Reduction Of Harmonics Constant Dc

Published Date: 02 Nov 2017

CHAPTER -1

INTRODUCTION

[Reduction of harmonics constant DC link load voltages and sinusoidal supply voltage and regeneration of power is difficult in power electronic converter system. Three- phase PWM power converters with DC-link are used in the DC drive system to alleviate harmonics, improve input sinusoidal voltage, power factor and regenerative capability. The model consists of PWM rectifier, current controller and voltage controller. The voltage control of the PWM converter is controlled with PI controllers. The PI voltage controller of PWM converter system is replaced with FPI voltage controller. The present chapter describe briefly about converters, harmonic issue, PWM rectifiers, controllers, current controllers, PI voltage controller and fuzzy PI controller.]

BASICS

In the electric power system, use of AC circuits has been a common practice nearly since the very inception of the interconnected power network. On such a system, the most familiar loads were constant power, constant impedance and constant current loads or a linear combination thereof. In these cases, the voltage and current wave shapes are nearly pure sinusoidal. However, this is no longer the case with modern electric drive system. Large-scale use of the nonlinear and time varying devices has led to distortion of voltage and current waveforms. Consequently, the issue of power quality has become very important recently. Both end users and electric utility of electric power are becoming increasingly concerned about the quality of power.

Power quality and harmonic distortion are the increasing concern for power system engineers. The term "Power Quality" has been used to describe the variation of the voltage, current and frequency on the power system beyond a limit. The outages, voltage drops, power factor, spikes and surges associated with transient conditions are the great problems for AC distribution systems. Due to advent of power electronics, installation of non-linear type loads are increasing day-by-day on the distribution system, causing adverse power system harmonics. The non-linear elements, connected to the power system have non-linear characteristics; this causes harmonic distortion. Some of the common sources of power system harmonics are listed below:

Transformer Saturation

Transformer inrush

Transformer neutral connections

MMF distributions in rotating machines

Electric arc furnaces

Fluorescent lighting

Computer Switch Mode Power Supplies (SMPS)

Battery chargers

Imperfect AC sources

Static VAR compensators

Variable Frequency Motor Drives (VFD)

DC converters

Inverters

Television power supplies, etc.

Among the above, the strongest sources of harmonic distortion are large (VFD) type loads.

1.2 ENERGY ISSUES AND NATURE OF LOADS

Energy consumption is increasing day-by-day. Approximately 70% to 80% of electrical energy is consumed by electric motors, and the rest of the energy is utilized for lighting and other miscellaneous uses. For efficient application of electric motors, more and more portions of motor loads are being applied using power electronic motor drives, which use phase controlled rectifiers to convert AC to DC. The power electronic motor drives yield 20% to 50% of economies and energy savings over the conventional motor electromechanical drive applications. Other benefits of using the power electronic motor drives are to obtain smooth ramp-up and ramp-down acceleration transitions, optimized mechanical load characteristics, equipment reliability, increased product qualities, reduced mechanical related maintenance, audible noise levels and physical space requirements. The motor drives optimize speed and torque of the motor, which saves energy costs.

In last few years, a large amount of equipment have been added to the drive system, controlled by electronics, and are not so tolerant of these variations. Some of the controls are directly through power conversion electronics, such as AC drives, DC drives and SMPS, while some of the electronic equipment is in the peripheral controls, such as computers, and Programmable Logic Controllers (PLC). With the availability of these sophisticated controls, much more precise control of the process have been developed, which make the process even more susceptible to the effects of power system disturbances. Many system disturbances, which have been considered normal for many years, may now cause disruption to the industrial power system with a resulting loss of production.

1.3 HARMONIC ISSUE

Use of adjustable AC and DC drives are increasing day- by- day leading to growing concern for harmonic distortion. The rectifiers of adjustable frequency AC motor drives produce harmonics in industrial electrical systems. The harmonic distortion causes transformer overheating, motor failure, fuse blowing, capacitor failures and mis-operation of controls. A revised version of IEEE standard-519, Recommended Practice for Harmonic Control in Electric Power Systems, provides the recommended limit for harmonics classified in two categories as per IEEE Standard 519 (120). The harmonic current limits are specified for individual customers and these are evaluated at the Point of Common Coupling (PCC) between the customer and the power system. Harmonic voltage limits are specified for the overall power system and provide an indication of the power quality for individual customers. These are designed into the power system to limit the harmonics, so that the resulting voltage distortion will be acceptable for all customers as discussed by McGranaghan and Mullellerr (63).

The harmonics generated by drive’s rectifiers/inverters are reduced using passive filters or active filters to bring the harmonics level below the IEEE-519 limits. The passive filters work on the basis of parallel resonance between capacitors and inductors. A number of passive LC filter steps are needed for required harmonics control. Three or four steps (for 5th, 7th, 11th or 5th, 7th, 11th, 13th harmonics) are sufficient to control the higher order harmonics components to the level specified in IEEE-519. Sometimes the passive filters may result in leading power factor and possible overvoltage. The other way to reduce the higher order of harmonics is to opt for 12-pulse or 18-pulse uncontrolled converters or electronically operated active filters as discussed by Keskar (68) and Reid (121).

Classically, shunt passive filters consist of tuned LC filters and/or high-pass filters to suppress the harmonics, and power capacitors are employed to improve the power factor of the utility/mains. However, these conventional methods have the limitations of fixed compensation, large size and can excite resonance conditions.

Use of three-phase, Multi Pulse AC-DC Converter (MPCs) improves the power quality to reduce harmonic in AC mains and ripples in DC output. These MPCs using diode rectifiers and transformers circuit configurations is in isolated and non- isolated topologies starting with 6 pulses to 12, 18, 24, 30 and higher numbers of pulses to maintain low THD of AC mains current and ripple free DC output.

