IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 23, NO.

1, JANUARY 2008

19

A Five-Level Symmetrically Dened Selective Harmonic Elimination PWM Strategy: Analysis and Experimental Validation
Vassilios G. Agelidis, Senior Member, IEEE, Anastasios I. Balouktsis, Member, IEEE, and Mohamed S. A. Dahidah, Member, IEEE

AbstractA ve-level symmetrically dened multilevel selective harmonic elimination pulsewidth modulation (MSHEPWM) strategy is reported in this paper. It is mathematically expressed using Fourier-based equations on a line-to-neutral basis. An equal number of switching transitions when compared against the well-known multicarrier phase-shifted sinusoidal PWM (MPSSPWM) technique is investigated. For this paper, it is assumed that the four triangular carriers of the MPSSPWM method have nine per unit frequency resulting in seventeen switching transitions for every quarter period. For the proposed MSHEPWM method, this allows control of sixteen harmonics and the fundamental. It is conrmed that the proposed MSHEPWM offers signicantly higher converter bandwidth in the standard range of the modulation indices. Moreover, when the bandwidth is reduced to be equal with the one offered with the MPSPWM, the modulation index can be increased resulting in a higher gain and at a reduced switching frequency overall. Selected solutions for the switching transitions are presented and veried experimentally in order to conrm the effectiveness of the proposed technique. Index TermsMultilevel Converter, optimization, phase-shifted sinusoidal pulsewidth modulation (PSPWM), pulsewidth modulation (PWM), selective harmonic elimination (SHE).

Fig. 1. Generalized staircase waveform suitable for multilevel systems and related angles of transition between the various voltage levels.

I. INTRODUCTION ELECTIVE harmonic elimination pulse-width modulation (SHEPWM) techniques have been mainly developed for two or three-level schemes [1][12]. The main challenge associated with such techniques is to obtain the analytical solutions of the non-linear transcendental equations that contain trigonometric terms which naturally exhibit multiple sets of solutions [5]. There have been many approaches to this problem reported in the technical literature including: sequential homotopy-based computation [6], resultants theory [7], optimization search [8],

Manuscript received April 20, 2006; revised April 25, 2007. This paper was presented in part at the IEEE PESC05, Recife, Brazil, June, 2005. Recommended for publication by Associate Editor A. Trzynadlowski. V. G. Agelidis is with the School of Electrical and Information Engineering, The University of Sydney, Sydney NSW 2006, Australia (e-mail: v.agelidis@ee. usyd.edu.au). A. I. Balouktsis is with the Department of Informatics and Communications, Technological Institution of Serres, Terma Magnesias, Serres 62124, Greece (e-mail: tasosb@teiser.gr). M. S. A. Dahidah was with the Faculty of Engineering, Multimedia University, Selangor, Malaysia. He is now with the School of Electrical and Electronic Engineering, The University of Nottingham (Malaysia Campus), Selangor, Malaysia (e-mail: mohamed.dahidah@nottingham.edu.my). Color versions of one or more of the gures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identier 10.1109/TPEL.2007.911770

Walsh functions [9], [10] and other optimal methods [11] including genetic algorithms (GAs) [12], [13]. The bipolar waveform when symmetry is requested through the denition of the problem has been treated in detail in [14] where a minimization technique is employed along with a biased optimization search method [8] to get the multiple sets as predicted in [6]. Interestingly, symmetry, which was widely assumed in the past in all reported methods and solutions, can be relaxed and this results in different solutions and a more generic way to dene the problem as reported in [11], [15]. Recently, solutions trajectories of the harmonic elimination problem were mathematically addressed in [16]. On the other hand, multilevel converters based on solid-state technologies have been investigated for more than three decades. Initially, when the switching frequency was restricted to line frequency, the generic question associated with the SHEPWM approach has mainly been the way the staircase multilevel waveform is generated in order to control the amplitude of the fundamental frequency and eliminate the maximum number of harmonics from the waveform [17]. A generalized staircase waveform suitable for a multilevel converter is shown in Fig. 1, where the transition angles are linked with the level change. In a generic denition this does include equal and non-equal dc levels [13], [18]. Recently, a number of technical papers have appeared addressing the multilevel waveform using similar theories used previously for the bipolar (two-level) and unipolar (three-level) waveforms. Specically, the theory of resultants and its performance for a multilevel staircase waveform was reported in [18]. A unied approach was presented in [19]. More recently, the use of symmetric polynomials is combined with the resultant theory for a multilevel converter [20]. Previous work [18] has

