ABSTRACT

Since its development, Permanent Magnet Synchronous Machines (PMSM) has been used in a number of specific applications. However, in recent years PMSMs have become more attractive due to developments in new materials for permanent magnets and in semiconductor technology for converter design. Recent investigations have proposed some implementations applying the Direct Torque Control (DTC) technique to PMSM motor drives, offering a fast and accurate control. This paper presents a modified DTC scheme, using fuzzy logic with Pulse Wide Modulation (PWM) to improve stator flux and the electric torque by significantly reducing their ripple. The proposed method effectiveness has been verified by computer simulations and experimental tests on a laboratory prototype. These results are compared with the ones obtained with a modified DTC using a PI controlled PWM with current limit.

INTRODUCTION

During the last decade, the use of PMSMs has been steadily growing in industrial applications, replacing DC and induction machines. The principal advantages of these machines are their low inertia and high efficiency, power density and reliability. Additionally, PMSMs are ideal for applications where fast and accurate torque control is required. The use of space vectors and field oriented transformations has been thoroughly developed for applications requiring fast dynamical response. The main limitations of these techniques are due to the transformation dependence upon machine model parameters and the need for a rotor position sensor, increasing the system cost. The use of DTC in PMSMs was proposed in the late nineties; since then there have not been major contributions for improving this control technique. DTC improves the machine controller performance and reduces the influence of parameter variation during its operation, and uses an inverter bridge switching table that sets the converter output depending on the flux/torque errors and flux angle. The bang-bang behavior produced by the limited number of states available in the inverter bridge (only seven different states)s produces a prominent electrical ripple torque. To solve this problem, in this paper a space vector PWM-DTC with fuzzy control is proposed to produce an effect equivalent to the inclusion of additional states in the inverter bridge. This is achieved by altering the resulting amplitude of the stator voltage. The results of the proposed fuzzy-DTC method are compared with the ones obtained using a standard DTC with on-off current limitation, and with a modified DTC using a PI controlled PWM with current limit. s Control technology of ac motor drives has improved dramatically during the last two decades. This trend owes its progress to the new control techniques and philosophies developed by several researchers around the world. The implementation of those concepts was possible due to technological developments such as the DSP controllers and the new power semiconductors. Variable speed or adjustable torque control of electrical motor drives are crucial components in almost all modern industrial manufacturing processes. Traditionally variable speed electric machines were based on dc motors, but, for the last 20 years, the inverter-fed ac motor has largely taken over as the preferred solution for variable speed applications.

For low performance applications, open loop voltage/frequency control strategies are employed. Considering high-performance motion control, Field Oriented Control (FOC), or more recently Direct Torque Control (DTC) are used. Using these techniques, both the induction motor and the permanent magnet synchronous motor (PMSM) can be applied even in high performance servo applications that were once the exclusive domain of the dc machine. Both FOC and DTC are strategies that allow torque and flux to be decoupled and controlled independently. Compared to the dc machine, the ac motor drive control strategy effectively\ becomes similar. In the dc machine, this decoupling is obtained in an electromagnetic way by orienting the current with respect to the stator flux using a commutator. In ac machines, this decoupling is obtained by implementing mathematical transformations, thereby avoiding problems due to the commutator. The mentioned control techniques have undergone considerable research over the last 15 years, but several problems remain: FOC is very dependent on knowledge of the rotor time constant when using an induction machine. DTC, in its traditional form, results in a non constant inverter switching frequency, which may result in high inverter/motor losses. Techniques that fix this problem result in parameter dependent solutions. This chapter provides a brief overview of the basic operating principles using DTC and FOC. Furthermore, a comparative study of FOC and DTC is presented. As shown, both control approaches have theirown advantages and drawbacks, respectively.

1.2 Electrical Motor Applications

Electric motors are the workhorse of the manufacturing industry and they also play a key-role in the transportation industry. Variable speed drives are used in all industries to control precisely the speed of electric motors driving loads ranging from pumps and fans to complex drives on paper machines, rolling mills, cranes and similar drives. In order to understand the requirements for adjustable speed drives, an overview about the fields of electrical motor applications is given. Of the total number of electrical machine applications, less than 10 % require an adjustable speed drive . Due to the increased demand for flexibility in manufacturing and due to the urge for rational use of electrical energy, the percentage is rapidly growing

 Constant torque: The requirement is to maintain torque constant independently of the speed. This is used in conveyors, mixers, screw feeders, extruders and positive displacement pumps for example, accounting for the majority of the electrical machine applications. These applications have high friction loads with little or no inertia, requiring starting torque much higher than rated.  Variable torque (continuously variable load): Low torque is needed at low speeds and high torque at high speeds. This is used for centrifugal loads such as fans, pumps and blowers, being second place in number of applications. Speed can be controlled to adjust air or liquid flow.  Constant power: High torque is needed at low speeds and low torque at high speeds.This is used in the machine tool industry, e.g. cutters and lathes, where usually dc motors are used. Motor applications for electric vehicles could also be classified under this category. Crucial factors, affecting the choice of an adjustable speed drive, include among others: rating, cost, speed range, efficiency, speed regulation, braking requirements, reliability, power factor, power supply availability, environmental consideration. The basic applications of electrical motors show that motion control of anelectrical machine can be focused to one or more of the following object. In the past, dc motors were used extensively in applications where high performance variable speed operation and controlled torque were required. In a separatelyexcited dc motor, flux and torque can be easily controlled by the field and armature currents. The main drawback of the dc motor is the use of a commutator and brushes. This leads to constant maintenance, limited use in explosive environments and limited use in high-speed and high-voltage applications. Alternating current motors present several advantages over their dc counterparts: they are more robust and smaller for the same power rating. This means lower rotating mass that can be translated into higher operating speeds and faster acceleration. Among the ac machines, the squirrel-cage induction motor has additional advantages: it is simple and rugged and is one of the cheapest machines available. The induction motor advantages also include the lack of electrical connections that can generate sparks or dust. Thus, they are suitable for explosive environments, clean-room areas and for low maintenance applications. Today, the induction motor is even more dominating the drive market. However, a majorimpetus to the use of the permanent magnet synchronous motor (PMSM) has been caused by the introduction of neodymium-iron-boron (NdFeB) materials in 1983.

