Title of Invention

MATRIX CONVERTER AND CONTROL METHOD FOR THE MATRIX CONVERTER

Abstract

An output voltage command signal for outputting a specified three-phase ac output voltage is generated by a line voltage control command signal generating section 11, and a signal representing a current flow ratio is generated by a current flow ratio generating section 12 based on a specified input current command signal. The output voltage command signal is corrected by a command signal computing section 13 based on the output voltage command signal generated by the line voltage control command signal generating section 11 and the signal representing the current flow ratio generated by the current flow ratio generating section 12. A PWM conversion signal is generated by a PWM conversion signal generating section 15, 17 based on the corrected output voltage command signal and a carrier signal. Based on the generated PWM conversion signal, a three-phase ac input voltage is converted into a specified three-phase ac input voltage by a conversion section.

Full Text

have been increasingly gaining attentions as next- generation power converter apparatuses. A three-phase to three-phase matrix converter is to obtain ac output voltages of variable voltages and variable frequencies by direct switching of a three-phase ac input voltage. As a typical modulation method, an analog base modulation method in which, with a virtual DC link formed, sinusoidal modulation is performed by carrier comparison is proposed in Document 1 (Jun OYAMA and other five, "VWF On-Line Control of PWM Cycloconverter", IEEJ Transactions D, IEEJ, Vol. 116 (1996), No. 6, pp. 664-651). Also, a control method for solving current distortion, which is an issue in asynchronism between output voltage command signal and carrier signal, is disclosed in Document 2 (JP 11-341807 A) . The control methods of Documents 1 and 2, in either case, is based on an amplitude modulation which is accomplished by introducing a variable that is a current distribution ratio, i.e. a ratio of a mid phase current to a maximum phase current, and by multiplying a factor on each of an output voltage command signal and a carrier signal. Further, the control method of Document 2 involves two steps of control for suppressing changeover of switching patterns, which serve as information updating part, so that current distortion in carrier asynchronism is avoided. Due to this, matrix converters of Documents 1 and 2 have an issue that control configuration and computation become more complex. SUMMARY OF THE INVENTION It is a primary object of the present invention to provide a matrix converter, as well as a matrix converter control method, which allows the control-related circuit construction to be simplified without involving such computations as ripple computation or carrier amplitude modulation. In order to achieve the above object, the present invention provides a matrix converter for converting a three-phase ac input voltage into a specified three-phase ac output voltage, comprising: an output voltage command signal generating section for generating an output voltage command signal to output the specified three-phase ac output voltage; a current flow ratio generating section for generating a signal representing a current flow ratio based on a specified input current command signal; a signal correcting section for correcting the output voltage command signal based on the output voltage command signal generated by the output voltage command signal generating section and the signal representing the current flow ratio generated by the current flow ratio generating section; a PWM conversion signal generating section for generating a PWM conversion signal based on the output voltage command signal corrected by the signal correcting section and a carrier signal; and a conversion section for converting the three- phase ac input voltage into the specified three-phase ac output voltage based on the PWM conversion signal generated by the PWM conversion signal generating section. According to this matrix converter, based on the output voltage command signal generated by the output voltage command signal generating section and the signal representing the current flow ratio generated by the current flow ratio generating section, the output voltage command signal is corrected by the signal correcting section. Based on the corrected output voltage command signal and the carrier signal, the PWM conversion signal is generated by the PWM conversion signal generating section. With such a modulation method by input current flow ratio, a constant average virtual dc link voltage can be obtained and, as a result, the control-related circuit construction can be simplified without involving such computations as ripple computation or carrier amplitude modulation. In one embodiment, given that the input current command signals are represented by ir * , is* , it * and a voltage command is represented by Vs * , assuming that the lowest voltage phase or the highest voltage phase out of the output voltage command signal is a reference voltage, the current flow ratio generating section generates the current flow ratio of the maximum phase by and generates a sum of the current flow ratio of the maximum phase and the current flow ratio of the mid phase by In this embodiment, the current flow ratio of the maximum phase as well as the sum of the current flow ratio of the maximum phase and the current flow ratio of the mid phase are generated by the current flow ratio generating section based on the input current command signal, and the output voltage command signal is corrected by using the current flow ratios. Thus, the virtual