CHAPTER 1
INTRODUCTION
Background
Power quality is considered to be very important in the distribution system. The type of load used will determine the quality of the power supply. However the major causes of harmonic generation are industrial electronic devices and non-linear loads. These loads degrade the power quality as they draw non-sinusoidal current from the mains thus causing harmonic distortion. There was less harmonic distortion in the past since the designs of power system were rather simple and conservative. But, nowadays harmonic distortion has drastically increased due to the use of complex design in the industry.
Harmonic distortion causes both the voltage and current waveform to change from the ideal sinusoidal waveform. The current distortion is produced by electronic loads, which are also called non- linear loads. These non-linear loads might be single phase loads or three phase loads. Moreover the harmonic currents generated by the non-linear loads increase the power system heat losses and power bills for the end users.
Active and Passive power filtering are the two main approaches used to reduce harmonic distortion in power distribution systems. Passive filter causes resonance, thus affecting the stability of the power system. Hence Active Power Filter (APF) which is connected in parallel with the non-linear load is a better solution for the reduction of harmonics. The APF uses power electronics topologies to produce current components which cancel the harmonic current components caused by the non-linear loads. Active filters have many advantages as they provide lower disturbances, lower current consumption, lower carbon dioxide emission through improved energy efficiency and increased production stability.
1.2 Problem Statement
Harmonics is a major cause of power supply pollution
Modern electronic equipments such as personal computers, printers, copiers, medical test equipment, fluorescent lighting, adjustable speed drives and other non-linear loads generate harmonics in the electrical distribution system. There are two types of non-linear loads: single-phase and three-phase. A non-linear load consumes electricity and thus draws a non-sinusoidal current from the supply when supplied with sinusoidal voltage. The harmonic currents produced by the non-linear load generate additional losses in electrical installations and can cause thermal overload.
Solution to reduce harmonics in power distribution system
There are many ways to reduce harmonics depending on the cost and load. Active Power Filter (APF) is considered to be the best solution to reduce harmonics in the power system since they have no resonance problems and are inherently current limiting.
Design and Simulation of active power filter is still an issue
Even though many research and simulation based on active power filter have been done, there still have no perfect method to reduce harmonics effectively. Active power filters are still an open problem since they are limited in their frequency range and cannot handle large amount of power. They require D.C power supply for their operation.
Aims and Objectives
Aims
The main aim of this project is to explain the effects of harmonics in the power system and designing active power filters to remove high value harmonics which causes non-sinusoidal waveform.
Objectives:
To study about active power filter and identify the method to reduce harmonic effects using active power filter.
To design circuits of active power filter.
To design and simulate single- phase shunt active power filter for harmonic reduction using Matlab/Simulink.
To design and simulate three-phase shunt active power filter for harmonic reduction using Matlab/Simulink.
1.4 General structure of active filtering scheme
Active Power Filter is used to compensate current-based distortion and voltage-based distortion such as voltage harmonics, voltage sag and swells. The Active Power Filter injects equal current or voltage distortion in the network but in opposite magnitude. Thus the injected current or voltage cancels the actual distortion present in the circuit. Active Power Filter uses fast switching insulated gate bipolar transistors (IGBTs) that produce output current which cancels the original load generated harmonics when injected into the Ac lines.
Figure 1.1 shows the different component of an Active Power Filter system
Figure 1.1
Shunt Active Power Filtering (Jarupula Somlal, 2011)
In this configuration, the filter is connected in parallel with the load being compensated. The harmonic control is possible due to the voltage source inverter in the active filter. This inverter generates a signal that will cancel the harmonics from the non-linear load.
Types of active power filters
Shunt Active Power Filters:
It compensates current harmonics by injecting equal-but-opposite harmonic compensating current. It operates as a current source injecting the harmonic components generated by the load but phase shifted by 180deg (ARUN KUMAR N.K, 2013).