Magnets are used to produce the magnetic field rather than employing a magnetizing component of stator current as in the induction motor. Furthermore, there is no equivalent of the loss in the rotor bars of an induction motor. Significant advantages arise from the simplification in construction, the reduction of losses and the efficiency improvement. PMSM with high fluxdensity magnets, as NdFeB, are particularly suited for high-performance drives with high acceleration required in applications such as robots and machine tools. Currently, induction motors and PMSM are being used extensively in applications requiring high-dynamic and accurate control of both speed and position. Apart from induction motor drives and PMSM, there is a wide variety of different electrical ac motors available for use with variable speed drives, e.g. brushless dc motor, switched reluctance motor, synchronous reluctance motor. No single motor type is ideal for all applications. However, the former are usually used for highperformance drive applications and they form the motor types considered extensively within literature. In the last three decades, important advances in the power semiconductor and control technology areas have led to new adjustable speed-drives for ac motors. Unfortunately, ac motor mathematical models are much more complex than those of the dc motor and thus require more complex control schemes and more expensive power converters to achieve speed and torque control.

Using advanced control approaches like field-oriented control (FOC) or direct torque control (DTC), a dynamic performance at least equivalent to that of a commutator motor can be achieved. As shown in the next paragraphs, illustrating the basics of torque control with FOC and DTC, the torque control of induction motor and PMSM are similar. Usually, a speed control loop, determining the torque reference, is outside of the torque control loop providing a variable motor speed drive.

1.3. AC Motors and Their Trends Unlike DC brush motors, AC motors such as Permanent Magnet AC motors (PMSM, and BLDC motors), and Induction Motors (IM) are more rugged meaning that they have lower weight and inertia than DC motors. The main advantage of AC motors over DC motors is that they do not require an electrical connection between the stationary and rotating parts of the motor. Therefore, they do not need any

mechanical commutator and brush, leading to the fact that they are maintenance free motors. They also have higher efficiency than DC motors and a high overload capability. All of the advantages listed above label AC motors as being more robust, quite cheaper, and less prone to failure at high speeds. Furthermore, they can work in explosive or corrosive environments because they dont produce sparks. All the advantages outlined above show that AC motors are the perfect choice for electrical to mechanical conversion. Usually mechanical energy is required not at a constant speeds but variable speeds. Variable speed control for AC drives is not a trivial matter. The only way of producing variable speeds AC drives is by supplying the motor with a variable amplitude and frequency three phase source. Variable frequency changes the motor speed because the rotor speed depends on the speed of the stator magnetic field which rotates at the same frequency of the applied voltage. For example, the higher the frequency of the applied voltage the higher the speed. A variable voltage is required because as the motor impedance reduces at low frequencies the current has to be limited by means of reducing the supply voltage. Before the days of power electronics and advanced control techniques, such as vector control and direct torque control AC motors have traditionally been unsuitable for variable speed applications. This is due to the torque and flux within the motor being coupled, which means that any change in one will affect the other. In the early times, very limited speed control of induction motors was achieved by switching the three-stator windings from delta connection to star connection, allowing the voltage at the motor windings to be reduced. If a motor has more than three stator windings, then pole changing is possible, but only allows for certain discrete speeds. Moreover, a motor with several stator windings is more expensive than a conventional three phase motor. This speed control method is costly and inefficient. Another alternative way of speed control is achieved by using wound rotor induction motor, where the rotor winding ends are connected to slip rings. This type of motor however, negates the natural advantages of conventional induction motors and it also introduces additional losses by connecting some impedance in series with the stator windings of the induction motor. This results in very poor performance. The goal was to achieve an adjustable speed drive with good speed characteristics compared to the
DC brush motor. Even after discovering of the AC asynchronous motor, also named induction motor, in 1883 by Tesla, more than six decades later of invention of DC brush motors, capability of adjustable speed drives for induction motors is not as easy as DC brush motors.

Speed control for DC motors is easy to achieve. The speed is controlled by applied voltage; e.g. the higher the voltage the higher the speed. Torque is controlled by armature current; e.g. the higher the current the higher the torque. In addition, DC brush motor drives are not only permitted four quadrant operations but also provided with wide power ranges. Recent advances in the development of fast semiconductor switches and cost-effective DSPs and micro-processors have opened a new era for the adjustable speed drive. These developments have helped the field of motor drives by shifting complicated hardware control structures onto software based advanced control algorithms. The result is a considerable improvement in cost while providing better performance of the overall drive system. The emergence of effective control techniques such as vector and direct torque control, via DSPs and microprocessors allow independent control of torque and flux in an AC motor, resulting in achievement of linear torque characteristics resembling those of DC motors.

