement.
Low cost.
Accurate response.Overview
-
rotor pole is directed on de axis
we = wr + wslPhasor diagram of Indirect Vector control
:Overview
-
Field component Iqs should be aligned on the de axis &Torque
component of current Ids should be on the qe axis
The total finalized equations :
WhereSlip speedOverview
-
The voltage equations of induction motor are time varying and
complex in nature.A transformation is used which converts stator
and rotor variables of induction motor to a frame of reference that
rotates at an arbitrary angular velocity. fqd0s =Ks f abcsThe above
equation transforms 3 phase variables of stationary circuit
elements to arbitrary reference frame.
Three phase stationary reference frame (as-bs cs) variables into
two phase stationary reference) frame (ds qs)neglect the Zero
sequence component.
Modelling of the Induction MotorOverview
-
Complete Induction Machine Model:Overview
-
Synchronously rotating reference frame dynamic model
to represent both ds -qs circuits and dr -qr their variables in
a synchronously rotating frame de qe
Dynamic de -qe equivalent circuit of machine qe axis
circuit:Overview
-
Dynamic de - qe equivalent circuit of machine - de axis
circuit
Overview
-
Voltage equations :
Overview
-
Flux linkages equations:
qse = Lls iqse + Lm(iqse +iqre )
dse = Lls idse + Lm(idse +idre )
qre = Llr iqre + Lm(iqse +iqre )
dre = Llr idre + Lm(idse +idre )
qme = Lm(iqse +iqre )
dme = Lm(idse +idre )
Overview
-
Dynamic model state equations
dwr / dt = (P/2J)* (Te-TL)Overview
-
Conventional PI ControllerController: Modify the error signal
& achieve better control action.Modify the transient response
& the steady state error of the system.The aim of a PI
controller is to determine the stator voltage frequency that will
make the measured output (speed of the rotor) reach the reference.A
proportional (kp) term,which is equal to the product of the error
signal by a constant called the proportional gain.
The integral (Ki) term of the controller is used to eliminate
small steady errors.The I term calculates a continuous running
total of the error signal.Overview
-
PI Controller:Set backs:
High Peak OvershootProlong SettlingtimeOverview
-
Neural Proportional Integral Controller:The Neural PI controller
is same as the Conventional PI controller but the gain blocks are
realised using Neural network techniques.ANN: It is the most
generic form of AI for emulating the human thinking process.The
basic structure of a Neuron is shown below:Overview
-
MODEL OF A NEURON A Neuron is the fundamental building block of
nervous system that performs computational and communication
function.
Overview
-
Artificial Neural Network :-
It can be defined as a highly connected ensemble of processing
elements called neurons or nodes.An artificial neuron is a
multi-input, single output processing element consisting of a
summation operation and an activation function. Overview
-
Neural Network SubsystemLayer SubsystemOverview
-
Neural networks can perform massively parallel operations.Neural
networks exhibit fault tolerance since the information is
distributed in the connections throughout the network.Self learning
capabilityReal time operation.By using Neural PI controller the
Peak overshoot is reduced and the system reaches the steady state
quickly when compared to a conventional PI controller.
Advantages:Overview
-
Neural PI ControllerApplications:
Sales ForecastingIndustrial Process control Customer Research
Data Validation Risk Management Target Marketing
Overview
-
SIMULATION RESULTS:
Overview
-
PARAMETERS:3 PHASE Induction Motor RATING @ 50hp,
460V,4pole,50HzStator resistance [ohm]Stator leakage inductance
[H]Rotor resistance [ohm]Rotor leakage inductance [H]Magnetizing
inductance [H]Number of
poles0.0870.8e-30.2280.8e-334.7e-34Overview
-
#Case 1: No-LoadPI Controller: Speed(rad/s) Vs
Time(s)Overview
-
Peak Overshoot:5%Torque(N-m) Vs Time(s)Overview
-
Current (A) Vs Time(s)Overview
-
At No-Load:NN Controller: Speed(rad/s) Vs Time(s):Overview
-
Torque(N-m) Vs Time(s)Overview
-
Current (A) Vs Time(s) Overview
-
#Case 2: Step Change in -LoadPI Controller: Speed(rad/s)Vs
Time(s)Overview
-
Torque(N-m) Vs Time(s)Overview
-
Current (A) Vs Time(s)Overview
-
NN Controller: Speed(rad/s) Vs Time(s):Overview
-
Torque(N-m)Vs Time(s)Overview
-
Current (A) Vs Time(s)Overview
-
#Case 3 : Speed Reversal:PI Controller: Speed(rad/s) Vs
Time(s):Overview
-
Torque(N-m) Vs Time(s) Overview
-
Current (A) Vs Time(s)Overview
-
NN Controller: Speed(rad/s) Vs Time(s):
Overview
-
Torque(N-m)Vs Time(s)Overview
-
Current (A) Vs Time(s)Overview
-
ConclusionNeural Controllers are fast acting & more
accurate.
Avoids prolonged Settling time & High Peak overshoots.
Future Scope
Implementation of NEuro-Fuzzy CONtroller (NEFCON) for further
better performance. NEFCON combines the merits of Fuzzy systems and
Neural networks.
Overview
-
BIBILOGRAPHY
1.C.M.Liaw.Y.S.Kung and M.S.Ouyang Identification and control of
inductionmachines using artificial neural networks.IEEE Trans
Ind.Applicat. vol.31.pp.612-619,1995.2. C.M.Liaw.Y.S.Kung and
C.M.Wu Design and implementation of a high performance field
oriented Inductionmotor drive.IEEE Trans Ind.Applicat.
vol.38.pp.275-282,1991.3.M.A.Wishart and R.G.Harley Identification
and control of inductionmachines using artificial neural
networks.IEEE Trans Ind.Applicat. vol.34.pp.412-419,1994.4.Levin
and K.S.Narendra Control dynamics systems using neural networks
controllability and stabilization. IEEE Trans on Neural
networks.NN-1. 1,4-27,1990.5..Artificial neural networks B.Yagna
narayana6.An introduction to neural networks
JA.Anderson7.Electrical machines P.S.Bimbra8.Electrical machines
S.K.Bhatta charya9.Machine modelling Krause10.Electrical drives----
Vedam subramanyam11.Modern power electronics and ac
drives-----Bimal.K.BoseOverview
*The aim of a PI controller is to determine the stator voltage
frequency that will make the measured output (speed of the rotor)
reach the reference.PI stands for Proportional and Integral,two
terms which describe two distinct elements of the controller.A
proportional term,which is equal to the product of the error signal
by a constant called the proportional gain.The proportional term
mainly describes the short-term behavior of the controller since it
determines how the controller strongly reacts to reference
changes.The integral (I) term of the controller is used to
eliminate small steady errors.The I term calculates a continuous
running total of the error signal.Therefore,a small steady state
error accumulates into a large error value over time.this
accumulated error signal is multiplied by an I gain factor and
becomes the Ioutput of the PI controller
*The neural PI controller is same as the conventional PI
controller but the gain blocks are realised using neural network
techniques.The basic structure of a neuron is shown below:
Neural networks can perform massively parallel operatiops.Neural
networks exhibit fault tolerance since the information is
distributed in the connections throughout the network.By using
neural PI controller the peak overshoot is reduced and the system
reaches the steady state quickly when compared to a conventional PI
controller.
The advantages of neural network implementation of the speed
controller are as follows:
**ADVANTAGES OF NEURAL PI CONTROLLER*SPEED VS TIME PLOT USING PI
CONTROLLER*