dc link voltage can be maintained as a constant dc voltage. In one embodiment, given that the input current command signals are represented by ir* , is * , it * and that a phase angle of the input current command signals ir* , is * , it * relative to the output voltage command signal is represented by Ø , the current flow ratio generating section generates the current flow ratio by In this embodiment, in generating a current flow ratio based on the input current command signal, the current flow ratio generating section looks up to a single sine table, allowing further simplification to be achieved. In one embodiment, the carrier signal is a triangle wave signal having a generally constant amplitude. In this embodiment, the matrix converter is easily applicable to digital counters and comparators by virtue of its constant carrier amplitude, and suitable for digital control systems by virtue of its capability of maintaining the PWM modulation resolution constant. Moreover, the matrix converter, in which the triangle wave signal suitable for PWM modulation is used as the carrier signal, allows the circuit for pulse width modulation to be simplified. In one embodiment, the carrier signal is a sawtooth wave signal having a generally constant amplitude. In this embodiment, the matrix converter is easily applicable to digital counters and comparators by virtue of its constant carrier amplitude, and suitable for digital control systems by virtue of its capability of maintaining the PWM modulation resolution constant. Moreover, the matrix converter, in which the sawtooth wave signal is used as the carrier signal, allows carrier generation and modulation processing to be simplified. One embodiment further comprises: a holding section for holding the output voltage command signal corrected by the signal correcting section and power supply voltage information, wherein the holding section updates the output voltage command signal and the power supply voltage information at a timing of a peak of the carrier signal, and the PWM conversion signal generating section generates the PWM conversion signal based on the output voltage command signal and the power supply voltage information held by the holding section. In this embodiment, update of the output voltage command signal and the power supply voltage information is synchronized at the timing of the peak of the carrier signal, thus making it possible to prevent occurrence of current distortions as well as to achieve simplification of the control-related construction. According to the present invention, there is provided a matrix converter control method for converting a three-phase ac input voltage into a specified three-phase ac output voltage, comprising the steps of: generating an output voltage command signal for outputting the specified three-phase ac output voltage by an output voltage command signal generating section; generating a signal representing a current flow ratio by a current flow ratio generating section based on a specified input current command signal; correcting, by a signal correcting section, the output voltage command signal generated by the output voltage command signal generating section based on the output voltage command signal and the signal representing the current flow ratio generated by the current flow ratio generating section; generating a PWM conversion signal by a PWM conversion signal generating section based on the output voltage command signal corrected by the signal correcting section and a carrier signal; and converting the three-phase ac input voltage into the specified three-phase ac output voltage by a conversion section based on the PWM conversion signal generated by the PWM conversion signal generating section. In this matrix converter control method, based on the output voltage command signal generated by the output voltage command signal generating section and the signal representing the current flow ratio generated by the current flow ratio generating section, the output voltage command signal is corrected by the signal correcting section. Based on the corrected output voltage command signal and the carrier signal, the PWM modulation signal is generated by the PWM modulation signal generating section. With such a modulation method by input current flow ratio, a constant average virtual dc link voltage can be obtained and, as a result, the control-related circuit construction can be simplified without involving such computations as ripple computation or carrier amplitude modulation. As apparent from the above description, according to the matrix converter and the matrix converter control method of the invention, there can be realized a matrix converter in which the control-related circuit construction can be simplified without involving such computations as ripple computation or carrier amplitude modulation. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a constructional view of a conversion section of a matrix converter according to an embodiment of the present invention; Fig. 2 is a block diagram of a control section of the matrix converter; Figs. 3A-3F are charts showing waveforms of individual sections of the matrix converter; Fig. 4 is a chart showing turn-on timing of the matrix converter; Figs. 5A and 5B are charts of signal update timing of the matrix converter; Figs. 6A-6D are charts showing signal update timing of occurrence of distortion; Fig. 7 is a diagram showing a logic circuit of a pulse pattern circuit; Fig. 8 is a block diagram of a control section in a matrix converter which is a comparative example; Figs. 9A-9F are charts showing waveforms of individual sections of the matrix converter; Fig. 10 is a chart showing turn-on timing of the matrix converter; Fig. 11 is a diagram showing a logic circuit of the matrix converter; Figs. 12A-12D are charts showing signal update timing of occurrence of distortion; Figs. 13A-13D are charts showing effects of carrier cycle in various switching states; DETAILED DESCRIPTION OF THE INVENTION Hereinbelow, the matrix converter and the matrix converter control method according to the present invention will be described in detail by way of embodiments thereof illustrated in the accompanying drawings. Fig. 1 is a constructional view of a matrix converter according to one embodiment of the invention. As shown in Fig. 1, the three-phase to three-phase matrix converter has three ac switches connected to individual phases of a three-phase power supply, respectively, in steps of output phases, with a view to obtaining variable voltages and variable frequencies by controlling the connection phase and time as required. This matrix converter, as shown in Fig. 1, includes a converter section 1 composed of switches Sur^ Sus / Su t r Sv r i Sv s r Svt / Sw r i Sws / Swt / and a controi section 2 (shown in Fig. 2) for outputting signals to turn on and off the switches S SWs/ Swt of the converter section 1. In the converter section 1, a phase voltage vr out of three-phase ac input voltages derived from a three- phase ac power supply 3 is inputted to one end of each of the switches Sur, Sv c , Swr, a phase voltage vs out of the three-phase ac input voltages is inputted to one of each of SUs/ Svs, Sws, and a phase voltage vt out of the three- phase ac input voltages is inputted to one end of each of respectively. The other ends of the switches Sur, Sus, Sut are connected to an output terminal of the phase voltage vu, the other ends of the switches Svr, Svs/ Svt are connected to an output terminal of the phase voltage vr , and the other ends of the switches Swr, S¥s/ Swt are connected to an output terminal of the phase voltage vw. Also, Fig. 2 shows a block diagram of the control section 2. The control section 2, as shown in Fig. 2, includes: a line voltage control command signal generating section 11 as an example of an output voltage command signal generating section for generating a line voltage control command signal based on output voltage command signals Vu * , Vv * , Vw * and an input voltage Vmld; a current flow ratio generating section 12 for computing a current flow ratio based on input current command signal ir * , i3 * , it * ; a command signal computing section 13 as an example of a signal correcting section for computing a maximum phase + mid phase command signal and a mid phase command signal based on a line voltage control command signal generated by the line voltage control command signal generating section 11 and a current flow ratio computed by the current flow ratio generating section 12; an S/H (sample-hold) section 14 for updating and holding a maximum phase + mid phase command signal and a mid phase command signal derived from the command signal computing section 13 at a timing of a peak of a carrier signal; a carrier comparison section 15 which has a carrier signal generation circuit 20 for generating the carrier signal and which performs a comparison between the maximum phase + mid phase command signal held by the S/H section 14 and the carrier signal, as well as a comparison between the mid phase command signal and the carrier signal; an S/H (sample-hold) section 16 for updating and holding power supply voltage phase information Vmax, Vmid, Vmin and a reference voltage X/Y at a timing of a peak of a carrier signal derived from the carrier signal generation circuit 20; and a pulse pattern generating section 17 for generating a pulse pattern (PWM conversion signal) for turning on and off the SWitCheS Sur ; SUSf Out/ ^v r / ^vs / Svtj Sw r i ^>w s i ^w t Daseu. on comparison results from the carrier comparison section 15 and the power supply voltage phase information Vmax, Vj,ld, Vmin and the reference voltage X/Y held in the S/H section 16. The command signal computing section 13 has a maximum phase + mid phase command signal computing section 13a for multiplying a line voltage control command signal derived from the line voltage control command signal generating section 11 by a maximum phase + mid phase command signal, and a mid phase command signal computing section 13b for multiplying the line voltage control command signal by a mid phase command signal. The carrier comparison section 15 and the pulse pattern generating section 17 constitute a PWM conversion signal generating section. Also, the S/H section 14 and the S/H section 16 constitute a hold section. (Comparative Example) Fig. 8 shows a block diagram of a control section in a matrix converter which is a comparative example. It is noted that the matrix converter shown in Fig. 8 is given for an easier understanding of the present invention, and neither a known or prior art nor a matrix converter of the invention. The control section of the matrix converter in this Comparative Example, as shown in Fig. 8, includes: a line voltage control command signal generating section 101 for generating a line voltage control command signal based on output voltage command signals Vu * , Vv * , Vw * and a current distribution ratio 'a'; a voltage ripple computing section 102 for computing a voltage ripple based on input voltages Vmax, Vmid, Vmin; a current distribution ratio computing section 103 for computing a current distribution ratio 'a' based on input current command signals imax * / imid* , imin* : a command signal computing section 104 for computing maximum phase + mid phase command signal and a maximum phase command