1 DIRECT TORQUE CONTROL Direct Torque and Flux Control (DTFC), also termed Direct Torque Control(DTC), has been developed by German and Japanese researchers for use in torque control of high power servo drives. The publication of the DTC theory actually goes back to 1971. Recently, it has provided an industrial alternative to the field oriented control strategy. DTC is a control philosophy exploiting the torque and flux producing capabilities of ac machines when fed by a simple voltage source inverter that does not require current regulation loops, still attaining similar performance to that obtained from a vector control drive [Buja 98]. Three control techniques have been employed for implementing DTFC drives: The Switching Table (ST), the Direct Self Control (DSC) and the Direct Vector Modulation Control (DVMC) [Buja 98]. ST and DSC regulators are of the hysteresis (discrete) type, whereas the DVMC uses digitally implemented analog controllers such as the PI controllers. The basic DTC scheme for ac motor drives is shown in figure 1.1. The main blocks are the stator flux and electromagnetic torque estimator and the switching vector look-up table.

For implementing the control loop, the actual stator flux (amplitude and orientation) and electromagnetic torque are calculated by an estimator from the stator voltages and currents in a similar fashion to the reconstruction approach of the direct vector control philosophies (DFOC and SFOC). The stator flux is an integral of the stator EMF:

Thus, the flux magnitude strongly depends on the stator voltage. As the stator voltage changes, the stator flux follows rapidly whereas the rotor flux (rotor current) changes are slower and less pronounced than that of the stator flux. This effect modifies the angle between stator and rotor fluxes and consequently the torque increases or decreases. Thus, stator flux and developed torque can be directly controlled by proper selection of the stator voltage, that is selection of consecutive inverter states without coordinate transformations like the ones performed by vectorcontrol strategies. The developed torque is obtained by the product of stator current and flux.

In most DTC control schemes, the torque is compared in a threelevel hysteresis controller defining the error torque state LT in modulus and sign. The stator flux is compared to its reference value and is fed into a two-level hysteresis controller defining the error flux state L. With this information, a voltage selector determines the stator voltage that is required to increase or decrease the variables (torque or flux) according to the demands. The affiliated look-up table and switching states are summarized in table 1.1. The used switching states are sector dependent and applied according to the error states LT and Lgiven by the flux and torque comparators. Note, that the voltage sector definition of DTC, indicated by the dashed lines in figure 1.1, is shifted by 30rclockwise when compared to the sector definition of the later described space vector modulation. More advanced look-up tables, overcoming DTC related problems as demagnetization at low motor speed, are described in literature, e.g. [Dam 97].

The classical DTC scheme is shown in figure 1. The torque and flux estimator uses the DC converters voltage and currents, measured directly on the machine terminals.

The flux space vector can be adjusted with the proper selection of one of the seven stator voltage vectors produced by the inverter bridge. The stator flux can be adjusted by increasing or decreasing its magnitude, and rotated clockwise or anticlockwise to obtain the required torque. The seven different voltage space vectors are function of the bridge connectivity, as described in figure 2. Stator flux is obtained integrating the electromotive force in the stator windings using the following relation,

In general, using a conservative transformation for the active power, space vectors are defined as:

The electric torque is calculated using the estimated flux space vectorand the stator current space vector obtained from measurements:

The classic DTC technique selects as the new voltage vector the one that maximizes the flux and torque error correction, and minimizes the number of commutations required to change from the old to the new inverter output. Each vector will produce a different change depending on the angular region where stator flux is located. Table 1 shows the DTC switch selection algorithm.

The torque of the permanent magnet synchronous motor is controlled by inspecting the armature current since electromagnetic torque is proportional to the armature current. For high dynamic performance, the current control is applied on rotor flux (dq) reference system that is rotated at synchronous speed. In this system, if the change of the back electromotor force (emf) and theinductance are sinusoidal, armature circuit inductance and magnet magnetic flux are constant. The main principle of DTC is to select the appropriate voltage vectors according to the stator magnetic flux, difference between the reference and real torque. The current control circuit that

is constituted with the pulse width modulation (PWM) comparator circuit is not used in DTC. Therefore, if the DTC method is compared to PWM current control, it\ yields advantages such as; less parameter dependence and fast torque response. If the initial position of the rotor is known, it is possible to work with DTC without sensors.

2. What is a PMSM motor?

If we take a PMSM motor, and spin it as a generator, that is, if we spin it with another motor, or even with our hands, these are the measurements we will get from the motor windings. These voltages are sinusoidal in a PMSM motor, whereas in a BLDC motor they are trapezoidal.

Here are some alternate names we find for Permanent Magnet Synchronous motors: 1. Brushless Motors with Sinusoidal Back EMF 2. Synchronous AC Motors 3. BLAC, or Brushless AC Motors 4. and PMSM, Permanent Magnet Synchronous Motors
2.1. Permanent Magnet Synchronous Motors (PMSMs) Recent availability of high energy-density permanent magnet (PM) materials at competitive prices, continuing breakthroughs and reduction in cost of powerful fast digital signal processors (DSPs) and micro-controllers combined with the remarkable advances in semiconductor switches and modern control technologies have opened up new possibilities for permanent magnet brushless motor drives in order to meet competitive worldwide market demands [1]. The popularity of PMSMs comes from their desirable features [4]: High efficiency High torque to inertia ratio High torque to inertia ratio High torque to volume ratio High air gap flux density High power factor High acceleration and deceleration rates Lower maintenance cost

 Simplicity and ruggedness Compact structure Linear response in the effective input