0885-8993/$25.00 2007 IEEE

20

IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 23, NO. 1, JANUARY 2008

shown that the transcendental equations characterizing the harmonic content can be converted to polynomial equations which are then solved using the method of resultants from elimination theory. A difculty with this approach [18], as suggested in [20], is that when there are several dc sources, the degrees of the polynomials are quite large, thus making the computational burden of their resultant polynomials (as required by elimination theory) quite high. An interesting method to overcome the previously mentioned drawbacks was reported in [20], where the theory of symmetric polynomials is exploited to reduce the degree of the polynomial equations that must be solved which in turn greatly reduces the computational burden. The use of optimization techniques [8] including GAs [12], [13] has been shown to overcome all known obstacles of previous approaches. On the topic of optimization techniques, interesting theoretical ideas were reported in [31][34]. Specically, reference [31] deals with the single-phase inverter optimal PWM where the standard mathematical formulation of the problem can be reformulated and that the sought solution can be found by computing the roots of a single invariable polynomial for which algorithms are readily available. The approach is similar to the one using polynomial approximation reported in [7]. In reference [32] a simplex homotopic xed-point algorithm for the computation of optimal PWM pattern was reported. Regarding computational algorithms, in [33], an algorithm was proposed that searches solutions so that two consecutive angles are placed way apart to allow easy implementation. The approach used in [33] is as follows; the problem is rst transformed inot a constraint optimization problem and then used differential evolution algorithm to nd the roots with the necessary distance separation. A new algorithm was presented in [34] for pushing the rst crest of the surplus harmonics backwards, ameliorating the amplitude frequency spectrum distribution of the output waveform and therefore reducing the impact of surplus harmonics in programmed PWM techniques. Looking at the SHEPWM problem as an optimization problem rather than an elimination one naturally creates new opportunities to extend SHEPWM methods to a new territories and the method reported here is one example. Finally, an interesting hybrid method between a carrier based modulation and SHE is proposed where the harmonic content and the switching angles are both controlled through the control of the frequency of the carrier waveform avoiding the need for solving the transcendental equations all together and obtaining at the same time the required switching angles [35]. On the other hand, multilevel SHEPWM systems have been controlled using the unipolar approach, where the waveform takes a positive, a negative and a zero value, and phase-shifted techniques are used to build multilevel systems [21]. Other approaches have also been reported including one where the harmonic elimination is combined with a programmed method [22] and another where a criterion based on power equalization between various cascaded connected H-bridge multilevel converters is used to obtain the angles of the harmonic elimination method [23]. The multicarrier phase-shifted sinusoidal PWM (MPSSPWM) has been used to increase the bandwidth of multiple connected converters. This can be traced back in the 1980s in a conference paper [24] and then in a journal publication which appeared in [25]. It was then used in many other works reported in the technical literature as it provided an

Fig. 2. Five-level symmetrically dened (line-to-neutral) MSHEPWM wave5, form shown for a distribution ratio of 5/12 ( 12, and 17).

k= m=

N = k+m =

opportunity to cancel a number of harmonics if the phase-shift between the numerous carriers within the same converter and among the carriers controlling the other converters is chosen carefully [28]. However, although the MPSSPWM technique has been extensively used for multilevel systems, no comparison with a MSHEPWM technique has been reported to clarify the performance of each method against each other. For instance, higher gain and improved bandwidth for the programmed SHEPWM techniques have been reported but these gains were analyzed only for the two-level systems [4]. So far, there exists limited information reported if such superior performance can also be attributed to the MSHEPWM methods. Moreover, thus far there has been limited documentation regarding a complete set of angles for relatively high frequency multilevel waveform for all modulation indices, except references [22], [26], [27]. The objective of this paper is to propose a symmetrically dened MSHEPWM strategy for a ve-level waveform as shown in Fig. 2. The various switching transitions as a function of the modulation index are also presented for the standard modulation range, and in the overmodulation region where the bandwidth of the strategy is reduced to be able to obtain higher amplitude for the fundamental component. The proposed technique is then compared against the well-known MPSSPWM technique using four carriers in order to create a ve-level line-to-neutral switching pattern [26]. Selected experimental results are also presented to conrm the validity of the proposed method. The paper is organized as follows. Section II presents in detail the proposed symmetrically dened ve-level MSHEPWM in both the standard modulation index region and in the overmodulation one as well. In Section III the MPSSPWM technique suitable for the ve-level system is also briey presented. The comparison between the two techniques along with selected experimental results taken from a low power laboratory prototype are presented and discussed in Section IV. Finally, conclusions are summarized in Section V. II. PROPOSED FIVE-LEVEL MSHEPWM STRATEGY The proposed symmetrical ve-level MSHEPWM strategy is dened according to the waveform shown in Fig. 2 and represents the line-to-neutral waveform of the three-phase multilevel converter. The number of levels of the waveform is assumed to be ve, i.e., 1 p.u., 2 p.u., 0 p.u., 1 p.u. and 2 p.u. Let be

