Field Oriented Control For Three Phase Induction Motor Based On Full Neural Estimator And Controller

Closed loop speed control for an I.M is somewhat complex strategy, the complexity is gradually increases according to the demand performance degree. There are many types of control strategies: scalar, direct torque, adaptive, sensor less, and vector or Field Oriented Control (FOC). This paper proposes the FOC strategy in details. Rotor flux, unit vector, and electromagnetic torque estimation are considered by using Digital Signal Processing (DSP). Artificial Neural Network (ANN) becomes a powerful tool for control nonlinear system in present time. This study proposes the using of ANN in stead of DSP to estimate the flux, unit vector


Introduction:
Open loop Variable Voltage Variable Frequency (VVVF) control of an induction motor is by far the most popular method, because of its simplicity, and these types of motors are widely used in industrial applications.
Traditionally, the Volt/Hz is controlled by means of D.C-A.C pulse width modulation inverter in order to maintain the airgap flux approximately at its rated value, this type of control called "scalar control".Unfortunately, scalar control is somewhat simple to implement, but the inherent coupling effect for both torque and flux are functions of voltage or current and frequency (i.e. during volt/Hz variation both torque and flux be varied too), which gives sluggish response and the system is easily prone to instability because of high order system effect [1].
This problem can be solved by vector or field-oriented control, which invented at 1970s, and the demonstration that; an induction motor can be controlled likes a separately excited D.C. motor, brought a renaissance in the high performance control of A.C. drives, vector control is also known as decoupling, orthogonal, or transvector control.The principle of this method is to rotate the direct axis always in the same direction of the rotor pole axis, or by aligned the rotor flux vector on the d-axis.This can be done by estimating the exact unit vector (cos and sin ).In this way it's essential to estimate the flux vector and the unit vector by using complex digital signal processing (DSP) and microcomputer analysis [1,2].Nowadays, the Artificial Intelligent System (AIS) has penetrated deeply into electrical engineering field and their applications in power electronics and motion control appears very promising [1,3].The artificial neural system is one of the most powerful intelligent system with improved design and performance features, which can be implemented in this paper to estimate the flux vector , the unit vector (cos and sin ), and the electromagnetic torque T e instead of using the complex microprocessor DSP analysis.

Brief Review:
Before entering deeply in fieldoriented control, some principles of A.C. drives and system stags must be briefly reviewed to understand system operation.The overall system is consisting of three main stages as shown in figure (1): • Voltage-fed Sinusoidal Pulse Width Modulation (SPWM) inverter.• Three phase induction motor.
• Controller.The last one will be discussed in details later.

Constant Volt/Hz Control:
The characteristics of 3-phase induction motor have different forms as the variation of input voltage and frequency.
The operation characteristics can be divided into three regions [1,4] The pulse width modulation techniques have wide simple and complex types, the Sinusoidal Pulse Width Modulation (SPWM) is the most popular used method of A.C. drives, but it's not the efficient method [1,5].Voltage source inverter (VSI) should have a stiff source at the input [1], that is, its Thevenin impedance ideally is zero.Thus, a large capacitor can be connected at the input if the voltage source is not stiff.A practical (VSI) consist of power bridge devices with three output legs, each consist of two power switches and two freewheeling diodes, the inverter is supplied from D.C. voltage source via LC or C filter.In sinusoidal PWM the three output legs considered as three independent push-pull amplifiers [5]   V an = is the output phase voltage measured to isolated neutral of threephase load such as induction motor. Where: Where: = fundamental frequency; c =carrier frequency; M and N are integers and M+N =odd, m=modulation index is defined as [1,5]: Where: V P = peak of modulating signal, V T = peak of triangle signal.At m= 1, the maximum value of fundamental peak = 0.5V d which is 78.54% of the peak fundamental voltage of the square-wave (2V d / ) which called the linear modulation region.To further increase the amplitude of the output voltage, the amplitude of the modulating signals exceeds the amplitude of the carrier signal which leads to enter into quasi-PWM region called "over modulating region" causing increase in the low order harmonics.Further increasing modulation index tends to obtain PDF created with pdfFactory Pro trial version www.pdffactory.com

Field Oriented Control for Three Phase Induction Motor Based On Full Neural
Estimator and Controller square wave at maximum possible output fundamental (2V d / ) [1,5].