`
Figure 1.2
Compensation characteristics of active power filter (ARUN KUMAR N.K, 2013)
Series active power filter:
Series active power filter functions mainly as a voltage regulator and harmonic isolator between the non-linear load and the utility system. In order to protect the consumer from a poor supply voltage quality, the series active power filter is considered. The series active power filter injects a voltage component in series with the supply voltage and therefore can be regarded as a controlled voltage source, compensating voltage sags and swells on the load side (Dr. Hopkins, 2011). Figure 1.3 below shows the series active power filter topology.
Figure 1.3 Series active power filter (NUR IZZATI NADIAH BINTI ISHAK, 2010)
Hybrid active power filters:
Hybrid Filters are more complex filters which are made up of combination of active and passive filters. They are used for the conditions where both voltage and current are leading to deterioration in power system.
Figure 1.4
Hybrid active power filter (NUR IZZATI NADIAH BINTI ISHAK, 2010)
1.6 Dissertation Outline
Chapter 2 is about the generation of harmonics and their effects on power quality. It presents the different techniques used to mitigate harmonics. It also includes a description of how switch mode power supplies affect line voltage waveforms and power factor.
Chapter 3 consists of the design of shunt active power filter for a single phase system. The shunt active power filter is implemented using simulation scheme. All the stages in the system are thoroughly analyzed mathematically.
Chapter 4 presents the design of shunt active power filter for a three phase system. Emphasis is given on the implementation of shunt APF. It also presents the mathematical model for the current sensing stage, filter, modulator, inverter and the interconnection back to ac lines.
Chapter 5 discusses the simulation test and results for single phase and three phase shunt APF. From these results discussions are made.
Chapter 6 consists of the conclusion on the work achieved in the project and summarizes the research undertaken. It also offers recommendations for further research.
Chapter 7 contains all the references that have been used in this project.
CHAPTER 2
HARMONICS GENERATION AND EFFECTS ON POWER QUALITY
2.1 Overview of Harmonics
Harmonics are currents or voltages with frequencies that are integer multiples of the fundamental power frequency being 50 or 60Hz (50Hz for European power and 60Hz for American power), for example if the fundamental frequency is 60 Hz, the second harmonic will have a frequency of 120 Hz (260 Hz) and the third harmonic will have a frequency of 180 Hz (360 Hz) and so on. Harmonics are electric currents and voltages found in the power system as a result of non-linear loads. The power system is deteriorated when harmonics cause distortion in the current and voltage waveforms. If large quantities of harmonics are present in the power system, this can cause malfunctioning of the system. Harmonic frequencies in the power grid are a frequent cause of power quality problems (Sankaran, C. ,1995-10-01). "Effects of harmonics on power systems’). Harmonics can be measured up to 63rd harmonic in modern test equipment.
Figure 2.1 shows the graph of a current wave affected by harmonic distortion.
Figure 2.1
Example of a current containing harmonics and expansion of the overall current into
its harmonic orders 1 (fundamental), 3, 5, 7 and 9 (Schneider Electric Industries SAS, 2008).
The frequency spectrum is a practical graphical means of representing the harmonics contained in a periodic signal. The graph shows the amplitude of each harmonic order. This type of representation is also referred to as spectral analysis. The frequency spectrum indicates which harmonics are present and their relative importance (Schneider Electric Industries SAS, 2008).
Figure 2.2 shows the frequency spectrum of the signal presented in figure 2.1.
Figure 2.2
Spectrum of a signal comprising of a 50 Hz fundamental and harmonic orders 3 (150 Hz), 5 (250 Hz), 7 (350 Hz) and 9 (450 Hz) (Schneider Electric Industries SAS, 2008).