Figure1.1 Most Popular Three-Phase Harmonic Reduction Techniques of Current

A) Harmonic reduction of already installed non-linear load

B) Harmonic reduction through linear power electronics load installation

1.4 RECTIFIERS

The rectification unit may be either controlled or uncontrolled. Since the utility voltage is sinusoidal, which alternates as a function of time, the first task is to convert it into a useful and reliable constant (DC) voltage for the successful operation of electronic circuits and direct-current machines. The conversion process is called the rectification. Although there are other semiconductor devices suitable for rectification, diodes are frequently employed. A three-phase voltage is converted into a unidirectional voltage using a three diode for a half-wave rectification, and six diodes for a full–wave rectification. The maximum value of a rectified voltage is equal to the maximum value of the input time-varying voltage minus the forward voltage drop across each diode in the rectifying circuit.

Rectifiers employing diodes are called uncontrolled rectifiers, because its average output voltages are fixed DC voltages depending on the magnitude of input supply voltage. The value of output DC voltage is not controlled; its amplitude can vary with the variation in the AC power supply. Six-pulse rectifier is composed of six or multiple of six diodes configured to form a three-phase double-way bridge for AC to DC conversion. The output of this configuration has six pulses per AC cycle and a DC voltage ripple content of 4.17% as per Mohan, et al. (7).

The power converters, also known as rectifiers, are fed from three-phase AC supply for power rating above few kilowatts. Rectifiers employing thyristors are called controlled rectifiers. Here the average output voltage can be controlled. To control the load voltage, the half-wave rectifier uses three, common-cathode thyristor arrangement. The thyristor will conduct when the anode-to-cathode voltage is positive, and a firing current pulse is applied to the gate terminal. Delaying the firing pulse by an angle α, controls the load voltage. The firing angle α is measured from the crossing point between the phase supply voltages. Possible range for gating delay is between α=0° and α=180°, but in real situations, because of commutation problems, the firing angle is limited to a range of 5° to 170°. When the load is resistive, the load current id has the same waveform of the load voltage. When α is smaller than 90°, output voltage Vo is positive, and when α becomes larger than 90°, the average DC voltage becomes negative. In such a case, the rectifier begins to work as an inverter, and the load needs to have the capability to generate power reversal by reversing its DC voltage. For this rectifier, the power transformer has to be oversized 21% at the primary side, and 48% at the secondary side. Then, a special transformer has to be built. In terms of average VA, the transformer needs to be 35% larger that the rating of the DC loads. The larger rating of the secondary respect to primary is because the secondary carries a DC component inside the windings. Besides, the transformer is oversized because of the circulation of current harmonics, which do not generate active power. The core saturation, due to DC components inside the secondary windings, also needs to be taken into account for iron oversizing.

In three-phase full-wave rectifier, parallel connection via inter-phase transformers permits the implementation of rectifiers for high current applications. Series connection for high voltage is also possible. With this arrangement, it can be observed that the three common cathode valves generate a positive voltage with respect to the neutral, and the three common anode valves produce a negative voltage. The result is a DC voltage twice the value of the half wave rectifier. Each half of the bridge is a three-pulse converter group. This bridge connection is a two-way connection, and alternating currents flow in the valve-side transformer windings during both half periods, avoiding DC components into the windings, and saturation in the transformer magnetic core. The configuration does not need any special transformer, and works as a six-pulse rectifier. The series characteristic of this rectifier produces a DC voltage twice the value of the half-wave rectifier.

The need of the converter is to convert the fixed frequency, fixed amplitude input power from utility into a variable frequency, variable amplitude power required for the DC load. In a DC to AC converter, topologies with three-phase full bridge with IGBT(Insulated Gate Bipolar Transistor) or MOSFETs(Metal Oxide Semiconductor Field Effect Transistor) or BJTs(Bi-Polar Junction Transistor ) or MCTs(Metal Oxide Semi-Conductor Controlled Transistor) are most popular because of their own different advantages, as mentioned in Dubey (4) and Dubey, et al. (5).

Half-wave converters are used for low performance while full-wave [H-bridge] converters are used for high performance applications. Half-wave converter topology has half the number of switches as compared to the full-wave topology. The half-wave topology can deliver power only for motoring in both directions without regenerative capability. With the use of two controlled switches having anti-parallel diodes per phase, bi-directional power flow is achieved in the full bridge converter. Topology based on this concept is very common for three-phase AC drives, as specified in Mohan, et al. (6) and Mohan, et al. (7).

It has been observed that the number of pulses can be increased to higher side, above 24 pulses, such as 38 pulses or more, as discussed by Hammond, et al. (60), by the use of autotransformer. The modern drive specifications are required to meet the new IEEE-519 standard, to avoid current and voltage harmonic distortion of utility. Presently, it is being assumed that voltage distortion is the responsibility of utility, and current distortion is the responsibility of the user, as specified by McGranaghan, et al. (63).

Full-wave (midpoint) rectifiers with double star, Zigzag, T-connection, tapped winding in transformer, multiple secondaries for phase shifting and pulse doubling, pulse multiplication, using interface transformer, and additional devices are chosen to suit the requirements. Therefore, these unidirectional AC/DC converters are used as full-wave and half bridge converters based on the number of pulses and circuits in either isolated and no isolated between AC input and DC output, as described by Bhim Singh, et al. (8).