AGELIDIS et al.: FIVE-LEVEL SYMMETRICALLY DEFINED SELECTIVE HARMONIC ELIMINATION PWM STRATEGY

21

the number of total switching transitions (angles) of the waveform sought within the quarter of the period of the waveform. This number can be either odd or even. However, there are reis chosen to be strictions once it is chosen. For this paper, an odd number. Let be the number of the switching transitions placed between the 0 p.u. level and the 1 p.u. This number can only be an odd number since the rst switching transition is chosen to be from 0 p.u. to 1 p.u. level. Let be the number of the remaining switching transitions placed between 1 p.u. and 2 p.u. levels. This number can be either even or odd depending upon . In this case, since has chosen to be odd and is only odd, can only be even. Then, for the proposed MSHEPWM strategy, can only be an odd number

TABLE I SUMMARY OF STANDARD MODULATION RANGE REGION WHERE SOLUTIONS FOR ALL SWITCHING ANGLES EXIST AS A RELATIONSHIP TO THE VARIABLE RATIO OF DISTRIBUTION OF THE SWITCHING ANGLES 17 BETWEEN THE MULTIPLE LEVELS (k=m) N

1,3 can only be an even number since an odd number, hence

(1) has been chosen to be (5)

(2) and the total sum of both and must always be equal to the maximum number of switching transitions, hence where (3) For any ratio of then a different set of transcendental equations describing the Fourier equations linked to the amplitude of the harmonics that can be eliminated needs to be written down. The idea here is the classic SHEPWM method that tries to nd switching angles in order to eliminate a number of harmonics and control simultaneously the fundamental component. The challenge of the proposed method is that it is applied to a true multilevel waveform, i.e., ve-level waveform (Fig. 2). The typical half-wave and quarter-wave symmetries are respected for the waveform, i.e., simply when the switching angles for all modulation indices are obtained for the angles between zero and 2, the usual reection occurs to nd the rest of the angles. Since there are switching angles (i.e., , where ), 1 harmonics can be eliminated if solutions can be found. For a three-phase inverter, the non-triplen odd harmonics can be eliminated from the wave3 2) form (i.e., fth, seventh, 11th, 13th, -th where and then the strategy relies on the structure of the power circuit in order to remove the triplen ones from the line-to-line voltage waveforms. In a generalized form, the set of equations that need to be solved is as follows: (7) (8) is the amplitude of the fundamental component to be If generated, then (9) 2 the square-waveform of 2 p.u. amplitude can genwhen erate 8 per unit maximum value at fundamental frequency. The minimization technique proposed in [8] has been applied and software is used to investigate the proposed method [29]. The distribution ratio varies for each case where a new set of equations describing the new waveform are written. The equations relating to the minimization of the transcendental equations describing the Fourier coefcients are then solved. The standard modulation regions where solutions exist for a given distribution ratio are summarized in Table I. For this paper, 17. However, this number and the various ratios can be changed as desired and the minimization method [8] would provide the respective solutions provided they exist. The overmodulation range is also discussed later. The constraint for the solutions sought is as follows: (10) A. Standard Modulation Range (4) 17 (6)

In this paper, a selected set of solutions for every region are reported although many more might exist. Therefore, it is beyond the scope of this paper to treat the cases where multiple

22

IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 23, NO. 1, JANUARY 2008

TABLE II SUMMARY OF THE OVERMODULATION REGIONS WHERE SOLUTIONS FOR ALL SWITCHING ANGLES EXIST AS A RELATIONSHIP TO THE VARIABLE RATIO OF DISTRIBUTION OF THE SWITCHING ANGLES BETWEEN varies THE MULTIPLE LEVELS (k=m) N

Fig. 3. (Continued) Standard modulation region: Switching angles in degrees versus modulation index M for the various ratios of k=m and cases which are also summarized in Table I for N = 17. (i) Case VII: 5/12 (1.43 M 1.56, Set 2). (j) Case VII: 5/12 (1.29 M 1.42, Set 3) (k) Case VII: 3/14 (1.1 M 1.31, Set 4). (l) Case VIII: 3/14 (1.38 M 1.51, Set 1). (m) Case VIII: 3/14 (1.46 M 1.59, Set 2). (n) Case IX: 1/16 (1.37 M 1.6).

Fig. 3. Standard modulation region: Switching angles in degrees versus modulation index M for the various ratios of k=m and cases which are also sum0.46, Set 1). (b) marized in Table I for N 17. (a) Case I: 17/0 (0 M Case I: 17/0 (0 M 0.9, Set 2). (c) Case II: 15/2 (0.63 M 0.9). (d) Case III: 13/4 (0.7 M 0.91). (e) Case IV: 11/6 (0.92 M 0.98). (f) Case V: 9/8 (0.96 M 1.04). (g) Case VI: 7/10 (1.04 M 1.36). (h) Case VII: 5/12 (1.39 M 1.46, Set 1.