signal based on a line voltage control command signal generated by the line voltage control command signal generating section 11 and a current distribution ratio 'a' computed by the current distribution ratio computing section 103; a carrier comparison section 105 which has a carrier signal generation circuit 120 for generating a carrier signal and which performs a comparison between the maximum phase + mid phase command signal derived from the command signal computing section 104 and the carrier signal, as well as a comparison between the maximum phase command signal and the carrier signal; a switching section 106 for outputting a switching signal based on a comparison result from the carrier comparison section 105; an S/H section 107 for updating and holding voltage command phase information 9r derived from the line voltage control command signal generating section 101 and power supply voltage phase information Vmax, Vmid, Vmln based on a Bas-level switching signal from the switching section 106; an S/H section 108 for updating and holding a reference voltage X/Y based on a Top-level switching signal from the switching section 106; and a pulse pattern generating section 109 for generating a pulse pattern (PWM conversion signal) to turn on and off the switches Sur, SUS/ Sut/ Svr, Svs, Svt, Swr, Sws, Sw t of the conversion section based on a comparison result from the carrier comparison section 105 as well as the voltage command phase information 8r and the power supply voltage phase information Vmax, Vmid, Vmin and the reference voltage X/Y held in the S/H sections 107, 108. The carrier signal generation circuit 120 of the carrier comparison section 105 performs amplitude modulation of a carrier signal based on a voltage ripple signal derived from the voltage ripple computing section 102. Figs. 9A-9F show waveforms of individual sections by the matrix converter control method of the Comparative Example, where the input power factor is assumed to be 1 and the output frequency is assumed as equal to that of the power supply for simplicity's sake. In this case, with regard to a magnitude relationship of three-phase ac potential, the mid phase potential is inverted in polarity in every 60-degree phase angle. Therefore, under the condition that based on the polarity of the mid phase voltage (positive: region X, negative: region Y), a reference waveform (minimum phase for the region X, maximum phase for the region Y) is changed, potentials with respect to the reference waveform are X : E max = V max - V min , E mid = V mid - V min Y : E max = V max - V min , E mid = V mid - V min where it is understood that two potentials can be obtained. Meanwhile, with respect to the command signal, a two-phase modulation waveform resulting from dividing a three-phase voltage command by a minimum phase is used. In this case, in the region Y, in which the potential becomes negative, the polarity is inverted to change over the waveform to one resulting from division by the maximum phase. From matching of the output voltage polarity in this way, the virtual dc link voltage results in a relationship expressed by the following equations: X : E max = V max - V min , E mid = V mid - V min Y : E max = -(V min - V max) , E mid = -(V mid - V max) where pulsating currents by two potentials shown in Fig. 9B are modulated. Next, the method for apportioning the command signal into two signals is defined by the current distribution ratio, which is a ratio of the mid phase current to the maximum phase current. Given a semi-cycle T of the carrier signal, a time T0 from a peak of the carrier signal to a switching to the mid phase, and a time T1 from a peak of the carrier signal to a switching to the maximum phase, the current distribution ratio 'a' is expressed by In this case, because of the expression with a power factor of 1, the ratio of phase voltage is used in Fig. 9A. Time ratios of connection to the maximum phase voltage and the mid phase voltage are determined by the following equations. In this connection, since the maximum phase value is multiplied by a voltage command value and the mid phase value is multiplied by a current distribution ratio, an average virtual dc link voltage is defined by a value resulting from addition of the mid voltage to the maximum voltage at a current distribution ratio. Max. phase : (T-T,)/T = V*S/Vtri ••• (1) Mid phase : (T1-T0)/T = aV*S/Vtri ••• (2) Max. + mid phase : (T-T0)/T = (l+a)V*S/Vtri ••• (3) Carrier amplitude : Vtri = Emax + aE mid In this case, for the region X, the phase of the lowest voltage out of the output voltage command signals Vu * , Vv * , Vw * is assumed as the reference voltage Vs * , while for the region Y, a voltage command in which the phase of the highest voltage out of the output voltage command signals Vu * , Vv * , Vw * is taken as the reference voltage is assumed as Vs * . Therefore, the pulsating current is compensated by amplitude modulation of the carrier signal with the average virtual dc link voltage, and then PWM modulation through carrier comparison is performed with two command signals at conduction timing shown in Fig. 10. In this case, patterns of gating to one-phase portions of the maximum, mid and minimum phases are isolated by using the following logical equations: Max. phase : X = (Vtri Min. phase : Z = (Vtri>(l + a)V*s) Mid phase : Y = (XuZ). With respect to the input phase of connection, as shown in Fig. 11, gate signals for the respective switches are apportioned by logical circuits based on logic signals showing the magnitude relationships among the individual phases. In the matrix converter of this Comparative Example, since modulation is performed by multiplying