Simply defined, rectification is the conversion of AC to direct DC. Sometimes, the method of rectification is referred to by counting the number of DC "pulses" output for every 360o of electrical "rotation". A three-phase full-wave rectifier would be called a 6-pulse unit. In a poly-phase circuit, it is possible to obtain more pulses than twice the number of phases in a rectifier circuit. Through the creative use of transformers, sets of full-wave rectifiers may be paralleled in such a way that more than six pulses of DC are produced for three phases of AC. A 30o phase shift is introduced from primary to secondary of a three-phase transformer when the winding configurations are not of the same type. In other words, a transformer connected either Y-Δ or Δ-Y will exhibit this 30o phase shift, while a transformer connected Y-Y or Δ-Δ will not. This phenomenon may be exploited by having one transformer connected Y-Y feed a bridge rectifier, and have another transformer connected Y-Δ feed a second bridge rectifier, then parallel the DC outputs of both rectifiers. Since the ripple voltage waveforms of the two rectifier outputs are phase-shifted 30o from one another, their superposition results in fewer ripples than either rectifier output considered separately i.e., 12 pulses per 360o instead of just six. A connection with more pulses on the output voltage has lower ripple and harmonic content; however, they are proportionally more costly.

1.5 PULSE WIDTH MODULATION (PWM) TECHNIQUES

PWM switching patterns are used to improve the input and output performance of the converter. PWM AC/DC voltage-source converter, as compared with the widely used phase-controlled converter, merits attention because of its ability to deliver near-sinusoidal currents at unity power factor. There are many forms of modulation used for communicating information. When a high frequency signal has amplitude varied in response to a lower frequency signal, we have Amplitude Modulation (AM). When the signal frequency is varied in response to the modulating signal, we have Frequency Modulation (FM). These signals are used for radio modulation because the high frequency carrier signal is needed for efficient radiation of the signal. When communication by pulses was introduced, the amplitude, frequency and pulse width become possible modulation options. In many power electronic converters where the output voltage can be one of two values, the only option is modulation of average conduction time. 

Linear Modulation is the simplest modulation to interpret where the average ON time of the pulses varies proportionally with the modulating signal. The advantage of linear processing for this application lies in the ease of de-modulation. The modulating signal can be recovered from the PWM by low pass filtering. For a single low frequency sine wave as modulating signal, modulating the width of a fixed frequency (fs) pulse trains the spectra. Clearly, a low pass filter can extract the modulating component of frequency (fm).

Sawtooth PWM is the simplest analog form of generating fixed frequency PWM by comparison with a linear slope waveform such as a sawtooth. The output signal goes high when the sine wave is higher than the sawtooth. This is implemented using a frequency whose output voltage goes to logic HIGH when one input is greater than the other one.  Other signals with straight edges can be used for modulation. A rising ramp carrier will generate PWM with Trailing Edge Modulation.

Regular Sampled PWM is the scheme that generates a switching edge at the instant of crossing of the sine wave and the triangular wave. This is an easy scheme to implement using analog electronics but suffers from the imprecision and drifts of all analog computation as well as having difficulties of generating multiple edges when the signal has even a small added noise. Many modulators are now implemented digitally but there is difficulty in computing the precise intercept of the modulating wave and the carrier. Regular sampled PWM makes the width of the pulse proportional to the value of the modulating signal at the beginning of the carrier period. The intercept of the sample values with the frequency determines the edges of the pulses.

The hysteresis modulation is a feedback current control method where the actual current tracks the reference current within a hysteresis band. The controller generates the sinusoidal reference current of desired magnitude and frequency that is compared with the actual current. The actual current is forced to track the reference current within the hysteresis band.

There are many ways to generate a PWM signal other than fixed frequency sine sawtooth. For three-phase systems, the modulation of a Voltage Source Inverter (VSI) can generate a PWM signal for each phase leg by comparison of the desired output voltage waveform for each phase with the same sawtooth. One alternative, which is easier to implement in a digital computer and gives a larger modulation depth, is using Space Vector Modulation (SVM).

General method to generate a PWM signal is to use an oscillator to generate a triangle or sawtooth waveform. Frequencies of 30-200 Hz are commonly used. A potentiometer is used to set a steady reference voltage. A comparator compares the sawtooth voltage with the reference voltage. When the sawtooth voltage rises above the reference voltage, a power transistor is switched on. As it falls below the reference, it is switched off. This gives a square wave output. If the potentiometer is adjusted to give a high reference voltage, the sawtooth never reaches it, so the output is zero. With a low reference, the comparator is always on, giving full power.

In the SVM method, the switching states of the converters are determined by using the space-vector PWM method. The converter produces only six nonzero space vectors and two zero vectors of the switching function for steady-state sinusoidal operation of the converter, which the switching function must satisfy.

1.6 PWM RECTIFIERS

To alleviate drawbacks of diode or thyristor rectifiers, different types of high power factor rectifiers for different applications are required. Nowadays, PWM rectifiers are replacing conventional diode bridge and thyristor bridge rectifiers, used for drives load. The line commutated thyristor converters require filters for improving power factor and harmonics. Uses of the diode or thyristor bridges are economical, but reduction in prices of high-power BJTs, MOSFETs and Gate Turn On (GTOs), the PWM converters are also becoming cost effective. The PWM converter is capable to deliver nearly sinusoidal input current with phase angle operating range between 0-3600, also with regulated DC link voltage and regenerative capability, hence proving most suitable for four-quadrant speed control of drives.

Control of phase angle and amplitude of converter voltage are indirectly the control of phase and amplitude of line current. In this way, average value and sign of DC current is proportional to active power conducted through converter. The reactive power can be controlled independently with shift of fundamental harmonic current with respect to voltage. It is being observed that the voltage vector is higher during regeneration (up to 3%) than rectifier mode. It means that these two modes are not symmetrical. Main circuit of bridge converter consists of three legs with IGBT or, in case of high power GTO. The bridge converter voltage can be represented with eight possible switching states.

A drive with PWM converter used as front-end converter maintains unity input power factor and sinusoidal input current, thus feeding negligible current harmonics to the supply system. The output of the converter can be treated as regulated DC voltage supply. The inverter of the drive draws power from the regulated DC voltage terminal called as DC link. A large capacitor connected at the DC link terminal to filter ripples in the DC voltage, is called as DC-link capacitor. DC-link capacitor serves two main purposes-

It maintains a DC voltage with small ripples in steady state.