Fourier theory tells us that any repetitive waveform can be interpreted in terms of summing sinusoidal waveforms which are integer multiples (or harmonics) of the fundamental frequency. For the purpose of a steady state waveform with equal positive and negative half-cycles, the Fourier series can be expressed as follows (Allen-Bradley, 2001):
f(t)= ‘_(n=1)^”A_n .sin”(nt/T)’ (2.1)
Where
f(t) is the time domain function
n is the harmonic number (only odd values of n are required)
An is the amplitude of the nth harmonic component
T is the length of one cycle in seconds
Power sources behave as non-linear loads, drawing a distorted waveform that contains harmonics. These harmonics can create problems ranging from telephone transmission interference to degradation of conductors and insulating material in motors and transformers. Therefore it is necessary to measure the total effect of these harmonics. The summation of all harmonics in a system is known as total harmonic distortion (THD).
Total harmonic distortion, or THD, is the summation of all harmonic components of the voltage or current waveform compared against the fundamental component of the voltage or current wave (apt associated power technologies, 2011):
‘THD’_i= ‘(‘_(h=2)^”I_harmonic^2 )/I_(1(fundamental)) (2.2)
‘THD’_v='(‘_(h=2)’V_harmonic^2 )/V_(1(fundamental)) (2.3)
The formulae for the calculation for THD on a current and voltage signal are shown above. The final result is a percentage comparing the harmonic components to the fundamental component of a signal. The higher the percentage, the more distortion that is present on the mains signals (apt associated power technologies, 2011).
The voltage THDv shows the distortion of the voltage wave. A THDv value of less than 5 % is considered normal and there is virtually no risk of equipment malfunctions (Schneider Electric Industries SAS, 2008). A THDv value between 5 % and 8 % shows significant harmonic distortion. Malfunctioning of equipment may take place. A THDv value greater than 8 %shows high harmonic distortion. Again equipment malfunctions are possible.
The current THDi displays the distortion of the current wave. The current THDi must be measured on the incomer and the outgoers of the different circuits in order to identify the load causing the disturbance. The measured THDi can give information on phenomena noticed in the installation. A THDi value of less than 10 % is considered normal and there is virtually no risk of equipment malfunctions (Schneider Electric Industries SAS, 2008). A THDi value between 10 % and 50 % shows significant harmonic distortion. A rise in temperature is possible thus cables and sources must be oversized. A THDi value more than 50 % shows high harmonic distortion. There is possibility of malfunctions of equipments. Therefore an attenuation system must be set.
2.2 Causes of Harmonics
Harmonics are caused by non-linear loads, that is loads that draw a non-sinusoidal current from a sinusoidal voltage source (Allen-Bradley, 2001). The two categories of non-linear loads are single-phase load and three phase load. The examples of single-phase loads are the switch mode power supplies (SMPS), the electronic fluorescent lighting ballasts and the small uninterruptible power supplies (UPS). Variable speed drives (VSD) and the large uninterruptible power supply (UPS) units are examples of three phase loads.
Most power systems can contain a certain level of harmonic currents. However when harmonics become a significant component of the overall load the systems will experience problems. These higher frequency harmonic currents when flowing through the power system can cause communication errors, overheating and hardware damage such as:
Overheating of electrical distribution equipment, cables, transformers, standby generators etc.
Harmonic resonance causes high voltages and circulating currents.
Malfunctioning of equipment due to excessive voltage distortion.
False tripping of branch circuit breakers.
Component failure and shortened life span caused by increased internal energy losses in connected equipment.
Metering errors.
Fires in wiring and distribution systems.
Generator failures.
Crest factors and related problems.
Low system power factor, resulting in penalties on monthly utility bills.
2.2.1 Switch Mode Power Supplies (SMPS)
Many modern electronic units make use of a switch mode power supplies (SMPS). In the past alternating current was converted to direct current by step-down transformer and rectifier. But nowadays it is replaced by direct controlled rectification of the supply to charge a reservoir capacitor to generate a direct current to the load depends on the voltage and current at the output of the load. The advantages of switch mode power supplies are that the power unit can be made in almost any required form factor, reduced size, less cost, and light weight. The drawback of using switch mode power supplies is that the power unit will draw pulses of current that contain high amount of 3rd harmonic and significantly higher frequency components as shown in Figure 2.2 below. To bypass the high frequency components from line and neutral to ground a simple filter is introduced at the supply input.