both command signal and carrier signal by factors with the use of the current distribution ratio 'a', there has been a problem that the control configuration or computations would be complicated. In the matrix converter of the Comparative Example, the ratio of the mid phase to the maximum phase would be defined as the current distribution ratio. However, in the present invention, the current is distributed with a normalized current value. Figs. 3A-3F show normalized current waveforms in the matrix converter of the invention, where it is assumed that shunting with Cl is applied to the maximum phase and shunting with C2 is applied to the mid phase. Since the maximum phase is assumed as 1 in the method of the foregoing Comparative Example, multiplying by a factor of 2/V3 to set 1 on the boundary of the region X and the region Y causes flow ratios among the individual phases to be expressed by the following relationship: In this case, for the region X, the phase of the lowest voltage out of the output voltage command signals Vu * , Vv * , Vw * is assumed as the reference voltage Vs * , while for the region Y, a voltage command in which the phase of the highest voltage out of the output voltage command signals Vu * , Vv * , Vw * is taken as the reference voltage is assumed as Vs * . Although a sum of flow ratios of the maximum phase and the mid phase becomes over 1, the cos30° = — 2 voltage control rate of the maximum voltage command results from the virtual dc link voltage waveform as: Thus, the maximum value of flow ratios results in 1 eventually. Multiplying two potentials of the virtual dc link by the above flow ratio results in an average virtual dc link voltage waveform. Since a constant dc voltage is obtained as shown in Fig. 3F, it can be understood that amplitude modulation of the carrier signal as in the Comparative Example is unnecessary. Fig. 4 shows a chart for explaining the conduction timing described above. At the conduction timing shown in Fig. 4, PWM modulation by carrier comparison is performed. Next, it is shown that the same results as in the Comparative Example can be obtained by the method of the invention. It is noted that the power factor is assumed as 1, and the description will be given by phase voltage. For intervals with the phase angle ranging from 30° to 60°, it follows that Emid = V^-V^ = Vs-Vt = -j|=(cos0,-cos0,) = cos(rf-*/3). In this case, since Comparative Example of Fig. 8, becomes constant in average virtual dc link voltage, eliminating the needs for blocks associated with the voltage ripple computation and the carrier amplitude modulation, so that its construction can be simplified. Although Equations (4) and (6) are shown as blocks for determining current flow ratios, yet it is also possible to further simplify the construction by looking up to a single sine table for the determination of mid and maximum values with the use of a phase angle Ø within the region X or region Y: Also, a case with the power factor of 1 only is shown above, and the switching of the region X and Y is performed by referencing the polarity of the mid phase of the power supply voltage (three-phase ac input voltage). However, when the input power factor is variable, it is also possible to control the power factor by switching with a phase of the mid phase to which a phase difference is added. In the matrix converter of the Comparative Example of Fig. 8, because the voltage command has a two- phase modulation waveform, the maximum phase and the mid phase out of the voltage command are subjected to PWM modulation, and apportioning to the individual phases is performed by using phase information 9r. Further, as a measure for a case where the synchronism between power supply frequency and carrier frequency is not ensured, updating timing is limited to the reference voltage X/Y, the voltage command phase information 9r and the power supply voltage phase information Vmax, Vmid, Vmin. Figs. 12A to 12D show timings at which distortion occurs in a case where the matrix converter of the Comparative Example is asynchronous. In this case, distortion occurs to three points of change, i.e. change of reference voltage, phase shift of the two-phase modulation waveform, and phase shift of the power supply voltage (current). Figs. 13A to 13D show effects on the above changes in individual switching states (maximum, mid and minimum) of the carrier cycle. These figures are shown by single-phase output for simplicity, where the phase having only one phase short-circuited by bold line is the 120° conduction phase of the two-phase modulation waveform while the other is any two phases. In the initial state, it is assumed that the maximum, mid and minimum voltages correspond to r phase, s phase and t phase, respectively. (1) State of conduction to the maximum phase (shown in Fig. 13B) As the reference voltage X/Y changes, voltage command values of U and V phases are inverted while switching signals of r and t phases are switched over. As a result, the state changes from X (+-) to Y (-+) , whereas neither U nor V phase changes in switching state to the maximum and minimum phases. In a case where the conduction phase and the switching phase have changed due to a phase shift of the voltage command phase information 9r, switching between U phase and V phase causes the state to change from X(+-) to X(-+) so that the output changes from r phase to t phase. With regard to the power supply voltage phase information Vmax, Vmid, Vmin, when the maximum phase and the mid phase of the input voltage have been switched (r phase ↔ s phase) , the switch for the U phase to conduct changes from Sur to Sus• (2) State of conduction in the mid phase (shown in Fig. 13C) When the reference voltage X/Y has changed, the state changes from X(+-) to Y(-+), so that both U phase and V phase change in switching state. As a result of a phase shift of the voltage command phase information 9r, the state changes from X(+-) to X(-+), so that both U phase and V phase change in switching state. As to the power supply voltage phase information Vmax / Vmid/ Vmin / when the maximum phase and the mid phase of the input voltage have been switched, the switch for the U phase to conduct changes from Sus to Sur. (3) State of conduction to the minimum phase (shown in Fig. 13D) When the reference voltage X/Y has changed, the state changes from X(+-) to Y(-+), so that both U phase and V phase change in switching state. As a result of a phase shift of the voltage command phase information 9r, the state changes from X(+-) to X(-+), whereas the switching state does not change because both U phase and V phase are in the minimum phase. With regard to the power supply voltage phase information Vmax, Vmid, Vmin, when the maximum phase and the mid phase of the input voltage have been switched, the switching state does not change because both U phase and V phase are in the minimum phase. As shown above, in the matrix converter of the Comparative Example, because of the presence of a mode in which a change of the signals does not cause a change of the switching state, it is permitted to update the reference voltage X/Y during the period of conduction to the maximum phase (Top), and the voltage command phase information 8r and the power supply voltage phase information VmaX , Vmid, Vmin during the period of switching to the minimum phase (Bas), so that balance of current distribution within the carrier cycle is ensured. Also, for the updating periods therefor are used a voltage command mid phase signal that minimizes the conduction period to the power supply maximum phase (Top) as well as a voltage command maximum phase signal that minimizes the conduction period to the power supply minimum phase (Bas). In the matrix converter of the Comparative Example, since the voltage command has a two-phase modulation waveform, the three-phase voltage command is not used, and only the maximum phase and the mid phase are subjected to PWM modulation, where phase information 9r is used for apportioning to the individual phases. Therefore, not only the reference voltage X/Y and the power supply voltage phase information (Vmax, Vmid, Vmin), which are subject to update limitations to avoid distortions due to asynchronism, but also phase information needs to be subject to update limitation. Further, update control involves the use of two signal levels, i.e. a mid phase signal and a maximum phase signal, giving rise to a problem of complicated construction of the control system. In the digital control of inverters, signal input and output are controlled in synchronization with the timing of PWM carrier cycles in order to take matching of signal delay time. In this connection, in the matrix converter of the invention, on the assumption that a signal wave to be modulated is updated and held at the timing of a peak of the carrier signal as shown in Fig. 5, the signal is switched in a state of conduction to the minimum phase. In this case, the voltage command phase information 9r and the power supply voltage phase information Vmax, Vmid/ v"min correspond to updatable timings, whereas the invention uses the three-phase voltage command, containing phases, so that update control for the phase information 9r becomes unnecessary. The power supply voltage phase information Vmax/ Vmid, Vm i n is updated in synchronization with the timing of a peak of the carrier signal. Also in this invention, the reference voltage X/Y is updated at the timing of a peak of the carrier signal as well. The timing of a peak of the carrier signal corresponds to the minimum phase, where update control would be performed by level signals, in conventional cases, so that the input current distribution within the carrier cycle is unbalanced as shown in Fig. 6A and 6B. On the other hand, as to the update in the maximum phase, there occurs no changes in switching state with respect to the changeover of the reference voltage X/Y as shown in Fig. 6C and 6D. Changes in switching state by update timing in this invention are shown in Fig. 5A and 5B. In this case, with respect to the minimum phase level signal, waveform update is performed at the timing of a peak of the carrier signal. As a result, whereas the switching state of the maximum phase and the minimum phase is changed, the waveform is maintained symmetrical, making it understood that current distribution of a half carrier cycle keeps equal, as compared with the updated waveform in the maximum phase. As shown above, update of the reference voltage X/Y and the power supply voltage phase information Vmax, Vmid, Vmin is synchronized at the timing of a peak of the carrier signal in which the voltage command is updated and held, thus making it possible to prevent occurrence of current distortions as well as to achieve simplification of the control-related construction. In addition, the update of control information in the pulse pattern generating section 17 of Fig. 2 can be provided by using such a circuit as shown in Fig. 7 as an example. According to the matrix converter of this embodiment, the average virtual dc link voltage is made constant by the modulation method with the use of input current flow ratio, eliminating the needs for ripple computation and computations for carrier amplitude modulation, so that the control-related construction can be simplified. Further, the matrix converter is