=       

      

implies that a three-level technique covering this region can be implemented and the result conrms that there are more than one set of solutions. The rst is a discontinuous set and the second covers the entire range. Further results for all combinations of ratios are plotted in Fig. 3 and summarized in Table I. The last non-triplen harmonic that can be eliminated according to (4)(6) is the 49 p.u. and this is conrmed in Fig. 4 where the rst non-triplen signicant harmonic present is the 53 p.u. since the 51 p.u. is a triplen one. Specically, Fig. 4(b) shows that the spectrum of the line-to-neutral voltage waveform has only multiple of triplen harmonics which are cancelled out in a three-phase system for the line-to-line waveforms. This is conrmed and the line-to-line voltage waveform is shown in Fig. 4(c). The bandwidth of the normalized line-to-line voltage waveform is proved to be according to the theory, which is that the harmonics up to the 49 p.u. are all zero. Since the 51 p.u. harmonic is a triplen one, the rst signicant non-triplen harmonic present turns out to be the 53 p.u. [Fig. 4(d)]. B. Overmodulation In order to increase the modulation index for the proposed technique and since the available bandwidth is higher than the one when compared with the MPSPWM technique, the overmodulation region is also investigated and solutions are reported in this paper. It is shown that controlling up to nine non-triplen harmonics results in the rst signicant non-triplen harmonic to be the 31 p.u. The ratio of is changed and the same optimization algorithm is used to seek the switching transitions. Fig. 5 presents the switching transitions in degrees as a function of the modulation index for various combinations. It should

sets of solutions can be found, although some overlap and multiple solutions are also reported. The aim here is to nd solutions that provide PWM waveform realization for the entire region of modulation indices. It should be noted that the result obtained from the proposed method when the ratio is chosen to be 17/0 provides solutions for all modulation indices up to 0.9, which has been reported in other cases to be the maximum attainable value for similar approaches (i.e., the three-level or unipolar waveform) [7]. This

AGELIDIS et al.: FIVE-LEVEL SYMMETRICALLY DEFINED SELECTIVE HARMONIC ELIMINATION PWM STRATEGY

23

Fig. 4. Selected waveforms implemented with the switching angle solutions 17, non-triplen harmonics to be elimigiven by the proposed method, N nated up to 49 p.u. (n = 3N 2), and taken from the case where k=m = 5/12, Set 3). (a) Line-to-neutral voltage waveform (M = 1.5)). (b) Spectrum of the line-to-neutral voltage waveform (M = 1.5). (c) Line-to-line voltage waveforms (M = 1.5). (d) Spectrum of the line-to-line voltage waveform (M = 1.5).

be noted that the number of the overall transitions varies. This investigation is similar to the pulse-dropping methods reported in the past as the lower the number of the angles sought the higher the effective value of the resulting waveform (see Fig. 6). III. MPSSPWM TECHNIQUE The MPSSPWM technique has been used in many applications in order to increase the bandwidth of the system [24], [25]. The harmonics are controlled through the separate SPWM controlled systems but due to the phase-shift effect between the various modulators the overall harmonic spectrum is further improved as the number of carrier waveforms is increased [28]. In this paper, the MPSSPWM technique with four carrier waveforms is considered. This way a ve-level (line-to-neutral) PWM voltage waveform is generated so that to accommodate its comparison with the MSHEPWM technique, presented in Section II. The carrier frequency is chosen to be 9 p.u. Fig. 7 presents this technique. Specically, Fig. 7(a) shows the reference signal (sinusoidal) with the four triangular signals with each having the same frequency (9 p.u.). Each carrier is phaseshifted by 1/4 of its period. This ensures that the line-to-neutral voltage waveform generated and shown in Fig. 7(b) has an increased bandwidth which is equal to four times the per unit frequency. This is the result of the phase-shift introduced which allows the cancellation of the switching frequency harmonics and the associated sidebands. Closer observation of the line-to-neutral waveform reveals that the switching transitions of the waveform for the modulation index shown 1.5 5 (from 0 p.u. level to 1 p.u.) and 12 (from 1 p.u. level to 2 p.u.), respectively. When looking at the spectrum, as the theory of PWM suggests, the rst signicant harmonics will be centered around the 4 9 36 p.u. frequency. For the 50-Hz system shown in Fig. 7, the rst point of interest in the spectrum becomes the 36 p.u. frequency which is 1800-Hz frequency although this frequency is not present in the spectrum. The sidebands are present and these include harmonics of 35 p.u., 33 p.u., 31 p.u., and 29 p.u. This