2.3 Harmonic Mitigation Techniques
The solutions to compensate for and mitigate harmonics in the power system, with varying degrees of effectiveness and efficiency are:
Oversize the neutral wiring
In modern facilities, the neutral wiring should have the same capacity as the power wiring or larger. For a call center, the neutral wiring must be specified such that it exceeds the phase wire capacity by about 200 percent. Careful attention should be paid to wiring in office cubicles. This approach does not protect the transformers; it protects only the building wiring.
Use separate neutral conductors
On three-phase branch circuits, separate neutral conductors are used for each phase conductor instead of installing a multi-wire branch circuit sharing a neutral conductor. Thus the capacity and ability of the branch circuits to handle harmonic loads are increased. This approach successfully removes the addition of harmonic currents on the branch circuit neutrals, but the panel board neutral bus and feeder neutral conductor must still be taken into consideration.
Use DC power supplies, which is not affected by harmonics
In the typical data center, the power distribution system uses a transformer to convert 480-volt AC utility power to 208-volt power that feeds racks of servers. This AC input is converted into DC voltage by the power supplies within each server. The internal power supplies generate substantial heat and are not energy efficient. The heat dissipation causes the number of servers to be limited in a data center. According to a recent article in Energy and Power Management magazine, ”Computers and servers equipped with DC power supplies instead of AC power supplies produce 20 to 40percent less heat, reduce power consumption by up to 30 percent, increase server reliability ,offer flexibility to installations, and experience decreased maintenance requirements.’ That sounds good but the conversion of AC to DC power is not justified when considering the cost, compatibility, reliability and efficiency for most data centers.
Use K-rated transformers in power distribution components
A standard transformer cannot be used when high harmonic currents produced by non-linear loads flow in the power system. The transformer will overheat and fail when connected to these loads. The K-rated transformers were developed when harmonics were introduced into electrical systems at levels that showed detrimental effects. The K-rated transformers are used to handle the heat produced by harmonic currents and are very efficient when used under their k-factor value. The range of the k-factor is between 1 and 50. If the k-factor is high, the transformer is able to handle more heat from the harmonic currents. The value of the k-factor must be properly chosen as it affects cost and safety.
Passive filters
The passive filters use resistors, inductors and capacitors and they do not depend on any type of external power source. Passive filters make use of devices that provide low-impedance paths to divert harmonics to ground and devices with high-impedance paths to reduce the flow of harmonics. The passive filters cannot adapt to changes in the electrical systems. Thus they can be overloaded or create resonances when changes are made in the electrical system (for example the addition of more non-linear loads). This causes harmonics to amplify rather than diminish.
Passive filters are simple and less expensive but inherit some shortcomings. The filter components are quite big because the harmonics that need to be suppressed are of the low order. Furthermore the filtering characteristics are affected by frequency variation of the power distribution system and tolerances in components. If the frequency variation is high, the size of the components becomes impractical. Therefore passive filters are not able to mitigate harmonics in the electrical systems, thus retrofit of new filters are required.
Active power filters
Active power filter is a better solution to reduce harmonics compared to the other solutions. It has many advantages even though it is relatively new and costly. Active power filters have no resonance problems and are inherently current limiting. They can be configured to mark specific harmonics or to improve the full spectrum of harmonics.
Advantages and Disadvantages of Active filter over Passive filter
Advantages of Active Filter over Passive Filter:
Active filters can be used to minimize the effects of harmonics of more than one order.
Active filters are also useful in flickering problems that are caused in the power system.
Disadvantages of Active Filter over Passive Filter:
Active filters are more expensive than the passive filters
Active filters cannot be used for small loads in a power system
Due to the presence of harmonics in both current and voltage, active filter may not be able to solve the problem in certain typical applications.