It serves as an energy storage element to supply real power difference between load and source during transients.

Irrespective of the drives connected across the DC-link, the PWM converter always draws sinusoidal current at unity power factor from the AC utility. An adjustable speed AC/DC drive with front-end, as PWM converter, is the most suitable drive system for industrial applications, as it helps in maintaining the utility power quality also.

1.7 LINE INDUCTOR

Inductors connected between input of rectifier and supply lines are an integral part of the PWM converter circuit. It brings current source character of input circuit and provides boost feature of converter. The line current is controlled by the voltage drop across the inductance interconnecting two voltage sources (line and converter). It means that the inductance voltage equals the difference between the line voltage and the converter voltage as discussed by Abdelli et, al.(2002)[212]. The source inductor provides a power transfer link between three-phase AC system, i.e. (i) three-phase supply (ii) three-phase AC regenerative/ back EMF of PWM converters. The converter is operated with outer DC-link voltage control loop and inner HCC loop.

1.8 CONTROLLERS

To achieve the ripple free DC link output and draw sinusoidal current from supply, the converter needs to use of Controllers. These are fast inner loop current controllers and outer loop voltage controllers.

1.8.1 Current Controllers

Different Current Control Strategies include

Phase and Amplitude Control (PAC)

HCC

Predicted Current Control with Fixed Switching Frequency (PCFF)

PAC is a good switching pattern to reduce steady state current harmonics and output voltage ripple. However, it has a DC current component (current offset) that appears on the AC side of the converter, which deteriorates the DC load current and voltage waveforms during transients. The PCFF shows a fast dynamic response and has a good switching pattern that reduces the switching device stresses. However, it is sensitive to parameter variations.

The basic implementation of HCC derives the switching signals from the comparison of the current error with a fixed hysteresis band. The HCC has a fast dynamic response, good accuracy, no DC offset, and high robustness. It produces a varying modulation frequency for the power converter. This is, in general, responsible for various problems, e.g., uneven and random switching pattern resulting in additional stresses on switching devices, difficulty in designing the input filters to the generation of unwanted resonances on the utility grid. Another negative aspect of the HCC is that its performance is negatively affected by the phase current’s interaction, which is typical of three-phase systems with insulated neutral. Thus, HCC are used to control the PWM converter three-phase input currents. The peak amplitude of the reference current is multiplied with three unit vectors. The three unit vectors are generated using three-phase supply voltages obtained from the common coupling point. The three-phase supply voltage templates are divided by their peak amplitudes resulting in Ua, Ub and Uc, three unit vectors in phase with phase ‘a’, phase ‘b’ and phase ‘c’ voltages respectively. These unit vectors Ua, Ub and Uc are multiplied with peak value of the reference current (obtained from the voltage controller) to give out Ia*, Ib* and Ic* respectively.

1.8.2 PI Voltage Controller

The voltage control of the PWM converter is controlled with PI controllers. Proportional Integral and Derivatives (PID) controller is avoided because differentiation can be problematic when input command is a step. The PI controller gains should be designed such that the converter response is well accepted under all operating condition. If the gains of the PI controller are fixed, it is not possible to obtain good response in all operating conditions. The PI controller gains depend on accurate model of the PWM converter for fixed parameters. Obtaining accurate model of the scheme is very difficult because of assumptions made to simplify the converter model. Also in actual operating conditions, the converter parameters are not fixed due to non-linear behaviour of converter. Fixed PI gains give good response only for fixed operating conditions and parameters for which it is designed and optimised. To cover a wide range of operating conditions of the drive, auto tuning of controller gains is required. The self-tuning controller methods are model-based and model independent. The model based self-tuning method works on an assumed or estimated model of the drive. Its effectiveness depends on accuracy of the assumed or estimated model. The estimated model does not include real conditions like system non-linearities, disturbance and noise etc., thus, causing inaccuracy in the estimated model. The inaccuracy of the estimated model is reflected on degraded controller performance.

1.8.3. Fuzzy PI voltage Controller

The controller gains are assigned by computed look-up table on a case study of the possible working conditions, or the look-up table is replaced by trained neural networks. Model free controllers are based on fuzzy control, sliding mode or variable structure control and neural/fuzzy control. The Fuzzy Logic Controllers (FLC) do not need accurate mathematical model of the drive, it can work with imprecise inputs, it can handle non-linearities and it is more robust than conventional non-linear controllers are. The fuzzy logic control applications fall mostly into two categories, (i) Fuzzy-PI (FPI) controller which generates an incremental output from the error and derivative of the error; (ii) Fuzzy Proportional-Derivative (FPD) controller which directly generates the control action from the error and the derivation of the error, as by Lee, et al. (42). The motor terminal voltage is derivative of the drive load. The derivative of error further amplifies noise, which leads to poor response of the voltage controller. Hence, FPI controller is preferred over FPD controller. In the look-up table method, the output current is pre-computed based on scaled error and change in error inputs, and these are stored in form of a look-up table. Based on the values of inputs, the output is read from the look-up table. The real time calculation needs large processing time than the simple look-up table method, but it gives results that are more accurate. The present drive scheme is implemented based on real time floating point value calculations of fuzzification, evaluation of rule base and defuzzification on the digital computer. The drive error and change in error in the sampling interval are inputs to the FPI control scheme. The input variables are multiplied by scale factor and saturated between prefixed upper and lower limits (-1 and 1). The inputs are fuzzified using a triangular membership function. The inference engine calculates outputs using the rule base matrix for each sampling of input variables. The output of the inference engine is defuzzified using mean of maxima method. The defuzzified output is multiplied by the output scale factor to obtain increment value of the command current. Integration of the command current increment gives out the current command for voltage control of the drive.