Fig. 5. Overmodulation region: Switching angles in degrees versus modulation index M for various ratios of k=m and cases which are also summarized in Table II. N also varies, and the method results in reduced switching frequency and reduced bandwidth and higher modulation index value when compared with the standard modulation range. (a) Case I: 1/9 N = 10, 1.42 M 1.56 (Set 1.7 (Set 2). (c) Case II: 1/10, N = 11, 1). (b) Case I: 1/9 N = 10, 1.6 M 1.54 M 1.63 (Set 1). (d) Case II: 1/10, N = 11, 1.54 M 1.68 (Set 1.68. (f) Case IV: 1/12, N = 13, 2). (e) Case III: 1/11, N = 12, 1.56 M 1.18 M 1.47 (Set 1). (g) Case IV: 1/12, N = 13, 1.4 M 1.55 (Set M 1.67 (Set 3). (i) Case V: 1.13, 2). (h) Case IV: 1/12, N = 13, 1.57 N = 14, 1.58 M 1.64. (j) Case VI: 1/14, N = 15, 1.42 M 1.6. M 1.62 (Set 1). (l) Case VII: 1/15, (k) Case VII: 1/15, N = 16, 1.52 N = 16, 1.21 M 1.62 (Set 2).

 

 

 

 

      

       

implies that the most signicant sideband harmonic which is not eliminated due to the PWM switching is the 29 p.u. or 1450 Hz.

24

IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 23, NO. 1, JANUARY 2008

TABLE III COMPARISON BETWEEN THE TWO TECHNIQUES: MPSSPWM AND MSHEPWM

IV. DISCUSSION OF RESULTSEXPERIMENTAL VALIDATION A. MSHEPWM and MPSSPWM Techniques: A Comparison

Fig. 6. Implementation of the proposed MSHEPWM strategy in the over1.7 (Set 2) and k=m 1/9, N 10. (a) modulation region where M Line-to-neutral voltage waveform. (b) Spectrum of the line-to-neutral voltage waveform (9 non-triplen harmonics eliminated with the rst non-triplen harmonic present being the 31 p.u.). (c) Line-to-line voltage waveform. (d) Spectrum of the line-to-line voltage waveform (the rst non-triplen harmonic present being the 31 p.u.).

Fig. 7. Multicarrier phase-shifted sinusoidal PWM technique for ve-level line-to-neutral switching pattern using four carriers (M = 1.5). (a) Reference and four carrier triangular signals with 9 p.u. frequency. (b) Line-to-neutral switching pattern directly controlled by the comparison of the signals. (c) Spectrum of the line-to-neutral waveform showing the bandwidth being almost four times the 9 p.u. as the theory suggests (around 36 p.u. minus the sidebands).

In this paper, the two techniques have been compared in order to identify any potential benets from using the proposed MSHEPWM. The MPSSPWM technique is obviously easier to implement as the switching transitions are directly controlled by the comparison of signals which can be easily varied to control the modulation index. The MSHEPWM offers a challenge from the calculation point of view since the system of equations that needs to be solved is not an easy one. However, in this paper, it is shown that when the two techniques are studied, the proposed MSHEPWM technique offers signicant benets which can justify the extra effort involved in solving the equations. The SPWM provides switching angles that are not optimum and mathematically are not calculated in order to eliminate the maximum number of possible harmonics from the spectrum. When the mathematical approach is followed, the maximum attainable modulation index increases to 1.7 p.u. when compared with the 1.57 p.u. possible with the MPSSPWM technique. If overmodulation is desired for the SPWM, low-order harmonics are introduced. This was resolved in two-level PWM systems with the introduction of third harmonic into the reference signal to increase the gain before bandwidth deteriorates [30]. In the multilevel case however, the most signicant harmonic in the case of the MPSSPWM is the 29 p.u. frequency. This gain can be increased if a third harmonic is also introduced in the reference waveform. With the proposed approach, the harmonic is tightly controlled for all modulation indices and the switching angles available can eliminate up to 49 p.u. The 51 p.u. happens to be a triplen harmonic therefore the rst signicant non-triplen harmonic becomes the 53 p.u. This results in an increased bandwidth without any further need to increase the switching transitions. These benets are summarized in Table III conrming that the mathematical approach to harmonic control in a multilevel system is a benecial approach although the way the angles are calculated requires effort and computing time. B. Experimental Results A low-power laboratory ve-level inverter prototype based on the two IGBT (IRG4BC20FD) H-bridges conguration shown in Fig. 8 was developed and tested to verify the feasibility and the validity of the theoretical and the simulation ndings. Each dc bus voltage is approximately 50 V. High-voltage high-speed drivers (IR2112) were used along with optically coupled isolators (SFH610). The pre-calculated PWM signals

This is shown in Fig. 7(c). Finally, the line-to-line voltage waveform for a three-phase system is shown in Fig. 7(d) and its spectrum in Fig. 7(e).

AGELIDIS et al.: FIVE-LEVEL SYMMETRICALLY DEFINED SELECTIVE HARMONIC ELIMINATION PWM STRATEGY

25

Fig. 10. Experimental results for an operating point within the overmodulation index range (a) proposed ve-level MSHEPWM line-to-neutral (M = 1.7 (Set 2) and k=m = 1/9, N = 10) and (b) spectrum of the proposed ve-level MSHEPWM line-to-neutral voltage waveform.