2.4 Harmonics refer to IEEE standard
Harmonics are sinusoidal voltage or current that contains the integer of multiple frequencies. They can produce excessive heating and pulsating and minimized torque in motors and generators. In capacitors they cause increased heating and voltage stress. Harmonics also cause malfunctioning in electronics, switch gear and relaying. In short, if equipment is not properly rated and applied and if the system is designed without consideration for harmonics, then the harmonics can lead to reduced equipment life.
However it is important to measure and limit harmonics in electric power systems. Thus a basis for limiting harmonics is provided by IEEE Std 519-1992, IEEE Recommended Practices and Requirements for Harmonics Control in Electric Power Systems (IEEE 519). The document defines the limits of harmonics but there are some applications issues that the reader need to focus on and use his or her own judgment.
The aim of IEEE 519 is to approve limits on harmonic distortion based on the two basic criteria namely:
A limitation is placed on the amount of harmonic current that a consumer can introduce in a utility network.
There is a limitation on the level of harmonic voltage that a utility can provide to a consumer.
2.5 Effects of Harmonics
Power system problems associated to harmonics are rare but a number of undesirable effects can occur. Effects such as increased transformer, capacitor, motor or generator heating, misoperation of electronic equipment (which relies on voltage zero crossing detection or is sensitive to wave shape), incorrect readings on meters, misoperation of protective relays, interference with telephone circuits, etc are caused by high levels of harmonic distortion. However if a resonant condition occurs the likelihood of such ill effects occurring is greatly increased. Resonance arises when a harmonic frequency generated by a non-linear load closely coincides with a power system natural frequency. Parallel resonance and series resonance are the two forms of resonance which can occur.
Parallel resonance takes place when the natural frequency of the parallel combination of capacitor banks and the system inductance drop near a harmonic frequency. This can produce substantial amplification of the harmonic current that flows between the capacitors and the system inductance and cause blowing or failure of the capacitor fuse or transformer overheating.
Series resonance is a outcome of a series combination of inductance and capacitance and gives a low impedance path for harmonic currents at the natural frequency. A high voltage distortion level between the inductance and capacitance is the effect of a series resonance.
Table 1 – Description of the harmonic effects on power system components
Component Effect of harmonics
Power generators – additional heating
– rotor heating
– appearance of pulsation torque
– increase in torsion force
Transformer and
inductors winding stray losses
– hysteresis losses increase
– additional heating
Conductors and
Cables – skin and proximity effect
– additional heating
– dielectric puncture
Circuit breakers
and fuses – skin and proximity effect
– additional heating due to overload
– failure in protection
– changes in rise-time and fall-time
Relays – changes in delay characteristic
– false tripping
Motors – increase of stator and rotor losses
– additional leakage field losses
– shaft torque oscillation
Capacitors -dielectric losses increase
– resonant over-voltage
– life expectancy decrease
Instrumentation
and electronic
equipment – possible erroneous operation
– erroneous measurement
Communication
and IT – interference over long wires
– data loss incidents
CHAPTER 3
DESIGN OF ACTIVE POWER FILTERS FOR HARMONIC MIGITATION FOR SINGLE PHASE SYSTEM
3.1 Active power filtering techniques
Active power filter (APF) is the suitable solution to suppress harmonics and compensate the reactive power. The APF use an inverter to inject currents or voltages harmonic components to cancel the load harmonic components. In order to achieve power factor correction (PFC), active power filter uses harmonic or current injection. Power factor correction (PFC) techniques use both passive and active solutions for removing harmonic distortion and improving power factor.
Shunt APF is the more usual configuration to introduce current harmonics into the point of common coupling (PCC). The APF can be placed in a low voltage power system to lower one or more loads, thus the propagation of current harmonics in the system are avoided. APF is also called active power line conditioners (APLC) since it compensates the reactive power and cancels the harmonics.
The three main characteristics of an active power conditioner are:
The configuration of power converter (the scheme and the topology of converter and the electronics device used).
The control strategy (the calculation of APLC control reference signals).
The control method used (how the power inverter follows the control reference).