The membership functions for input and output variables are chosen as triangular functions for simplicity. The membership functions are distributed evenly in the universe of discourse. The universe of discourse for voltage error input is (-1, 1), for change in voltage error input is (-1, 1) and also for the output variable it is (-1, 1). Five or seven term fuzzy system is sufficient for application of FLC to electrical drive systems; more fuzzy terms will not yield apparent improvement of the controller performance. The maximum amplitude of membership function is 1. Hence, the measurements of membership range of input variables are from 0 to 1. The formation of the rule-base matrix is based on experience of human operator expert in the area of drive control.

1.9 GENETIC ALGORITHMS (GAs)

GAs are search procedures based on the rule, "Survival of the fittest will win". GAs work on the natural evolution process. It (GA) is a search mechanism based on the principle of natural selection and population genetics, which is transformed by three genetic operators: Selection, crossover and mutation.

Each string (chromosome) has a possible solution to the problem being optimized and each bit (or group of bits) represents a value or some variable of the problem (gene). These solutions are further classified by an evolution function, giving better values, or fitness, to better solutions. Each solution must be evaluated by the fitness function to produce a value. In selection, a number of selected exact copies of chromosomes in the current population become a part of the offspring. In crossover, randomly selected subsection of two individual chromosomes is swapped to produce the offspring. In mutation, randomly selected genes in chromosomes are altered by a probability equal to specified mutation rate as specified in Masimoic, et al. (123). The evolution process is operated on chromosomes. The decoded structures of the chromosome are responsible for its performance, and the natural selection is linked with performance. In natural selection, the chromosomes having structures which provide desirable performance are reproduced more often those that having undesirable performance.

In the evolution process, combining materials from the chromosomes of parents produce new chromosomes in children. Mutation causes the chromosomes of the children to be different from the chromosomes of their parents. Generation after generation, the fittest chromosomes for a desirable performance are produced in the population and they also carry historical information. This natural evolution process is appropriately incorporated in computer algorithms to solve difficult search problems as discussed by Toliyat, et al. [122]. Therefore, it can be said that GAs are computerized search and optimization algorithms based on the mechanism of natural genetics and natural selection.

The use of GAs includes the knowledge of disciplines of biology, computer science, image processing and pattern recognition, physical sciences, social science and neural networks. In other words, it can be said that GAs are good at taking larger, potentially huge search space and navigating them looking for optimal combinations of things and solutions, which we may not find in a lifetime. GAs are very different from most of the traditional optimization methods. They need design space to be converted into genetic space. Therefore, GAs work with a coding of variables. The advantage of working with a coding of variable space is that coding discretizes the search space even though the function may be continuous. A most important difference between GAs and most of the traditional optimization methods is that it uses a population of points at one time in contrast to the single point approach by traditional optimization methods. Transition rules are used that are deterministic in nature but GAs use randomized operators. Random operators improve the search space in an adaptive manner as mentioned by Mohamed, et al. [155]. Three most important aspects of using GAs are:

Definition of objective function

Definition and implementation of genetic representation

Definition and implementation of genetic operators

APPLICATIONS

Most of the industrial DC drive loads require rectified power supply. Specific rectifiers and inverters may be required for the individual drives to meet out their power supply requirements. The rectified power supply being used for DC drive operations is a significant fraction of the generated electric power. The diode or thyristor rectifiers draw current non-sinusoidal in nature. The non-sinusoidal currents inject harmonics to power lines, causing significant harm to the power distribution system. The large harmonic content in the rectifier input current also causes low power factor. The large harmonic current leads to voltage distortion and EMI in power lines. Low power factor of the load causes losses in power distribution wiring, as well as in the utility power lines and equipment. Other power line users are badly affected by large harmonic currents and low power factor loads of the supply lines.

The development of new power semiconductor devices, new circuit topologies with their improved performance and their fall in prices have opened up wide field for the new applications of power electronics converter. Further use of semiconductor devices in conjunction with microprocessor or microcontrollers has further enhanced the control strategies and the capabilities of power electronic converter. The power flow received at the load can be controlled by power electronic converters, which are either fully on or fully off, as discussed by Bhimara (1).

Modern solid state devices can be used to control efficiently and safely. However, there is a considerable tendency to generate harmonics in the supply system as well as in the load circuit. In the load circuit, the operating performance of the different nature of the load is affected, for example, high harmonic content in the load circuit causes commutation problem in DC machine, increased motor heating and more acoustical noise in both DC and AC machines. These harmonics must be filter out from the output side of a converter, as mentioned in Bhimra (1) et al. and Gardy et al. (3).

Circuit power levels range from few watts to large mega watts. Although power electronic devices such as thyristors, IGBT, MOSFET etc. Power transistors have been available since around early 1960s. State-of-the-art power electronic devices based on semiconductor devices technology are continuously improving providing better, more reliable and more economical components, as discussed by Gardy et al. (3). The need of converter is to convert the fixed frequency, fixed amplitude input power from utility into a variable frequency, variable amplitude power required for the DC load. In a DC to AC converter topology with three-phase full bridge with IGBT or MOSFETs or BJTs or Metal Oxide Semiconductor (MOS) Controlled Thyristor (MCTs) are most popular because of their own different advantages, as mentioned by Dubey (4) and Dubey, et al. (5).

Solid state AC/DC converters are widely used in a number of applications such as Adjustable Speed Drives (ASD), High Voltage DC Transmission (HVDC), electro-chemical processes, in electro-plating, battery charging, telecommunication supplies, UPS, high power induction heating equipment, high capacity magnet power supplies, plasma power supplies, converters for renewable energy conversion system and for aircraft converter system.