Fig. 8. Schematic diagram of the converter topology used to verify the proposed ve-level (line-to-neutral) MSHEPWM method.

output placing the most signicant harmonics at 29 p.u. as the theory predicts. The overmodulation region for the proposed method is also experimentally veried and results presented in Fig. 10. Specifically, Fig. 10(a) shows the ve-level SHEPWM waveform for 1.7 (Set 2) and 1/9, 10. The spectrum of the waveform shown in Fig. 10(b) conrms that only triplen harmonics exist and the rst non-triplen one is the 31 p.u. as the theory suggests. V. CONCLUSION A symmetrically dened ve-level multilevel SHEPWM technique has been proposed in this paper. The various switching transitions are calculated using a minimization technique with a biased search optimization approach. This approach results in an efcient method to obtain the switching transitions. It is shown that these angles can be computed for all modulation indices by using a distribution ratio and search for solutions. When compared with the conventional MPSSPWM technique and keeping the number of transitions the same, the proposed method offers signicant benets for increased modulation index and higher bandwidth tightly controlled throughout the entire range. Extending its operating range to the overmodulation region where a higher gain can be obtained while compromising some of the bandwidth is also possible. Selected simulation and experimental results have been presented to conrm the theoretical ndings. REFERENCES

Fig. 9. Experimental results for an operating point within the standard modulation index range (a) Proposed ve-level MSHEPWM line-to-neutral (M = 1.5) (b) Spectrum of the proposed ve-level MSHEPWM line-to-neutral voltage waveform. (c) Line-to-neutral voltage waveform of the MPSSPWM technique (M = 1.5, four carriers, 9 p.u. frequency each) (d) Spectrum of the line-to-neutral voltage waveform of the MPSSPWM technique.

are implemented using low-cost high-speed Texas Instruments TMS320F2812 digital signal processor (DSP) board with an accuracy of 20 s. A digital real-time oscilloscope (Tektronix TDS210) was used to display and capture the output waveforms and using the feature of the fast Fourier transformer (FFT), the spectrum of each of the output voltage was obtained. Specically, Fig. 9(a) shows the line-to-neutral voltage waveform of the proposed ve-level MSHEPWM method for 1.5 and the associated spectrum is presented in Fig. 9(b). It is conrmed that only triplen harmonics are present in the waveform and the rst signicant non-triplen harmonic is the 53 p.u. as predicted. For completeness, the same ve-level waveform is presented when the MPSSPWM technique is used also for 1.5. Clearly, the method controls the bandwidth at the

[1] H. S. Patel and R. G. Hoft, Generalized harmonic elimination and voltage control in thyristor inverters: Part IHarmonic elimination, IEEE Trans. Ind. Appl., vol. IA-9, no. 3, pp. 310317, May/Jun. 1973. [2] H. S. Patel and R. G. Hoft, Generalized harmonic elimination and voltage control in thyristor inverters: Part IIVoltage control technique, IEEE Trans. Ind. Appl., vol. IA-10, no. 5, pp. 666673, Sep./ Oct. 1974. [3] P. Enjeti and J. F. Lindsay, Solving nonlinear equations of harmonic elimination PWM in power control, Electron. Lett., vol. 23, no. 12, pp. 656657, Jun. 1987. [4] P. N. Enjeti, P. D. Ziogas, and J. F. Lindsay, Programmed PWM techniques to eliminate harmonics: A critical evaluation, IEEE Trans. Ind. Appl., vol. 26, no. 2, pp. 302316, Mar./Apr. 1990. [5] T. Kato, Precise PWM waveform analysis of inverter for selected harmonic elimination, in Proc. IEEE Ind. Appl. Soc. Annu. Meeting, 1986, pp. 611616. [6] T. Kato, Sequential homotopy-based computation of multiple solutions for selected harmonic elimination in PWM inverters, IEEE Trans. Circuits Systems I, vol. 46, no. 5, pp. 586593, May 1999.