APF’s can be categorized based on converter type, topology, and the number of phases. The converter type is mainly two types:
Voltage source inverter (VSI)
Current source inverter (CSI)
The Voltage source inverter (VSI) is the most commonly used due to its popularity and simple in implementation. It is used to suppress harmonics current drawn by the non-linear loads. However, current source inverters (CSI) are used for the mitigation of voltage harmonics caused by non-linear loads.
The topology of active power filter is classified in to three types:
Series active power filters
Shunt active power filters
Hybrid active power filters
Basic configuration of single-phase shunt active power filter
Before non-linear loads are connected to the system, the source current and voltage are purely sinusoidal and in phase to each other. Single-phase shunt active power filter consists of two main parts, the power circuit and the control circuit. The power circuit comprises of non-linear load, IGBTs or MOSFETs switching semiconductor devices filter inductor, and DC side capacitor. The control part is the brain of the active filter, which saved as the control algorithm for the filter in order to mitigate the harmonics caused by the non-linear loads within the system. The Figure below shows the SAPF configuration.
The control circuit is the heart of the APF and is implemented in three stages. In the first stage, power transformers (PT’s or CT’s), Hall-effect sensors, and isolation amplifiers are used to sense the essential voltage and current signals to assemble accurate system information. In the second stage, compensating commands in terms of current and voltage levels are obtained
based on control methods and APF configurations. PWM, hysteresis, sliding-mode,
or fuzzy-logic-based control techniques are used to generate the gating signals for the solid-state devices of APF in the third stage of control. The important part of APF control is the development of compensating signals either in terms of voltages or currents and this affects their rating and transient, as well as steady-state performance. The control strategies to produce compensation commands are based on frequency-domain or time-domain correction techniques. Control strategy in the frequency domain is based on the Fourier analysis of the distorted voltage or current signals to remove compensating commands. Control methods of the APF’s in the time-domain are based on instantaneous derivation of compensating commands in the form of either voltage or current signals from distorted and harmonic-polluted voltage or current signal. Many theories have been developed in time domain and frequency domain.
3.2 Shunt Active power filter Implementation
The basic operating principle of shunt active power filter is shown in the schematic diagram below. The principle of shunt APF is to produce specific harmonic current components to cancel the harmonic current components generated by the load.
Figure 3.2 Generalized block diagram for APF.
All the necessary information concerning the harmonic currents and other system variables are passed to the compensation current/voltage reference signal estimator. The compensation reference signal from the estimator drives the overall system controller. However this produces the control for the gating signal generator. The power circuit is controlled by the output of the gating signal generator. Eventually, the power circuit in the generalized block diagram can be configured into different connections depending on the interfacing inductor/transformer used.
Shunt APF is the most important configuration commonly used in active filtering applications for lowering current harmonic and improving power factor. A shunt APF is made up of a controllable voltage or current source inverter. The voltage source inverter (VSI) based shunt APF is the most widely employed type, due to its simple installation procedure. SAPF functions as harmonic current source which introduce an opposite but equal magnitude of the harmonic and reactive current as that of nonlinear load. As a result components of harmonic currents produced by non-linear loads are cancelled and the source current remains sinusoidal and in phase with the respective phase to neutral voltage.
The principle configuration of a VSI based shunt APF is shown in Fig. 3.3. It is made up of a three leg six-switch bridge fixed at the point of common coupling (PCC) through an interfacing inductor (Lf) and with capacitor (Cf) connected on DC-side. A three phase diode bridge rectifier with an inductive load on DC side is taken as nonlinear load.
Figure 3.3 Principle configuration of a VSI based shunt APF.
Total instantaneous power drawn by the nonlinear load can be represented as:
pL(t) = pf (t) + pr (t) + ph (t)
Where,
pf(t) – instantaneous fundamental (real) power absorbed by the load,
pr(t) ‘ instantaneous reactive power drawn by the load, and
ph(t) ‘ instantaneous harmonic power drawn by the load.