easily applicable to digital counters and comparators by virtue of its constant carrier amplitude, and suitable for digital control systems by virtue of its capability of maintaining the PWM modulation resolution constant. Moreover, the matrix converter is enabled to reduce input current distortions even in cases of asynchronism of three-phase ac input voltage and carrier cycle, with a simple construction in which the output voltage command and the power supply voltage (current) information are updated and held at the timing of carrier peaks. WHAT IS CLAIMED IS: 1. A matrix converter for converting a three-phase ac input voltage into a specified three-phase ac output voltage, comprising: an output voltage command signal generating section (11) for generating an output voltage command signal to output the specified three-phase ac output voltage; a current flow ratio generating section (12) for generating a signal representing a current flow ratio based on a specified input current command signal; a signal correcting section (13) for correcting the output voltage command signal based on the output voltage command signal generated by the output voltage command signal generating section (11) and the signal representing the current flow ratio generated by the current flow ratio generating section (12); a PWM conversion signal generating section (15, 17) for generating a PWM conversion signal based on the output voltage command signal corrected by the signal correcting section (13) and a carrier signal; and a conversion section (1) for converting the three-phase ac input voltage into the specified three-phase ac output voltage based on the PWM conversion signal generated by the PWM conversion signal generating section (15, 17). 2. The matrix converter as claimed in Claim 1, wherein given that the input current command signals are represented by ir * , is * , it * and a voltage command is represented by Vs * , assuming that the lowest voltage phase or the highest voltage phase out of the output voltage command signal is a reference voltage, the current flow ratio generating section (12) generates the current flow ratio of the maximum phase by and generates a sum of the current flow ratio of the maximum phase and the current flow ratio of the mid phase by 3. The matrix converter as claimed in Claim 1, wherein given that the input current command signals are represented by ir * , is * , it * and that a phase angle of the input current command signals ir* , is* i it * relative to the output voltage command signal is represented by 4> , the current flow ratio generating section (12) generates the current flow ratio by 2 4. The matrix converter as claimed in Claim 1, wherein the carrier signal is a triangle wave signal having a generally constant amplitude. 5. The matrix converter as claimed in Claim 1, wherein the carrier signal is a sawtooth wave signal having a generally constant amplitude. 6. The matrix converter as claimed in Claim 4, further comprising a holding section (14, 16) for holding the output voltage command signal corrected by the signal correcting section (13) and power supply voltage information, wherein the holding section (14, 16) updates the output voltage command signal and the power supply voltage information at a timing of a peak of the carrier signal, and the PWM conversion signal generating section (15, 17) generates the PWM conversion signal based on the output voltage command signal and the power supply voltage information held by the holding section (14, 16). 7. A matrix converter control method for converting a three-phase ac input voltage into a specified three-phase ac output voltage, comprising the steps of: generating an output voltage command signal for outputting the specified three-phase ac output voltage by an output voltage command signal generating section (11); generating a signal representing a current flow ratio by a current flow ratio generating section (12) based on a specified input current command signal; correcting, by a signal correcting section (13), the output voltage command signal generated by the output voltage command signal generating section (11) based on the output voltage command signal and the signal representing the current flow ratio generated by the current flow ratio generating section (12); generating a PWM conversion signal by a PWM conversion signal generating section (15, 17) based on the output voltage command signal corrected by the signal correcting section (13) and a carrier signal; and converting the three-phase ac input voltage into the specified three-phase ac output voltage by a conversion section (1) based on the PWM conversion signal generated by the PWM conversion signal generating section (15, 17). An output voltage command signal for outputting a specified three-phase ac output voltage is generated by a line voltage control command signal generating section 11, and a signal representing a current flow ratio is generated by a current flow ratio generating section 12 based on a specified input current command signal. The output voltage command signal is corrected by a command signal computing section 13 based on the output voltage command signal generated by the line voltage control command signal generating section 11 and the signal representing the current flow ratio generated by the current flow ratio generating section 12. A PWM conversion signal is generated by a PWM conversion signal generating section 15, 17 based on the corrected output voltage command signal and a carrier signal. Based on the generated PWM conversion signal, a three-phase ac input voltage is converted into a specified three-phase ac input voltage by a conversion section.