26

IEEE TRANSACTIONS ON POWER ELECTRONICS, VOL. 23, NO. 1, JANUARY 2008

[7] J. Chiasson, L. M. Tolbert, K. McKenzie, and Z. Du, A complete solution to the harmonic elimination problem, IEEE Trans. Power Electron., vol. 19, no. 2, pp. 491499, Mar. 2004. [8] V. G. Agelidis, A. Balouktsis, and I. Balouktsis, On applying a minimization technique to the harmonic elimination PWM control: The bipolar waveform, IEEE Power Electron. Lett., vol. 2, no. 2, pp. 4144, Jun. 2004. [9] F. Swift and A. Kamberis, A new Walsh domain technique of harmonic elimination and voltage control in pulse-width modulated inverters, IEEE Trans. Power Electron., vol. 8, no. 2, pp. 170185, Apr. 1993. [10] T. J. Liang, R. M. OConnell, and R. G. Hoft, Inverter harmonic reduction using Walsh function harmonic elimination method, IEEE Trans. Power Electron., vol. 12, no. 6, pp. 971982, Nov. 1997. [11] J. R. Wells, B. M. Nee, P. L. Chapman, and P. T. Krein, Optimal harmonic elimination control, in Proc. IEEE Power Electron. Spec. Conf., 2004, pp. 42144219. [12] B. Ozpineci, L. M. Tolbert, and J. N. Chiasson, Harmonic optimization of multilevel converters using genetic algorithms, IEEE Power Electron. Lett., vol. 3, pp. 9295, Sep. 2005. [13] M. S. A. Dahidah and V. G. Agelidis, A hybrid genetic algorithm for selective harmonic elimination control of a multilevel inverter with non-equal dc sources, in Proc. 6th IEEE Power Electron. Drives Syst. Conf., Kuala Lumpur, Malaysia, Nov. 2005, pp. 12051210. [14] V. G. Agelidis, A. Balouktsis, I. Balouktsis, and C. Cossar, Multiple sets of solutions for harmonic elimination PWM bipolar waveforms: Analysis and experimental results, IEEE Trans. Power Electron., vol. 21, no. 2, pp. 415421, Mar. 2006. [15] J. R. Wells, B. M. Nee, P. L. Chapman, and P. T. Krein, Selective harmonic control: A general problem formulation and selected solutions, IEEE Trans. Power Electron., vol. 20, no. 6, pp. 13371345, Nov. 2005. [16] R. A. Jabr, Solution trajectories of the harmonic-elimination problem, Proc. Inst. Elect. Eng., vol. 153, pp. 97104, Jan. 2006. [17] P. M. Bhagwat and V. R. Stefanovic, Generalized structure of a multilevel PWM inverter, IEEE Trans. Ind. Appl., vol. IA-19, no. 6, pp. 10571069, Nov./Dec. 1983. [18] J. N. Chiasson, L. M. Tolbert, K. J. McKenzie, and Z. Du, Control of a multilevel converter using resultant theory, IEEE Trans. Contr. Syst. Technol., vol. 11, no. 3, pp. 345354, May 2003. [19] J. N. Chiasson, L. M. Tolberta, K. J. McKenzie, and Z. Du, A unied approach to solving the harmonic elimination equations in multilevel converters, IEEE Trans. Power Electron., vol. 19, no. 2, pp. 478490, Mar. 2004. [20] J. N. Chiasson, L. M. Tolbert, K. J. McKenzie, and D. Zhong, Elimination of harmonics in a multilevel converter using the theory of symmetric polynomials and resultants, IEEE Trans. Contr. Syst. Technol., vol. 13, no. 2, pp. 216223, Mar. 2005. [21] L. Li, D. Czarkowski, Y. Liu, and P. Pilay, Multilevel selective harmonic elimination PWM technique in series-connected voltage inverters, IEEE Trans. Ind. Appl., vol. 36, no. 1, pp. 160170, Jan./Feb. 2000. [22] Z. Du, L. M. Tolbert, and J. N. Chiasson, Harmonic elimination in multilevel converter with programmed PWM method, in Proc. IEEE Ind. Appl. Soc. Annu. Meeting, 2004, pp. 22102215. [23] J. Vassallo, J. C. Clare, and P. W. Wheeler, A power-equalized harmonic elimination scheme for utility-connected cascaded H-bridge multilevel converters, in Proc. IEEE IECON03, 2003, vol. 2, pp. 11851190. [24] J. Holtz, W. Lotzkat, and K. H. Werner, A high-power multitransistorinverter uninterruptible power supply system, in Proc. IEEE Power Electron. Spec. Conf., Vancouver, BC, Canada, 1986, pp. 311320. [25] J. Holtz, W. Lotzkat, and K. H. Werner, A high-power multitransistor-inverter uninterruptible power supply system, IEEE Trans. Power Electron., vol. PEL-3, no. 3, pp. 278285, Jul. 1988. [26] V. G. Agelidis, A. Balouktsis, I. Balouktsis, and C. Cossar, Fivelevel selective harmonic elimination PWM strategies and multicarrier phase-shifted sinusoidal PWM: A comparison, in Proc. IEEE 36th Power Electron. Spec. Conf. (PESC05), Recife, Brazil, Jun. 2005, pp. 16851691. [27] Z. Du, L. M. Tolbert, and J. N. Chiasson, Active harmonic elimination for multilevel converters, IEEE Trans. Power Electron., vol. 21, no. 2, pp. 459469, Mar. 2006. [28] V. G. Agelidis and M. Calais, Application specic harmonic performance evaluation of multicarrier PWM techniques, in Proc. IEEE Power Electron. Spec. Conf., Fukuoka, Japan, May 1998, pp. 172178. [29] Mathematica 5.0 Wolfram Research, Inc. [30] V. G. Agelidis, P. D. Ziogas, and G. Joos, Dead-band PWM switching patterns, IEEE Trans. Power Electron., vol. 11, no. 4, pp. 522531, Jul. 1996. [31] D. Czarkowski, D. V. Chudnovsky, G. V. Chudnovsky, and I. W. Selesnick, Solving the optimal PWM problem for single-phase inverters, IEEE Trans. Circuits Syst. I, vol. 49, no. 4, pp. 465475, Apr. 2002.