However, the higher initial cost, operating temperature limitations, and danger of demagnetization mainly due to the presence of permanent magnets can be restrictive for some applications. In permanent magnet (PM) synchronous motors, permanent magnets are mounted inside or outside of the rotor. Unlike DC brush motors, every brushless DC (so called BLDC) and permanent magnet synchronous motor requires a drive to supply commutated current. This is obtained by pulse width modulation of the DC bus using a DC-to-AC inverter attached to the motor windings. The windings must be synchronized with the rotor position by using position sensors or through sensorless position estimation techniques. By energizing specific windings in the stator, based on the position of the rotor, a rotating magnetic field is generated. In permanent magnet ac motors with sinusoidal current excitation (so called PMSM), all the phases of the stator windings carry current at any instant, but in permanent magnet AC motors with quasisquare wave current excitation (BLDC), which will be discussed in more detail later, only two of the three stator windings are energized in each commutation sequence [1]. In both motors, currents are switched in a predetermined sequence and hence the permanent magnets that provide a constant magnetic field on the rotor follow the rotating stator magnetic field at a constant speed. This speed is dependent on the applied frequency and pole number of the motor. Since the switching frequency is derived from the rotor, the motor cannot lose its synchronism. The current is always switched before the permanent magnets catch up, therefore the speed of the motor is directly proportional to the current switching rate [5]. Recent developments in the area of semiconductor switches and cost-effective DSPs and microprocessors have opened a new era for the adjustable speed motor drives. Such advances in the motor related sub-areas have helped the field of motor drives by replacing complicated hardware structures with software based control algorithms. The result is considerable improvement in cost while providing better performance of the overall drive system [6].

Vector control techniques, including direct torque control incorporating fast DSPs and micro-processors, have made the application of induction motor, synchronous motor, recently

developed PMSM, and BLDC motor drives possible for high performance applications where traditionally only DC brush motor drives were applied. In the past, such control techniques would have not been possible because of complex hardware and software requirement to solve the sophisticated algorithms. However with the recent advances in the field of power electronics, microprocessor, and DSPs this phenomenon is solved. Like DC brush motors, torque control in AC motors is achieved by controlling the motor currents. However, unlike DC brush motor, in AC motors both the phase angle and the modulus of the current has to be controlled, in other words the current vector should be controlled. That is why the term vector control is used for AC motors. In DC brush motors the field flux and armature MMF are coupled, whereas in AC motors the rotor flux and the spatial angle of the armature MMF requires external control. Without this type of control, the spatial angle between various fields in the AC motor will vary with the load and cause unwanted oscillating dynamic response. With vector control of AC motors, the torque and flux producing current components are decoupled resembling a DC brush motor, the transient response characteristics are similar to those of a separately excited DC motor, and the system will adapt to any load disturbances and/or reference value variations as fast as a brushed-DC motor [4]. The primitive version of the PMAC motor is the wound-rotor synchronous motor which has an electrically excited rotor winding which carries a DC current providing a constant rotor flux. It also usually has a three-phase stator winding which is similar to the stator of induction motors. The synchronous motor is a constant-speed motor which always rotates at synchronous speed depending on the frequency of the supply voltage and the number of poles. The permanent magnet synchronous motor is a kind of synchronous motor if its electrically excited field windings are replaced by permanent magnets which provide a constant rotor magnetic field.

2.1 PMSM Introduction

Permanent magnet (PM) synchronous motors are widely used in low and mid power applications such as computer peripheral equipments, robotics, adjustable speed drives and electric vehicles. The growth in the market of PM motor drives has demanded the need of simulation tools capable of handling motor drive simulations. Simulations have helped the process of developing new systems including motor drives, by reducing cost and time.

Simulation tools have the capabilities of performing dynamic simulations of motor drives in a visual environment so as to facilitate the development of new systems.

Modeling and simulation is usually used in designing PM drives compared to building system prototypes because of the cost. Having selected all components, the simulation process can start to calculate steady state and dynamic performance and losses that would have been obtained if the drive were actually constructed. This practice reduces time, cost of building prototypes and ensures that requirements are achieved. In works available until now ideal components have been assumed in the inverter feeding the motor and simulations have been carried out.

The voltages and currents in different parts of the inverter have not been obtained and hence the losses and efficiency can not be calculated. In this work, the simulation of a PM motor drive system is developed using Simulink. The simulation circuit includes all realistic components of the drive system. This enables the calculation of currents and voltages in different parts of the inverter and motor under transient and steady conditions. The losses in different parts are calculated. A comparative study associated with hysteresis and PWM control techniques in current controllers has been made. A speed controller has also been designed for closed loop operation of the drive. Nowadays, as in every area of the technology, a development process has been proceeded in industrial driving systems. The improvement of the switching speeds of the switching equipment has enabled control techniques which have high switching frequency and feasibility of high efficiency driving systems. Using complex control algorithms has become available with the development in microprocessor technology. The applications of vector control for induction and synchronous motors can be given an example of this. As a result of development of various algorithms for system

modelling and control applications, induction and synchronous motors are being used in applications where DC motors were used. However, induction motors efficiency changes with slip value, it needs reactive current, and not able to produce the high torque / weight ratio which needed for high performance applications such as robotics, therefore different solutions are being studied, and different motor designs have been developed. One of these recently developed motors is the permanent magnet synchronous motor. In applications where high performance is demanded, some properties of the permanent magnet synchronous motor such as high torque, high power, high efficiency and low noise have made it more popular compared to other alternating current motors [1]. Especially because of the high power density, permanent magnet synchronous motor is applicable for areas such as robotics, automation and aeronautics technologiesSince the excitation flux is supplied by the magnets and due to the magnet characteristics and location, permanent magnet synchronous motors have both of a synchronous machine and a direct current machine characteristics. Unloaded conditions, velocity is directly proportional to voltage and inversely proportional to the flux and loaded conditions, it is directly proportional to the current and flux. Synchronous motors have three phase windings in their stators, just like the induction motors. However, the rotor structure is different. By using permanent magnets in stead of windings on the rotor, disadvantages of the brush and collector are eliminated. Also, since the excitation losses are eliminated, thermal limits are expanded and higher power values can be obtained from a machine of same volume. Using high energy permanent

magnets such as Sm2Co17 or Nd-Fe-B on the rotor, keeps the air gap flux density at higher values than of wounded machines and eliminates the copper losses of the rotor windings, thus provides the higher efficiency compared to the induction motors at identical power value. Also the motor dimensions are considerably reduced [2]. Permanent magnet synchronous motor is an AC motor that has windings in the stator slots. The flux generated by stator currents is almost sinusodial. Therefore, the same control methods used for the induction motors can also be used for the permanent magnet synchronous motors [3]. These controls are; V/f control, field oriented control, and direct torque control. The choice of direct torque control from these methods gives advantages such as; faster torque control, high torque at low level speed and high speed sensitivity.