Documents:

4693-KOLNP-2008-(16-05-2014)-ABSTRACT.pdf

4693-KOLNP-2008-(16-05-2014)-ANNEXURE TO FORM 3.pdf

4693-KOLNP-2008-(16-05-2014)-CLAIMS.pdf

4693-KOLNP-2008-(16-05-2014)-CORRESPONDENCE.pdf

4693-KOLNP-2008-(16-05-2014)-FORM-1.pdf

4693-KOLNP-2008-(16-05-2014)-FORM-2.pdf

4693-KOLNP-2008-(16-05-2014)-FORM-5.pdf

4693-KOLNP-2008-(16-05-2014)-OTHERS.pdf

4693-KOLNP-2008-(16-05-2014)-PETITION UNDER RULE 137.pdf

4693-kolnp-2008-abstract.pdf

4693-kolnp-2008-claims.pdf

4693-KOLNP-2008-CORRESPONDENCE-1.1.pdf

4693-kolnp-2008-correspondence.pdf

4693-kolnp-2008-description (complete).pdf

4693-kolnp-2008-drawings.pdf

4693-kolnp-2008-form 1.pdf

4693-KOLNP-2008-FORM 13.pdf

4693-kolnp-2008-form 18.pdf

4693-kolnp-2008-form 2.pdf

4693-kolnp-2008-form 3.pdf

4693-kolnp-2008-form 5.pdf

4693-KOLNP-2008-INTERNATIONAL EXM REPORT.pdf

4693-kolnp-2008-international publication.pdf

4693-kolnp-2008-international search report.pdf

4693-kolnp-2008-others pct form.pdf

4693-KOLNP-2008-OTHERS.pdf

4693-kolnp-2008-pct priority document notification.pdf

4693-kolnp-2008-pct request form.pdf

4693-KOLNP-2008-REPLY TO EXAMINATION REPORT.pdf

4693-kolnp-2008-specification.pdf

4693-KOLNP-2008-TRANSLATED COPY OF PRIORITY DOCUMENT.pdf

abstract-4693-kolnp-2008.jpg