[32] H. Huang, S. Hu, and D. Czarkowski, A novel simplex homotopic xed-point algorithm for computation of optimal PWM patterns, in Proc. IEEE Power Electron. Spec. Conf., 2004, pp. 12631267. [33] H. Huang, S. Hu, and D. Czarkowski, Harmonic elimination for constrained optimal PWM, in Proc. 30th Annu. Conf. IEEE Ind. Electron. Soc., 2004, pp. 27022705. [34] H. Hu, H. Huang, and D. Czarkowski, Hybrid trigonometric differential evolution for optimizing harmonic distribution, in Proc. IEEE Int. Symp. Circuits Syst., 2005, vol. 2, pp. 13061309. [35] J. R. Wells, X. Geng, P. L. Chapman, P. T. Krein, and B. T. Nee, Modulation-based harmonic elimination, IEEE Trans. Power Electron., vol. 22, no. 1, pp. 336340, Jan. 2007.

Vassilios G. Agelidis (SM00) was born in Serres, Greece. He received the B.S. degree in electrical engineering from Democritus University of Thrace, Thrace, Greece, in 1988, the M.S. degree in applied science from Concordia University, Montreal, QC, Canada, in 1992, and the Ph.D. degree in electrical engineering from the Curtin University of Technology, Perth, Australia, in 1997. From 1993 to 1999, was with the School of Electrical and Computer Engineering, Curtin University of Technology. In 2000, he joined the University of Glasgow, Glasgow, U.K., as a Research Manager for the Centre for Economic Renewable Power Delivery. In addition, he has authored/coauthored several journal and conference papers as well as Power Electronic Control in Electrical Systems (2002). From January 2005 to December 2007, he was the inaugural Chair in Power Engineering in the School of Electrical, Energy and Process Engineering, Murdoch University, Perth, Australia. Since January 2007, he has held the EnergyAustralia Chair of Power Engineering at the University of Sydney, Australia. Dr. Agelidis received the Advanced Research Fellowship from the United Kingdoms Engineering and Physical Sciences Research Council (EPSRC-UK) in 2004. He was the Vice President Operations within the IEEE Power Electronics Society for 20062007. He was an Associate Editor of the IEEE POWER ELECTRONICS LETTERS from 2003 to 2005, and served as the PELS Chapter Development Committee Chair from 2003 to 2005. He is currently an AdCom member of IEEE PELS for 20072009. He will be the Technical Chair of the 39th IEEE PESC08, Rhodes, Greece. Anastasios I. Balouktsis (M99) was born in Serres, Greece, in 1955. He received the Dipl.-Eng. degree in electrical and mechanical engineering and the B.S. degree in mathematics from Aristotle University of Thessaloniki, Thessaloniki, Greece, in 1978 and 1982, respectively, and the Ph.D. degree in electrical engineering from Democritus University of Thrace, Xanthi, Greece, in 1986. From 1981 to 1986, he was a scientic collaborator and from 1987 to 1990 a Lecturer with the Department of Electrical Engineering, Democritus University of Thrace. Since 1990, he has been a Professor with the Technological Educational Institution (T.E.I.), Serres, Greece. He has been elected as the Head of School of Applied Technology and has been Vice President and President of the Institution (TEI). He is the author/coauthor of many scientic papers in various disciplines and his technical interests include renewable energy sources, power systems, power electronics, digital electronics, and mathematics. Mohamed S. A. Dahidah (M03) was born in Tripoli, Libya. He received the B.S. degree in electrical and electronic engineering from the Bright Star University of Technology, Briga, Libya, in 1998, the M.S. degree in applied science from Universiti Putra Malaysia, Malaysia, in 2002, and is currently pursuing the Ph.D. degree in the Faculty of Engineering and Technology, Multimedia University, Selangor, Malaysia. In November 2007, he was appointed Lecturer in the School of Electrical and Electronic Engineering, The University of Nottingham (Malaysia Campus), Selangor. He has authored or co-authored a number of refereed journal and conference papers. His research interests include power electronics, selective harmonic elimination techniques, and PWM converter control.