2.2 PMSM Operation

A synchronous motor differs from an asynchronous motor in the relationship between the mechanical speed and the electrical speed. In a synchronous motor, the supplied voltages have the same frequency as the mechanical motor speed. In an asynchronous motor, the end mechanical speed is different from the input frequency, and the relationship between input frequency and mechanical speed varies depending on mechanical load applied to the motor.

There are two main components in a PMSM motor: Permanent magnet and stator windings.

This is how the motor operates:

Stator windings generate a rotating magnetic field. Since the permanent magnet rotor has its own magnetic field, it will try to align with the stators magnetic field. The electronics will generate a rotating magnetic field in the stator that is always ahead of the rotor, thus keeping the rotor spinning. The angle between rotor and stator magnetic fields plays an important role on the torque generation of the motor. For simplicity a single pole pair motor is shown, which means that for every revolution of the magnetic field there is one revolution on the motor shaft. In reality most of the motors have multiple pole pairs, where the mechanical speed is the product of input frequency and number of pole pairs. For example, if the motor has five pole pairs, for every revolution on the motor shaft, there will be five revolutions on the magnetic field. A PMSM motor rotates because of the magnetic attraction between the rotor and stator poles.

When the rotor poles are facing stator poles of the opposite polarity, a strong magnetic attraction is set up between them.

The mutual attraction locks the rotor and stator poles together, and the rotor is literally yanked into step with the revolving stator magnetic field. At no-load conditions, rotor poles are directly opposite the stator poles and their axes coincide. At load conditions the rotor poles lag behind the stator poles, but the rotor continues to turn at synchronous speed; the mechanical angle (a) between the poles increases progressively as we increase the load.

2.3 PMSM Applications

Air Conditioner (AC) compressors Direct-drive washing machines Refrigerator compressors Automotive AC compressors

These are some applications where PMSM motors are used. 1. In order to boost the efficiency of air conditioning compressors, PMSM motors are used. Also, physical conditions of a compressor, where the motor is flooded in oil, position sensors are not allowed. Sensor less algorithms are required for compressors in general. 2. PMSM motors are also becoming popular in Direct Drive washing machine due to their torque available at very low speeds. FOC enables better dynamic response in a washing machine without increasing the overall system cost.

3. Refrigerator compressors also require better efficiency and torque performance at low speeds. These requirements are covered by PMSM motors as well. 4. Pumps and Air Conditioner compressors in the automotive applications are also transitioning to PMSM motors due to they efficiency gains, increased life time compared to DC motors, and high torque at low speeds.

The main advantages of using permanent magnets over field excitation circuit used by conventional synchronous motors are given below [4]: Elimination of slip-rings and extra DC voltage supply. No rotor copper losses generated in the field windings of wound-field synchronous motor. Higher efficiency because of fewer losses. Since there is no circuit creating heat on the rotor, cooling of the motor just through the stator in which the copper and iron loses are observed is more easily achieved. Reduction of machine size because of high efficiency. Different size and different arrangements of permanent magnets on the rotor will lead to have wide variety of machine characteristics.

PMAC motors have gained more popularity especially after the advent of high performance rare-earth permanent magnets, like samarium cobalt and neodymium-boron iron which surpass the conventional magnetic material in DC brush and induction motors and are becoming more and more attractive for industrial applications. The positive specific characteristics of PMAC motors explained above make them highly attractive candidates for several classes of drive applications, such as in servo-drives containing motors with a low to mid power range, robotic applications, motion control system, aerospace actuators, low integral-hp industrial drives, fiber spinning and so on. Also high power rating PMAC motors have been built, for example, for ship propulsion drives up to 1 MW. Recently two major sectors of consumer market are starting to pay more attention to the PM motor drive due to its features [7]. PMAC motors have many advantages over DC brush motors and induction motors. Most of these are summarized as [1]: High dynamic response

 High efficiency providing reduction in machine size Long operating life

Noiseless operation


High torque to volume ratio High air-gap flux density Higher speed ranges Better speed versus torque characteristics Lower maintenance cost Simplicity and ruggedness Compact design Linear response Controlled torque at zero speed

Compared to IMs, PMSMs have some advantages, such as higher efficiency in steadystate, and operate constantly at synchronous speed. They do not have losses due to the slip which occurs when the rotor rotates at a slightly slower speed than the stator because the process of electromagnetic induction requires relative motion called slip between the rotor conductors and the stator rotating field that is special to IM operation. The slip makes the induction motor asynchronous, meaning that the rotor speed is no longer exactly proportional to the supply frequency [1]. In IMs, stator current has both magnetizing and torque-producing currents. On the other hand, in PMSMs there is no need to supply magnetizing current through the stator for constant air-gap flux. Since the permanent magnet generates constant flux in the rotor, only the stator current is needed for torque production. For the same output, the PMSM will operate at a higher power factor and will be more efficient than the IM [1]. Finally, since the magnetizing is provided from the rotor circuit by permanent magnets instead of the stator, the motor can be built with a large air gap without losing