Active power filter must supply all the reactive and harmonics power demand of the load in order to achieve unity power factor operation and drawing sinusoidal currents from the utility. At the same time, active filter will draw real component of power (PLoss) from the utility, to provide switching losses and to keep in the DC link voltage unchanged. Active power filters should supply power components (reactive and the harmonic) i.e.
pc (t) = pr (t) + ph (t)
Mathematical model for the current sensing stage
From the single line diagram shown in fig.3.4
is(t) = iL(t) – ic(t) (1)
Where,
is(t), iL(t), ic(t) are the instantaneous value of source current, load current and the filter current.
And the utility voltage is given by
vs(t) = Vmsint (2)
Where,
vs(t) ‘ is the instantaneous value of the source voltage, Vm – is the peak value of the source voltage.
If non-linear load is connected then the load current will have a fundamental component and the harmonic components which can be represented as
i_L (t)= ‘_(n=1)^”’I_(n ) sin”(nwt+’_n)’ ‘
= I_(1 ) sin”(wt+’_1 )+’_(n=2)^”’I_n sin”(nwt+’_n)’ ” (3)
Where, I1 and ‘_1 are the amplitude of the fundamental current and its angle with respect to the fundamental voltage, and In and ‘_nare the amplitude of the nth harmonic current
and its angle.
Instantaneous load power pL(t) can be expressed as’
pL(t) = vs(t) iL(t)
= V_m sin”wt I_1 sin”(wt+’_1)’ ‘+ V_m sin”wt ‘_(n=2)^”’I_n sin”(nwt+’_n)’ ”
= pf (t) + pr (t) + ph (t) (4)
= pf (t) + pc (t) (5)
In the equation (4) and (5) pf (t) is the real power (fundamental),
pr (t) represents the reactive power and ph (t) represents the harmonic power drawn by the load.
For ideal compensation only the real power (fundamental) should be supplied by the source while all other power components (reactive and the harmonic) should be supplied by the active power filters i.e.
pc (t) = pr (t) + ph (t)
.
The total peak current supplied by the source
Imax = Ism + IsL (6)
Where,
I_sm= I_1 cos”’_1 ‘
and IsL is the loss component of current drawn from the source. If active power filter provide the total reactive and harmonic power, then is(t) will be in phase with the utility and pure sinusoidal. At this time, the active filter must supply the following compensation current:
Ic(t) = IL(t) ‘ is(t) (7)
It is very important to calculate the accurate value of the instantaneous current supplied by the source for the accurate and instantaneous compensation of reactive and harmonic power.
Is(t) = Imax sint (8)
Where, Imax (= I1 cos’_1+ IsL) is the amplitude of the desired source currents. The phase angles can be found from the source voltages. Hence, the waveform and phases of the source currents are known and only the magnitude of the source currents needs to be found. The peak value or the reference current Imax is gauged by adjusting the DC link voltage of the inverter. This DC link voltage is compared by a reference value and a PI controller processes the error. The output of the PI controller is taken as the amplitude of the desired source currents and the reference currents are approximated by multiplying this peak value with the unit sine vectors in phase with the source voltages.
CONTROL STRATEGIES FOR LOW VOLTAGE INVERTER BASED SAF
The control methodology of APF is implemented through two steps. First step is estimation of reference compensating signal and the second step is generation of firing signals for switching devices of VSI. [1]
Estimation of Reference Compensating Signals
The most important part of the active filter control is the estimation of compensating signal. It has great effect on compensation objectives, rating of active filter and its transient as well as steady state performance. In order to gather accurate information about the system variables, the reference signal estimation is initiated through the detection of essential voltage/current signals. AC source voltage and DC-bus voltage are the voltage variables of the APF. Typical current variables are load current, AC source current and compensation current. Based on these system variables compensation reference signal in terms of voltage/current is estimated in frequency-domain approach or in time-domain approach.