The power converters are fed from three phase AC supply for power rating above few kilowatts. In this way, power quality problems in terms of harmonic injections cause poor power factor, AC voltage distortion and output DC ripples. These problems in AC/DC conversion give several standards and guidelines, which are laid down and should be considered by designers, manufactures, and users. There are different methods used to mitigate these power quality problems in AC-DC converters. Sometimes the ratings of the filters used are close to the converter ratings which not only increase the cost but also the losses and component count resulting in reduced reliability of the system. Installation of filters in the next stage is preferred to modify the converter structure at design stage either using active or passive (magnetic) wave shaping of input current. In AC/DC converter, active wave shaping technique is widely used in a number of applications. However, passive wave shaping technique of these converters is considered a simple and economical method of improving the power quality in some applications, as discussed by Schaeffer et al. [9], IEEE Guide [10], IEEE Recommended Practice [11], IEC [12], IEEE Guide [13], EMC [14], Power Quality Measurements [15] and Choi et al. [16]

A number of novel configurations of MPC AC-DC converters are developed in view of great advantages for unidirectional and bidirectional power flow starting from twelve to a large number of pulses, as presented by Singh et al. [8]. Therefore, it is considered a timely attempt to present MPC AC/DC converters for engineers using them and dealing with reduced harmonics issues.

MPCs are developed to high accurate value for AC/DC converter with reduced harmonics contents and reactive power burden, low EMI and RFI in input mains and good quality reduced ripple in DC output with unidirectional and bidirectional electrical power flow from feeding loads from a few kilowatts to several hundred megawatts. With varying configurations the MPCs have evolved in over 50 years with varying configurations, reduced magnetic circuit integrations, different concepts such as pulse multiplications, phase staggering, varying connections such as T-connection, Zigzag, fork, extended delta and double star, polygon, reduced rating transformers, optimum interface reactors, pulse doubling circuits. Therefore, a number of circuit configurations of MPCs are explored to meet exact requirement of vast applications while maintaining a high level of power quality at the AC mains and output DC loads. Thus, they can be used for vast varying requirements of applications from a few kilowatts to hundreds of kilowatts large rating variable frequency AC/DC motors in compressors, blowers, pumps, etc. However, a bidirectional power flow is required into AC-DC conversion in some applications from a few kilowatts DC motor drives to several megawatts HVDC transmission system. The optimum value of DC link inductor and leakage reactance of input transformer reduces drastically the values of THD of AC mains current, thus resulting in improved power quality. Sometimes, MPC technology is considered superior to PWM technology, since it reduces EMI, RFI and switching losses due to low frequency soft switching caused by line/natural commutation, besides reduction of harmonics which results in high efficiency and low noise level in the converter system. The MPCs are most robust, reliable, and simple in construction as discussed by Singh, et al. [8].

For medium to large size drives, twelve- and eighteen-pulse converters are readily available. The choice of pulse number is a matter of economic versus harmonic control as discussed by Paice (1996) [34] and Verdelho and Marques (1998) [67]. Twelve-pulse unidirectional converters are extensively used in both isolated and non isolated circuit topology while considering voltage levels on the input AC mains and DC output. If output voltage required as much lower DC levels such as in electroplating, then isolated circuits topology are preferred from the protection point of view. However, in some cases, if voltage level of input AC and output DC is very close, and isolation is not required between AC input and DC output, then these converters of full-wave and bridge type are used with drastic reduction in weight, volume, cost, size, and losses in magnetic using autotransformer configurations. Thus, these converters are finding increased applications due to simplicity with high efficiency and high level of power quality, as discussed by Paulillo, et al. [33], Paice, et al. [34], Paice, et al. [46], Kamath, et al. [47], Niermann [48], Nishida [50], Oliver, et al. [53] and Toliyat, et al. [67]. Application of eighteen-pulse AC-DC converters is achieved to improve performance in terms of percentage THD low in AC mains current and to have low value of output voltage ripples. These are also used both in isolated and non-isolated circuit topologies depending upon the requirement of specific applications, as discussed by Paice, et al. [34], Fuentes, et al. [38], Hammond [59], and Rosa [61].

Similarly, 24-pulse AC-DC converters are used in high rating applications. It provides ripple-free DC output and almost sinusoidal AC current in AC mains, as discussed by Wiechmann, et al. [24], Guimaraes, et al. [29], Lee, et al. [42], Choi, et al. [43], Masukawa, et al. [54], Oguchi, et al. [55] and Domingues, et al. [57].

The thirty-pulse AC-DC converters use tapped delta connected auto transformers for conversion of three-phase AC mains to fifteen-phase AC source to feed 30 diode circuits to provide a DC thirty-pulse output. The thirty-pulse circuit is quite cost effective because of use of only autotransformer, and no other components. This results in a high level of performance with low percentage of THD of AC mains current with almost unity power factor and DC output with negligible ripple in output voltage, as stated by Traver, et al. [62].

1.11 CHALLENGES

Modern industries comprise of power electronic devices to control the voltage, speed or frequency of the devices used in industries. These power electronic converter circuits have power electronic controllers having low overload capacity. Regeneration of power is difficult in power electronic converter system. Conventional power controllers are based on thyratrons, mercury-arc rectifiers, magnetic amplifier rheostatic controller etc. These conventional power electronics converters have been replaced by power electronic controllers using semiconductor devices in almost all applications, as discussed by Bhimara (1).

It is being observed that in a three-phase system, unbalanced of supply voltage can be due to various reasons. Characteristics of the converter and input current quality are poor with the increase of unbalances. It has been observed that the level of unbalance plays a significant role in the converter characteristics, mainly at the lower output voltage range. By the phase 80’s and early 90’s the three phase AC-DC converter are widely used in industrial applications. The static power converters are non-linear in nature, that’s why they generate harmonics into the supply, as discussed by Gardy [2]. Angle controlled scheme may be used to analyze the effect of supply unbalance on the power factor of input and output currents, THD of output voltages and lower order harmonics.

