JP7100847B2 - Power converter control device - Google Patents

Description

Application of Article 30, Paragraph 2 of the Patent Law, Institute of Electrical Engineers of Japan, Institute of Electrical Engineers of Japan, pp. 67-72 (SPC-16-38, MD-18-38), published on January 19, 2018. Joint Study Group on Semiconductor Power Conversion / Motor Drive, Institute of Electrical Engineers of Japan, held on January 20, 2018

The present invention relates to a control device for a power conversion device, and more particularly to a control device for a power conversion device that performs commutation by switching.

When a storage battery is used to stabilize the power system for the mass introduction of renewable power sources, a power conversion device is often used between the power system and the storage battery. As a power conversion device of this type, a high frequency isolated AC / DC converter to which a three-phase single-phase matrix converter (MC) is applied is currently attracting attention (see Non-Patent Document 1). This is a type of bidirectional isolated AC / DC converter, and is attracting attention because it can achieve high efficiency with a small number of conversions and can be configured without an electrolytic capacitor, which is advantageous for system miniaturization and long life. , Is being actively researched.

In Non-Patent Document 1, as a specific configuration example of a high-frequency isolated AC / DC converter to which a three-phase single-phase matrix converter is applied, a transformer of a three-phase single-phase matrix converter and a full-bridge type AC / DC converter is provided. An MC type dual active bridge (DAB) type bidirectional isolated AC / DC converter formed by sandwiching and connecting to each other is disclosed. This MC type DAB type bidirectional isolated AC / DC converter is characterized by the fact that surges due to transformer leakage inductance do not occur, soft switching is possible with all semiconductor elements, and a large interconnection reactor is not required. It is considered to be particularly advantageous in terms of miniaturization, high efficiency, and low cost. As a specific control method of the MC type DAB type bidirectional isolated AC / DC converter, as disclosed in Non-Patent Document 1, the output power is controlled by phase shift modulation (PSM). On the other hand, there is a method of controlling the output voltage of the matrix converter by pulse width modulation (PWM), and as a result, controlling the input current of the matrix converter into a sinusoidal waveform (hereinafter referred to as "PSM + PWM method"). Used.

Non-Patent Documents 2 to 6 disclose research results for realizing high-efficiency operation in a wide voltage range for DC / DC converters.

Koji Shigeuchi, 4 outsiders, "Improvement of Input Current Waveform of Bidirectional Insulated AC / DC Converter by Applying Matrix Converter", Institute of Electrical Engineers of Japan, Institute of Electrical Engineers of Japan, 2016, Vol. SPC-16-153, p. 25-30 F. Krismer, 2 outsiders, "Performance Optimization of a High Current Dual Active Bridge with a Wide Operating Voltage Range", 2006 37th American Electrical and Electronic Engineering Power Electronics Specialists Conference, 2006. PESC '06. 37th IEEE, p. 1-7 W. Choi, 2 outsiders, "Fundamental Duty Modulation of Dual-Active-Bridge Converter for Wide-Range Operation", IEEE Trans. ), June 2016, Vol. 31, No. 6, p. 4048-4064 A. K. Jain, 1 outside, "PWM Control of Dual Active Bridge (Comprehensive Analysis and Experimental Verification)", Institute of Electrical and Electronics Engineers (IEEE) , April 2011, Vol. 26, No. 4, p. 1215-1227 G. G. Oggier, 2 outsiders, "Modulation Strategy to Operate the Dual Active Bridge DC-DC Converter Under Soft Switching in the Whole Operating Range) ”, American Electrical and Electronic Engineering Society Bulletin (IEEE Trans.), April 2011, Vol. 26, No. 4, p. 1228-1236 H. Wen, 2 outsiders, "Nonactive Power Loss Minimization in a Bidirectional Isolated DC-DC Converter for Distributed Power Systems", Institute of Electrical and Electronics Engineers, USA IEEE Trans., December 2014, Vol. 61, No. 12, p. 6822-6831

By the way, the conventional MC type DAB type bidirectional isolated AC / DC converter has a problem that the power factor decreases and the switching loss increases when operating in a wide voltage range.

That is, according to the above-mentioned PSM + PWM method, for example, when the ratio of the power supply voltage on the primary side and the power supply voltage on the secondary side deviates from the turns ratio due to a decrease in the power supply voltage on the secondary side, a large circulating current flows through the transformer. It will be. Since this circulating current does not contribute to power transmission, the power factor decreases.

Further, a snubber capacitor is connected in parallel to each semiconductor element constituting the MC type DAB type bidirectional isolated AC / DC converter, whereby zero voltage switching (ZVS) is realized. If the direction in which the current flows changes due to the circulating current, ZVS does not hold and the switching loss increases.

Non-Patent Documents 2 to 6 are examples of DC / DC converters instead of MC-type DAB type bidirectional isolated AC / DC converters, but disclose techniques for realizing high-efficiency operation over a wide voltage range. There is.