SPWM Inverter Simulation:
By using MATLAB/SIMULINK PROGRAM, the SPWM inverter can be simulated, firstly generation of the carrier triangle signal and the three modulating signals by using internal timer and the rated frequency (50 H Z ) to obtain the angular speed ( e t), then multiplying the angular speed and the amplitude of the signal by the frequency command (f * com ) and voltage command (V * com ) respectively.Secondly, compared the two signal sets to generate the switching signals of three switches used as three push-pull devices.The output of the switches gives (V ao , V bo , V co ) then the three phase to load neutral (V an , V bn , V cn ) can be achieved by implementing equation (1). Figure (3) illustrated the complete simulation of SPWM inverter, and the output phase voltage can be shown in figure (4).

Modeling and Simulation of Three Phase I.M.:
The mathematical representation of an induction motor can be looked on as transformer with moving secondary winding, where the coupling coefficients between the stator and rotor phases change continuously with the change of rotor position [1,5].The machine model can be described by differential equation with time varying mutual inductances, but such model tends to be very complex.Therefore, axis transformation is applied to transfer the three phase parameters (voltage, current and flux) to two-axis frame called (dq-axis stationary frame or park transformation).Park transformation is applied to refer the stator variables to a synchronously rotating reference frame fixed in the rotor, by such transformation the stator and rotor parameters rotate in synchronous speed and all simulated variables in the stationary frame appear as d.c.quantities in the synchronously rotating reference frame [1,5].
The per-phase equivalent circuit diagrams of an I.M. in two-axis synchronously rotating reference frame are illustrated in figure (5).From the circuit diagram the following equations can be written [1]:

A.C Drive Control Strategies:
To perform the desired load task, a close loop control system must be implemented.There are many types of controller, were built and employed according to specific strategies and the desired performance level.The most popular used strategies are [1,2]:

Scalar Control:
It's the simplest control method, easy to implement, and have been widely used in industry.As the name indicates, is due to the magnitude variation of the control variables only, and disregards the coupling effect of the machine.

Direct Torque Control:
It's complex than scalar control, in which the errors between the reference and the estimated values of torque and flux are directly control

Field Oriented Control for Three Phase Induction Motor Based On Full Neural
Estimator and Controller the inverter states in order to reduce the torque and flux errors.

Adaptive Control:
It is more complex than previous types, which deal with variable mechanical parameters load nature.Such controller requires adaptation or tuning of controller parameters in real time, depending on the plant parameter variation and load torque disturbance.

Sensors Less Control:
Ongoing researches have concentrating on the eliminating of the speed sensor at the machine shaft without deteriorating the dynamic performance of the drive control system.Speed estimation is an issue of particular interest with such controllers.

Vector or Field-Oriented Control:
Vector or Field-Oriented Control (FOC), allows a squirrel-cage induction motor to be driven with high dynamic performance.It transforms the dynamic structure of the A.C motor into that of separately excited D.C motor [1,2].For D.C motor, the field flux is proportional to the field current, if the field assumed to be constant and independent of armature current, the armature current provides direct control torque, so that:

….(20)
With the induction motor transformed to d-q plane, it looks like a separately excited D.C motor.The (FOC) technique decouples the two components of stator current; one providing the air-gap flux, and the other producing the torque.These current components provide independent control of flux and torque and the characteristic is linear [1,2].These components are transferred back to the stator frame before feeding back to the rotor.The two components are d-axis analogues to field current , and qaxis is analogues to armature current of the separately excited D.C motor [1,2].This strategy can be implemented by align the rotor flux vector along the d-axis of the stationary frame as shown by the phasor diagram in figure (7).The fundamentals of vector control implementation can be explained in figure (8), where the motor model is presented in a synchronously rotating reference frame, the voltage-fed SPWM inverter produces three-phase voltages ( ) according to the reference command voltages ( ) the flux component of stator current and the torque component of the stator current is used as a control signals to the system, which are inversely transformed to three-phase reference currents ( ), and then transferred to three-phase voltages ( ) through (PI) controller [1,2].The vector control can be implemented by either direct or indirect method, these methods are different essentially by how the unit vector (cos and sin ) is estimated for the controller.