In the supply system, the harmonics distort the voltage waveform and seriously influence the performance of the equipment connected to the same supply line. In addition, the harmonics in the supply AC line can also cause interference in audio and video equipment. It is, therefore, necessary to insert filter on the input side of a converter. Design of such filters is challenging.

The problems associated with the present of harmonics on power distribution systems are not just the power quality problems but they also affect the energy efficiency of the system. Typical problems include overheating of transformers, motors, phase and neutral conductors, causing unacceptable neutral-to-earth voltage, voltage distortion, EMI, capacitor bank failure, etc. Many of the problems are related to the proliferation of non-linear loads such as variable speed motor drives, rectifiers for direct-current power supplies, electronic ballasts in energy efficient lighting and SMPS in computers and other electronic office equipment.

To achieve the desired phase shift to cancel, eliminate, and to reduce harmonics in input AC mains feeding AC-DC converters, different methodologies used are zigzag, polygon, T-connection, Tapped winding, polarity of winding of isolated multi-winding transformers and autotransformers. In the unidirectional AC-DC converters, AC input to DC output is used in variable frequency AC drives for fans, compressors, pumps, waste-water treatment plants, electroplating, telecommunication, and power supplies, etc. Development of these converters is possible due to use of diodes and transformers with other additional components. Quality of power supply can be enhanced by further use of more number of pulses, which reduces cost, increases reliability and power quality, as discussed by Singh, et al. [8].

Front-end PWM converter is most suitable for the modern variable speed drive; its analysis and controller design is challenging. From the control structure point of view, the front end converters are put in three groups as the phase and amplitude control, the HCC and the utility voltage orientation control. The phase and amplitude control structure is simple and it provides a good switching pattern to reduce the steady state current harmonic and voltage ripple; it has limitations as it possesses offset DC current during a transient and also the stability of the system is difficult at very low damping factor. HCC is fast with no DC offsets. With HCC, the input line current tracks the command vector in negligible response time and the control technique is insensitive to time voltage and parameter variations. Later PWM converters were used extensively. They are most suitable for drive loads since they have the capabilities of DC bus voltage, regulation, bidirectional power flow and controlled power factor with reduced input current harmonics. Output DC link capacitor voltage is kept constant and ripple-free. The three-phase input currents are maintained sinusoidal and in phase with supply voltage. The PWM converters control the harmonics injected to the supply and ensure unity power factor operation, as discussed by Tiwari [66].

OBJECTIVE

The modern drive specifications are required to meet the new IEEE-519 standard to avoid current and voltage harmonic distortions of the utility. The main problem is in enforcing the standards. Present assumption is that the voltage distortion is the responsibility of the utility and the current distortion is the responsibility of the user. As a result of imposing the limits on voltage and current harmonics, the variable frequency drive market is moving towards the higher pulse converter as front ends. For medium to large size drives, twelve- and eighteen-pulse converters are readily available. The twenty-four and thirty-pulse converters are under development. The choice of pulse number is a matter of economic versus harmonic control. The large AC drives (100 HP and above) typically use thyristor converters for the bus voltage and line current control for the voltage regulation. The bus voltages are controlled by switching of the thyristors in the converter.

An integrated DC drive system using a front-end PWM rectifier link is most suitable for the modern variable speed drives. Three-phase PWM power converters with DC-link are used in the DC drive system. The input current maintains near sinusoidal waveform at the utility frequency with unity power factor and it can work with leading power factor also. It is also capable of regenerative operation, in integration with DC drives. The front-end PWM converters comply with the IEEE 519 standards for the drive unit. With the advent of high power and high frequency switching devices, the application of the front-end PWM converters is possible for medium to large power drives (up to 100 kW). Due to non-linear nature of the converter operation and non-linear nature of the drive load, the analysis and controller design of the PWM converter is challenging.

The present work aims to investigate in-depth the performance of the AC/DC PWM rectifier which includes:

Discussion and analysis of the power quality issues, its standards, different types of rectifiers including PWM rectifier and its harmonic analysis.

Mathematical modelling of the PWM converter.

Design and optimization of outer loop conventional PI voltage controllers for the rectifier.

Design and optimization of outer loop FPI based voltage controller and comparison with conventional PI.

WORK CARRIED

Power quality issues, problems, its standards, equipment causing power quality problems, harmonics mitigation techniques and harmonic effects were discussed. Different types of rectifiers, their configurations, connections and harmonic analysis have been observed. Advantages of PWM rectifier, modelling, and design of outer loop PI controllers along with inner loop HCC with fixed band are observed. Design and optimization of FPI voltage controllers are analyzed and compared with PI voltage controllers.

The simulation of the system has been carried out using MATLAB software. The research work has been accomplished by referring literature from renowned journals. Various infrastructure facilities that are available in the Institute have been used.

1.14 ORGANISATION OF THE THESIS

The thesis is organized in seven chapters. The work included in each chapter is briefly outlined as follows–

Chapter 1 includes different terminologies and topologies of power quality, rectifiers, different controllers, their configuration and applications, their effects on the power system while considering the topological and control methodological aspects.

Chapter 2 deals with extensive literature survey about PWM rectifiers and power quality from renowned journals.

Chapter 3 describes power quality issues, standards and performance. Different types of rectifiers and its harmonic analysis have been discussed.

Chapter 4 deals with a mathematical model of PWM Rectifier. The complete front-end PWM rectifier with voltage and current controller schemes is explained in details. The model consists of PWM rectifier, current controller and voltage controller. The PWM converter system is represented by a set of first order non-linear differential equations. The switching models are described in order to emphasize their behaviour on rectifier model.