performance. According to the above results, the PMSM has a higher efficiency, torque to ampere rating, effective power factor and power density when compared with an IM. Combining these factors, small PM synchronous motors are suitable for certain high

performance applications such as robotics and aerospace actuators in which there is a need for high torque/inertia ratio [1]. Induction machine drives have been widely used in industry. Squirrel-cage induction machines have become popular drive motors due to their simple, rugged structure, ease of maintenance, and low cost. In pumps, fans, and general-purpose drive systems, they have been a major choice. In servo applications, permanent magnet (PM) machine drives are more popular due to their high output torque performance. They are also actively used in high-speed applications, not only as motors/starters but also as generators, as in the flywheel energy storage system (FEES), electric vehicle (EV), and industrial turbo-generator system (ITG) [7]. The smaller the motor, the more preferable it is to use permanent magnet synchronous motors. As motor size increases the cost of the magnetics increases as well making large PMACs cost ineffective. There is no actual breakpoint under which PMSMs outperform induction motors, but the 1-10 kW range is a good estimate of where they do [8].

3. PERMANENT MAGNET SYNCHRONOUS MACHINE MODEL

The equations of the permanent-magnet synchronous machine are:

where va, vb, vc, and ia, ib, ic, and a, b, c, are the stator phase voltages, currents, and flux linkages, respectively, and Rs, is the stator phase resistance. The flux linkages are further defined

as:

where r is the rotor electrical angle, and pm is a coefficient which is defined as:

where pk rpm V / K is the peak line-to-line back emf constant, in rpm V / K (mechanical speed) and P is the number of poles. The stator self and mutual inductances are rotor position dependent, and are defined as:

where Lsl is the stator leakage inductance. The d-axis and q-axis inductances are associated with the above inductances as follow:

The developed torque can be expressed as (equation 12):

A mathematical model of the PMSM on the primitive frame is written in where the symbols used in the model are:

The voltage and current vectors are represented by:

The resistance and inductance matrix are:

The presence of trigonometric terms in these equations is due to the use of two different coordinate systems, one for the stator and another for the rotor. The angle position between these reference frames. By transforming to the stationary reference frame, a simpler relation can be obtained: is the relative

Where Lq and Ld are the stator inductances measured in the direct and quadrature directions. For non-salient rotor Ld is equal to Lq because the magnetic path does not change with the afor the flux depends on the rotors relative position, hence ngular position. For salient rotor machines Ld is not equal to Lq, and the magnetic path for the flux depends on the rotors relative position, hence:

Using (3), the electric torque equation is obtained as:

4. FUZZY INFERENCE SYSTEM

Fuzzy logic has been applied to the standard DTC algorithm to improve the drive performance, reduce ripple in the electromagnetic torque and in the stator currents. The

voltage vector to be used is selected from the switching Table 1, and its magnitude is determined with a three dimensional fuzzy inference system, using torque, torque error and stator current magnitude as input variables..

4.1 WHERE DID FUZZY LOGIC COME FROM?

The concept of Fuzzy Logic (FL) was conceived by Lotfi Zadeh, a professor at the University of California at Berkley, and presented not as a control methodology, but as a way of processing data by allowing partial set membership rather than crisp set membership or nonmembership. This approach to set theory was not applied to control systems until the 70's due to insufficient small-computer capability prior to that time. Professor Zadeh reasoned that people do not require precise, numerical information input, and yet they are capable of highly adaptive control. If feedback controllers could be programmed to accept noisy, imprecise input, they would be much more effective and perhaps easier to implement. Unfortunately, U.S. manufacturers have not been so quick to embrace this technology while the Europeans and Japanese have been aggressively building real products around it.

4.2 WHAT IS FUZZY LOGIC?

FL is a problem-solving control system methodology that lends itself to implementation in systems ranging from simple, small, embedded micro-controllers to large, networked, multi-channel PC or workstation-based data acquisition and control systems.

It can be implemented in hardware, software, or a combination of both. FL provides a simple way to arrive at a definite conclusion based upon vague, ambiguous, imprecise, noisy, or missing input information. FL's approach to control problems mimics how a person would make decisions, only much faster. The past few years have witnessed a rapid growth in the

number and variety of applications of fuzzy logic. The applications range from consumer products such as cameras, camcorders, washing machines, and microwave ovens to industrial process control, medical instrumentation, decision-support systems, and portfolio selection.

To understand the reasons for the growing use of fuzzy logic it is necessary, first, to clarify what is meant by fuzzy logic.

Fuzzy logic has two different meanings. In a narrow sense, fuzzy logic is a logical system, which is an extension of multivalued logic. But in a wider sense, which is in predominant use today, fuzzy logic (FL) is almost synonymous with the theory of fuzzy sets, a theory which relates to classes of objects with unsharp boundaries in which membership is a matter of degree. In this perspective, fuzzy logic in its narrow sense is a branch of FL. What is important to recognize is that, even in its narrow sense, the agenda of fuzzy logic is very different both in spirit and substance from the agendas of traditional multivalued logical systems.

In the Fuzzy Logic Toolbox, fuzzy logic should be interpreted as FL, that is, fuzzy logic in its wide sense. The basic ideas underlying FL are explained very clearly and insightfully in the Introduction. What might be added is that the basic concept underlying FL is that of a linguistic variable, that is, a variable whose values are words rather than numbers. In effect, much of FL may be viewed as a methodology for computing with words rather than numbers. Although words are inherently less precise than numbers, their use is closer to human intuition. Furthermore, computing with words exploits the tolerance for imprecision and thereby lowers the cost of solution.