Flux Vector Estimation:
There are two commonly methods of flux estimation; voltage model, and current model.The first one has strong performance in high speed regions but not in low speeds.Whereas, the second method has an accepted performance in both low and high speeds [1].
PDF created with pdfFactory Pro trial version www.pdffactory.com

Field Oriented Control for Three Phase Induction Motor Based On Full Neural Estimator and Controller
The current model depends on the main rotor equations in the two axes stationary frame d s -q s , superscript "s" referred to stationary frame quantities: PDF created with pdfFactory Pro trial version www.pdffactory.com

Field Oriented Control for Three Phase Induction Motor Based On Full Neural
Estimator and Controller

Artificial Neural Networks:
Nowadays, we have seen extensive researches and developments effort to use the intelligent system in many industrial applications, because of its strong features like: learning ability, massive parallelism, fast adaptation, inherent approximation capability, and high degree of tolerance [6,7].Neural Network Controller (NNC) was effectively introduced to improve the performance of nonlinear system, which are a powerful tools used to predict the optimum performance for both identification and control system.The universal approximation capabilities of multi-layer perceptron make it popular choice for modeling nonlinear systems and for implementing general-purpose controllers.The neural network is an intelligent system which could be learned or trained using an actual existing input and output table, learning or training can be achieved in real-time or off-time operation [1,7].
Thus, these features can be employed in this work to estimate the amount of the rotor flux vector and the unit vector (cos and sin ) instead of using DSP microprocessor, which reduce the complexity of the hardware, and reduce the Electro Magnetic Interference EMI in the microprocessor performance.Also, a neural-based PI controller can be implemented instead of hardware PI controller.

Design of Neural Network
Estimator:

( 2 )
illustrates the principles of SPWM; gating signals, phase voltage.The output phase voltage: V ao = is the output phase voltage measured to the center of the input D.C. voltage.

Field
superscript notation "e" referred to the synchronously rotating reference frame quantities.It's obviously that in squirrel cage I.M V qdr =0, then the pervious equation can be rewritten:PDF created with pdfFactory Pro trial version www.pdffactory.comgap flux and rotor current can be found as:T e =(3/2)(P/2) …..(12)By resolving the variables into d e -q e components:Where: r = is the rotor speed; P: no. of poles; J= rotor inertia; T L = load torque.The stator current can be found by:From the previous equations the dynamic model of an induction motor is simulated as shown in figure(6).
PDF created with pdfFactory Pro trial version www.pdffactory.com

5 .
Adding terms (L m R r /L r ) and (L m R r /L r )and simplifying, get:Where: Tr = is the rotor time constant.Equations(23 & 24)  give rotor fluxes as a function of stator currents and speed.Therefore, knowing these signals, the rotor flux and corresponding unit vector (cos and sin ) can be estimated by means of DSP microprocessor to implement the following equations[1,2]: the current model requires a speed encoder, but the advantage is that the drive operation can be extended down to low and zero speed.It's important to mention, that the input signals to the estimator ( and ) must be filtered by a low pass filter stage.The simulation of the current model estimator can be shown in figure (9).DSP Field-Oriented Control Simulation: As mentioned before, equations (25) implemented by means of DSP microprocessor to estimate the rotor flux and the unit vector (cos and sin ).Then two reference commands are obtained; first the torque command , and second the flux command , which are D.C values in synchronously rotating frame [1, 2].These two setting commands are compared with the actual currents and , the error signals of the two components are converted to voltage reference signals and by means of two PI controllers, which are converted to three phase reference voltages ( , , ) to drive the SPWM inverter at desired voltage and frequency.The flux command obtained from PI flux control loop for precision control of flux, and the torque command generated from the speed control loop.The overall close loop simulation of DSP-estimator with PIcontroller field oriented speed control is illustrated in figure (10).The flux command collected with the speed command the lockup table to program the flux during field weakening regions.The operation performance of the motor under different speed steps is shown in figure (11).

5 .
The decoupling of the stator current components is exactly aligned in Neural-estimator strategy.The torque component ( ) is exactly analogues the electromagnetic torque (T e ) as shown in figure (19), and the flux component current ( ) is constant and exactly analogues the air-gap flux as shown in figure (20).

com Field Oriented Control for Three Phase Induction Motor Based On Full Neural Estimator and Controller
PDF created with pdfFactory Pro trial version www.pdffactory.

Field Oriented Control for Three Phase Induction Motor Based On Full Neural Estimator and Controller
PDF created with pdfFactory Pro trial version www.pdffactory.com