Chapter 5 describes design of PI voltage controller for the PWM converter. The transfer function for PI voltage controlled DC Motor drive is obtained in discrete time domain. Stability of the drive is studied with variations of load parameters and different sampling periods. Optimum values of controller parameters are searched using GAs. For DC-link voltage PI controller design, closed loop transfer function is developed in discrete time domain. The stability of the control system with parameter variations is described. Optimum controller parameters are obtained using GAs. Simulation studies of transient performances of the PWM converter are observed with the designed controller parameters.

Chapter 6 describes implementation scheme of FPI voltage controller on the PWM converter model. The PI voltage controller of PWM converter system is replaced with FPI voltage controller. Design of FPI voltage controller with two input and an output variables is discussed. The off-line computation algorithm of the FPI controller is explained in detail. The input scale factors are optimised using GAs. Using the optimum values of the input scale factors, the complete scheme is simulated and implemented. The voltage response obtained by the FPI controller is compared with the voltage response of the PI voltage controller having offline optimised PI gains.

Chapter 7 summarizes the important conclusions of the work and states the scope for future research.

List of important references is given in the end.

1.15 SUMMARY

Since many decades, many power electronic converters utilizing switching devices are being widely used in industrial as well as in domestic applications. In addition to the numerous advantages, these power electronic converters suffer from the problem of drawing harmonics and reactive components of current from the source, and offer highly non-linear characteristics. Increase in such non-linearity causes different undesirable features like, low system efficiency, poor power factor, and disturbance to other consumers, and interference in nearby communication networks, etc. The current harmonics produced by these nonlinear loads further result in voltage distortion and leads to various power quality problems, which led to implementation of standards and guidelines such as IEEE 519-1992, for controlling harmonics in the power system. The power quality standards are operational characteristics, tolerances, and limits. These allow facilities containing power sensitivity and power disturbing loads to operate on the distribution system with a minimum interference to utility equipment or customer loads. The standards are important for determining degree of power quality of any electrical distribution system, feeder, etc. Harmonic distortion decides limits on current distortion caused by equipment or loads. Consumer electronics equipment causes a large part of the harmonic voltage distortion. The acceptable level of power quality can be obtained through a cooperative effort among power users, equipment manufacturers and utilities, and causes longer service life for equipment and lower cost for electric service. Power quality monitoring is an essential service to many utilities serving their industrial and other main commercial customers. Energy consumption is increasing day-by-day and approximately 70% to 80% electric energy is consumed by electric motors and the rest of the energy is utilized for lighting and other miscellaneous uses. The power electronic motor drives yield 20% to 50% of economies and energy reduction savings over that of conventional motor electromechanical drive applications.

The use of adjustable AC and DC drives is increasing day by day, leading to growing concern for harmonic distortion. The rectifiers of adjustable frequency AC motor drives produce harmonics in industrial electrical systems. The rectification unit may be either controlled or uncontrolled. Since the utility voltage is sinusoidal, which alternates as a function of time, the first task is to convert it into a useful and reliable constant (DC) voltage for the successful operation of electronic circuits and DC machines. Although there are other semiconductor devices suitable for rectification, diodes are frequently employed. Multi-pulse techniques are frequently found in high power applications. The thyristor rectifiers present the same robustness as that of the diode rectifiers. The complexity and the costs are little increased due to gate drive circuit. PWM Rectifiers use active rectification techniques which is the most promising rectifier technology from a power quality viewpoint. A unity power factor and a very low harmonic distortion can be achieved. Rectifiers employing thyristors are called controlled rectifiers. PWM switching patterns are used to improve the input and output performance of the converter. There are many forms of modulation used for communicating information. There are many ways to generate a PWM signal other than fixed frequency sine sawtooth. The PWM converter is capable to deliver nearly sinusoidal input current with phase angle operating range between 0-3600, also with regulated DC link voltage and regenerative capability, hence proving most suitable for four-quadrant speed control of drives. Inductors connected between input of rectifier and lines are integral parts of PWM converter circuit. The source inductor provides a power transfer link between three-phase AC system, i.e. (i) three-phase supply (ii) three-phase AC regenerative/back EMF of PWM converters. The converter is operated with outer DC-link voltage control loop and inner HCC loop. Different Current Control Strategies include the Phase and Amplitude Control (PAC), HCC and the Predicted Current Control with Fixed Switching Frequency (PCFF). The HCC has a fast dynamic response, good accuracy, no DC offset, and high robustness. It produces a varying modulation frequency for the power converter. Thus, HCC are used to control the PWM converter three-phase input currents. The voltage control of the PWM converter is controlled with PI controllers. FPI voltage Controller gains are assigned by computed look-up table on a case study of the possible working conditions or the look-up table is replaced by trained neural-networks. Model free controllers are based on fuzzy control, sliding mode or variable structure control and neural/fuzzy control. The inference engine calculates outputs using the rule base matrix for each sampling of input variables. The output of the inference engine is defuzzified using mean of maxima method. The defuzzified output is multiplied by the output scale factor to obtain increment value of the command current. Integration of the command current increment gives the current command for voltage control of the drive. GAs work on the natural evolution process. It (GA) is a search mechanism based on the principle of natural selection and population genetics which is transformed by three genetic operators: Selection, crossover and mutation. In the evolution process, combining materials from the chromosomes of parents produce new chromosomes in children. Mutation causes the chromosomes of the children to be different from the chromosomes of their parents. Generation after generation the fittest chromosomes for a desirable performance are produced in the population and they also carry historical information. This natural evolution process is appropriately incorporated in computer algorithms to solve difficult search problems. It needs design space to be converted into genetic space. So, genetic algorithms work with a coding of variables. The advantage of working with a coding of variable space is that coding discretizes the search space even though the function may be continuous. A most important difference between GAs and most of the traditional optimization methods is that it uses a population of points at one time in contrast to the single point approach by traditional optimization methods; transition rules are used that are deterministic in nature but GAs uses randomized operators. Random operators improve the search space in an adaptive manner.