Another basic concept in FL, which plays a central role in most of its applications, is that of a fuzzy if-then rule or, simply, fuzzy rule. Although rule-based systems have a long history of use in AI, what is missing in such systems is a machinery for dealing with fuzzy consequents and/or fuzzy antecedents. In fuzzy logic, this machinery is provided by what is called the calculus of fuzzy rules. The calculus of fuzzy rules serves as a basis for what might be called the Fuzzy Dependency and Command Language (FDCL). Although FDCL is not used explicitly in the Fuzzy Logic Toolbox, it is effectively one of its principal constituents. In this connection, what is important to recognize is that in most of the applications of fuzzy logic, a fuzzy logic solution is in reality a translation of a human solution into FDCL.

A trend that is growing in visibility relates to the use of fuzzy logic in combination with neurocomputing and genetic algorithms. More generally, fuzzy logic, neurocomputing, and genetic algorithms may be viewed as the principal constituents of what might be called soft computing. Unlike the traditional, hard computing, soft computing is aimed at an accommodation with the pervasive imprecision of the real world. The guiding principle of soft computing is: Exploit the tolerance for imprecision, uncertainty, and partial truth to achieve tractability, robustness, and low solution cost. In coming years, soft computing is likely to play an increasingly important role in the conception and design of systems whose MIQ

Among various combinations of methodologies in soft computing, the one that has highest visibility at this juncture is that of fuzzy logic and neurocomputing, leading to so-called neurofuzzy systems. Within fuzzy logic, such systems play a particularly important role in the induction of rules from observations. An effective method developed by Dr. Roger Jang for this purpose is called ANFIS (Adaptive Neuro-Fuzzy Inference System). This method is an important component of the Fuzzy Logic Toolbox.

Fuzzy logic is all about the relative importance of precision: How important is it to be exactly right when a rough answer will do? All books on fuzzy logic begin with a few good quotes on this very topic, and this is no exception. Here is what some clever people have said in the past.

4.3 HOW DOES FL WORK?

FL requires some numerical parameters in order to operate such as what is considered significant error and significant rate-of-change-of-error, but exact values of these numbers are usually not critical unless very responsive performance is required in which case empirical tuning would determine them. For example, a simple temperature control system could use a single temperature feedback sensor whose data is subtracted from the command signal to compute "error" and then time-differentiated to yield the error slope or rate-of-change-of-error, hereafter called "error-dot". Error might have units of degs F and a small error considered to be 2F while a large error is 5F. The "error-dot" might then have units of degs/min with a small error-dot being 5F/min and a large one being 15F/min. These values don't have to be

symmetrical and can be "tweaked" once the system is operating in order to optimize performance. Generally, FL is so forgiving that the system will probably work the first time without any tweaking. FL was conceived as a better method for sorting and handling data but has proven to be a excellent choice for many control system applications since it mimics human control logic. It can be built into anything from small, hand-held products to large computerized process control systems. It uses an imprecise but very descriptive language to deal with input data more like a human operator. It is very robust and forgiving of operator and data input and often works when first implemented with little or no tuning. Membership Functions

A membership function (MF) is a curve that defines how each point in the input space is mapped to a membership value (or degree of membership) between 0 and 1. The input space is sometimes referred to as the universe of discourse, a fancy name for a simple concept.

One of the most commonly used examples of a fuzzy set is the set of tall people. In this case the universe of discourse is all potential heights, say from 3 feet to 9 feet, and the word tall would correspond to a curve that defines the degree to which any person is tall. If the set of tall people is given the well-defined (crisp) boundary of a classical set, we might say all people taller than 6 feet are officially considered tall. But such a distinction is clearly absurd. It may make sense to consider the set of all real numbers greater than 6 because numbers belong on an abstract plane, but when we want to talk about real people, it is unreasonable to call one person short and another one tall when they differ in height by the width of a hair. ut if the kind of distinction shown above is unworkable, then what is the right way to define the set of tall people? Much as with our plot of weekend days, the figure following shows a smoothly varying curve that passes from not-tall to tall. The output-axis is a number known as the membership value between 0 and 1. The curve is known as a membership function and is often given the designation of . This curve defines the transition from not tall to tall. Both people are tall to some degree, but one is significantly less tall than the other.

Subjective interpretations and appropriate units are built right into fuzzy sets. If I say "She's tall," the membership function tall should already take into account whether I'm referring to a six-yearold or a grown woman. Similarly, the units are included in the curve. Certainly it makes no sense to say "Is she tall in inches or in meters?" Membership Functions in the Fuzzy Logic Toolbox

The only condition a membership function must really satisfy is that it must vary between 0 and 1. The function itself can be an arbitrary curve whose shape we can define as a function that suits us from the point of view of simplicity, convenience, speed, and efficiency.

A classical set might be expressed asA = {x | x > 6}

A fuzzy set is an extension of a classical set. If X is the universe of discourse and its elements are denoted by x, then a fuzzy set A in X is defined as a set of ordered pairs.A = {x, A(x) | x X}

A(x) is called the membership function (or MF) of x in A. The membership function maps each element of X to a membership value between 0 and 1.

The Fuzzy Logic Toolbox includes 11 built-in membership function types. These 11 functions are, in turn, built from several basic functions: piecewise linear functions, the Gaussian distribution function, the sigmoid curve, and quadratic and cubic polynomial curves. For detailed information on any of the membership functions mentioned below, turn to Functions Alphabetical List. By convention, all membership functions have the letters mf at the end of their names.

The simplest membership functions are formed using straight lines. Of these, the simplest is the triangular membership function, and it has the function name trimf. it is nothing more than a collection of three points forming a triangle. The trapezoidal membership function, trapmf, has a flat top and really is just a truncated triangle curve. These straight line membership functions have the advantage of simplicity.