Frequency-domain approach is obtained from the principle of Fourier analysis in which Fourier Transform is used to the captured voltage/current signal. The fundamental components are eliminated in order to separate the harmonic components of the captured voltage/current signal. Inverse Fourier Transform is then used to find the compensation reference signal in time domain.
Time-domain approaches are based on instantaneous estimation of reference signal in the form of either voltage or current signal from distorted and harmonic-polluted voltage and current signals [2]. These approaches can be used for both single- phase and three-phase systems.
Generation of Firing Signals for Switching Devices of VSI
The objective of APF control is to produce appropriate gating signals for the switching devices based on the estimated compensation reference signal. The work of an APF is influenced by the selection of control techniques. Therefore to achieve a satisfactory APF performance, the choice and implementation of the control technique is very important. This section briefly describes some of the frequently used control techniques and their basic features.
Linear Control Technique
This technique is executed by using a negative-feedback system as shown in Fig. 2.7. In this technique, the compensation current (if) or voltage (vf) signal is correlated with its estimated reference signal (if’ or vf’) through the compensated error amplifier to generate the control signal. A comparison of the resulting control signal with a saw tooth signal is made through a pulse width modulation (PWM) controller to produce appropriate gating signals for the switching transistors. The switching frequency is established by the frequency of the repetitive saw tooth signal. This frequency is constant in linear control technique.
Fig. 2.7 Block diagram of linear control technique.
When the control signal has a higher numerical value than the saw tooth signal and vice versa, the gating signal is set high as shown in figure 2.8 below. A faster response is obtained and a simpler implementation is made with analogue PWM circuit. However due to inherent problem of analogue circuitry, the linear control technique has a poor harmonic compensation performance. This is mainly due to the constraint of the attainable bandwidth of the compensated error amplifier.
Fig. 2.8 Gating signal generation by linear control.
Hysteresis Current Control Technique
This technique establishes a bang-bang type instantaneous control that pushes the APF compensation current (if) or voltage (vf) signal to follow its estimated reference signal (if’ or vf’) within a certain tolerance band. The block diagram of the control scheme is shown below. In this control scheme, a signal deviation (H) is imposed on if’ or vf’ to create the upper and lower limits of a hysteresis band. The if or vf is then calculated and compared with if’ or vf’ ; the resulting error is subjected to a hysteresis controller to find out the gating signals when surpasses the upper or lower limits set by (estimated reference signal + H/2) or (estimated reference signal – H/2). No switching action is taken as long as the error is within the hysteresis band. Switching takes place whenever the error hits the hysteresis band.
Fig. 2.9 Block diagram of hysteresis control technique.
Fig. 2.10 Gating signal generation by hysteresis current controller.
The APF is therefore switched in such a way that the peak-to-peak compensation current/voltage signal is restricted to a particular band determined by H as illustrated in the figure above.
In this specific work, a hysteresis current controller with a fixed H is implemented. To get a compensation current (if ) with very small switching ripples, the value of H can be decreased. However, this result in higher switching frequency and switching losses are increased. The excellent dynamic performance and controllability of the peak-to-peak current ripple within a specified hysteresis band makes the hysteresis current controller beneficial. Furthermore, the implementation of this control scheme is straightforward; this is apparent from the controller structure shown in Fig. 2.9. The technique of generating gating signals using hysteresis controller is shown in Fig. 2.10. However, this control scheme shows some inadequate features. The main disadvantage is that it creates uneven switching frequency which influences the APF efficiency and reliability. Due to its excellent dynamic performance, controllability and simplicity this technique is chosen in this project for producing gating signals for SAF.
DC Bus Voltage Control
The DC bus voltage must be kept at a constant value otherwise the source current will change and lapse from sinusoidal waveform. Various types of controllers like Proportional-Integral (PI), adaptive, Neuro and Fuzzy Logic Controller (FLC) for DC bus voltage regulation are well presented in literature. PI, PID, fuzzy logic based controllers are used for DC bus voltage control of shunt active power filter.