Two membership functions are built on the Gaussian distribution curve: a simple Gaussian curve and a two-sided composite of two different Gaussian curves. The two functions are gaussmf and gauss2mf.

The generalized bell membership function is specified by three parameters and has the function name gbellmf. The bell membership function has one more parameter than the Gaussian membership function, so it can approach a non-fuzzy set if the free parameter is tuned. Because of their smoothness and concise notation, Gaussian and bell membership functions are popular methods for specifying fuzzy sets. Both of these curves have the advantage of being smooth and nonzero at all points.

ERROR &

ERROR-DOT FUNCTION MEMBERSHIP

y

SUMMA

RY INTRODUCTIO N This is the fourth in a series of six to articles share and

intended information

experience in the realm of fuzzy logic (FL) and its application. This

article will continue the example from Part 3 by introducing membership functions and explaining how they work. The next two articles will examine FL inference and defuzzification processes and how they work. For further information, several general references are included at the end of this article.

MEMBERSHIP FUNCTIONS In the last article, the rule matrix was introduced and used. The next logical question is how to apply the rules. This leads into the next concept, the membership function. The membership function is a graphical representation of the magnitude of participation of each input. It associates a weighting with each of the inputs that are processed, define functional overlap between inputs, and ultimately determines an output response. The rules use the input membership values as weighting factors to determine their influence on the fuzzy output sets of the final output conclusion. Once the functions are inferred, scaled, and combined, they are defuzzified into a crisp output which drives the system. There are different membership functions associated with each input and output response. Some features to note are: SHAPE - triangular is common, but bell, trapezoidal, haversine and, exponential have been used. More complex functions are possible but require greater computing overhead to implement.. HEIGHT or magnitude (usually normalized to 1) WIDTH (of the base of function), SHOULDERING (locks height at maximum if an outer function. Shouldered functions evaluate as 1.0 past their center) CENTER points (center of the member function shape) OVERLAP (N &Z, Z&P, typically about 50% of width but can be less

Figures 3 and 4 show the membership function for each variable. The inference systems used with the electric torque and the torque error have a uniform distribution that gives a variable

value. The membership function for the fuzzy region labelled small (S) is an L shaped membership function, for the fuzzy region labelled medium is a triangular membership function, and for the fuzzy region labelled big (B) is a gamma shaped membership function. For the current inference system used to limit the stator current there are only two fuzzy regions labelled small and big, and using an L shaped and gamma shaped membership functions respectively.

Fuzzy sets and fuzzy logic are powerful mathematical tools for modeling and controlling uncertain systems in industry, humanity, and nature; they are facilitators for approximate reasoning in decision making in the absence of complete and precise information. Their role is significant when applied to complex phenomena not easily described by traditional mathematics.

The fuzzy inference rules are presented in Table 2. In this table Z stands for Zero, S for Small, M for Medium and B for Big. The rules were heuristically obtained, and had been tested and corrected in simulations.

Fuzzy Inference Systems discourse universe, torque and torque error

Fuzzy Inference Systems discourse universe, stator current magnitude

5. SIMULATIONS

Figure 5 shows the simulation scheme used in this work. Depending on the magnitude of the torque and flux errors for each stator flux zone, the commutation table selects the next bridge state, and calculates the input voltage applied to the machine model. Torque and flux are obtained directly from the machine model and compared with their corresponding references using two hysteresis comparators.

Three different control methods were simulated: Standard DTC with on-off current limit. In this case errors in torque or flux are corrected by applying the full amplitude of the selected voltage vector, regardless of the actual error amplitude. This step change in the applied voltage causes prominent torque and flux ripple, and these oscillations are reflected on the stator currents.

Modified DTC with PI controlled PWM current limit. The amplitude of the selected voltage vector is modulated by the PI controller as a function of error amplitude, avoiding the big step changes. Proposed Modified fuzzy DTC with inference based current limit.

The selected voltage vector has been modulated by the fuzzy controller. As can be seen, the results here are very similar to the ones produced in case (b), but actual control implementation is much simpler.

The Standard DTC with on-off current limit scheme is a fast and effective way to control electric torque and flux in a PMSM machine. Nevertheless, if no modulation on the vectors selected by the commutation table is used, its behavior is rather bang-bang like, producing prominent ripple on the generated torque and stator currents, which can cause instability and increase the losses problem in the whole system. DTC performance can be considerably enhanced if the amplitude of the voltage vector selected is modulated as a function of the error magnitudes. This modulation considerably reduces the ripple produced by DTC, and does not cause big delays on the overall system response but increases the inverter bridge frequency, reducing the efficiency. Both methods, classical PI and fuzzy logic modulation, show similar capabilities regarding initial peak stator current control, and both are able to reduce torque, flux and stator current ripples. In the simulations, the modified DTC with PI controlled PWM current limit and the new modified fuzzy DTC with inference based current limit schemes produced almost identical results. Experimental tests showed that the new modified fuzzy DTC with inference based current limit scheme outperformed the modified DTC with PI controlled PWM current limit scheme, producing the best results with the biggest reduction in torque, flux and stator current ripple. The advantages of the new Modified fuzzy DTC with inference based current limit schemes produced almost identical results. Experimental tests showed that the new modified fuzzy DTC with inference based current limit scheme outperformed the modified DTC with PI controlled PWM current limit scheme, producing the best results with the biggest reduction in torque, flux and stator current ripple. The advantages of the new Modified fuzzy DTC with inference based current limit scheme are due to the fact that fuzzy techniques inherently include the effects of irregularities or nonlinearitys of the system in their inferences, making this approach more flexible and easier to modify that those based on rigid deterministic models, such as the modified DTC with PI controlled PWM current limit scheme.