Specifically, First, Non-Patent Document 2 discloses a technique called TRM (TRiangular current mode Modulation) using a triangular wave-shaped trans current. According to this technique, the circulating current of the DC / DC converter can be minimized, but on the other hand, there is a problem that ZVS and zero current switching (ZCS) coexist. The effect of reducing the switching loss of ZCS in the MC type DAB type bidirectional isolated AC / DC converter is limited. This is because the charging / discharging loss of the snubber capacitor connected in parallel with the semiconductor element occurs during switching other than ZVS. Therefore, in the MC type DAB type bidirectional isolated AC / DC converter, it is desirable to realize ZVS in all switching.

Non-Patent Document 3 discloses a modulation method that minimizes reactive power by using FDM modulation (Fundamental Duty Modulation) using a fundamental wave. Since this method minimizes only the fundamental wave component, the circulating current is larger than that of the TRM described in Non-Patent Document 2. However, unlike TRM, ZVS is established in all switching, so in many cases, the efficiency is higher than that of TRM.

Non-Patent Document 4 discloses the result of a comprehensive study on various operations that can be assumed as the operation of the dual active bridge. The analysis results such as the maximum value and effective value of the transformer current and the range of establishment of ZVS are included in it, and the highest efficiency modulation can be achieved by optimizing by applying various parameters to this analysis result. Although it may be realized, it is considered difficult to implement because the analysis results include many cases and non-linearity.

Non-Patent Documents 2 to 4 disclose that a period of zero voltage is provided on both the primary side and the secondary side. On the other hand, Non-Patent Documents 5 and 6 disclose a modulation method in which a zero voltage period is provided only in the converter on the high voltage side. In particular, in Non-Patent Document 6, a method of analyzing the formation range of ZVS and minimizing the reactive power in that range is used.

However, all the techniques disclosed in Non-Patent Documents 2 to 6 relate to DC / DC converters, and there is no description regarding the above-mentioned problems of MC type DAB type bidirectional isolated AC / DC converters. For MC type DAB type bidirectional isolated AC / DC converter, (1) it is desirable not to use approximation in order to reduce input current distortion, and (2) numerical calculation is required to not use approximation. However, there is a constraint that further increase in computational cost is not desirable, so a solution by another approach was needed.

Therefore, one of the objects of the present invention is to provide a control device for a power conversion device capable of suppressing a decrease in power factor when operating an MC type DAB type bidirectional isolated AC / DC converter in a wide voltage range. be.

Another object of the present invention is to provide a control device for a power conversion device capable of suppressing an increase in switching loss when operating an MC type DAB type bidirectional isolated AC / DC converter in a wide voltage range. There is something in it.

The control device of the power conversion device according to the first aspect of the present invention is connected to a transformer having first and second coils magnetically coupled to each other and a fifth node having one end constituting one end of a DC power supply, and the like. One end is connected to a first one-way switch element connected to a third node constituting one end of the second coil, one end is connected to a sixth node constituting the other end of the DC power supply, and the other end is connected. A second one-way switch element connected to the third node, one end connected to the fifth node and the other end connected to a fourth node constituting the other end of the second coil. An AC / DC converter having a third one-way switch element and a fourth one-way switch element having one end connected to the sixth node and the other end connected to the fourth node, and one end. A first bidirectional switch element connected to a seventh node corresponding to the first phase of three-phase alternating current, the other end connected to a first node constituting one end of the first coil, and one end thereof. A second bidirectional switch element connected to an eighth node corresponding to the second phase of three-phase alternating current, the other end corresponding to the first node, and one end corresponding to the third phase of the three-phase alternating current. A third bidirectional switch element connected to a ninth node and the other end connected to the first node, one end connected to the seventh node, and the other end to the other of the first coil. A fourth bidirectional switch element connected to a second node constituting an end, a fifth bidirectional switch element having one end connected to the eighth node and the other end connected to the second node. A matrix converter having a sixth bidirectional switch element, one end of which is connected to the ninth node and the other end of which is connected to the second node, and the first node and the first coil. A control device for a power conversion device having a reactor inserted between and a mode in which the first and fourth bidirectional switch elements are turned on at the same time, the second and fifth bidirectional switches. A first matrix converter is used to execute any one of a mode in which the elements are turned on at the same time and a mode in which the third and sixth bidirectional switch elements are turned on at the same time at least once in each cycle. It is a control device that controls periodically with a cycle.

In the control device of the power conversion device according to the second aspect of the present invention, in the control device according to the first aspect, the phase voltage of the first phase is larger than the phase voltage of the second phase, and the phase voltage of the second phase is described. The first mode in which the third and sixth bidirectional switch elements are turned on at the same time in each cycle when is larger than the phase voltage of the third phase, and the first and sixth bidirectional switches. The second mode in which the elements are turned on at the same time, the third mode in which the second and sixth bidirectional switch elements are turned on at the same time, and the third and sixth bidirectional switch elements are turned on at the same time. A fourth mode, a fifth mode in which the third and fourth bidirectional switch elements are turned on at the same time, and a sixth mode in which the third and fifth bidirectional switch elements are turned on at the same time. It is a control device configured to execute the seventh mode in which the third and sixth bidirectional switch elements are turned on at the same time in this order.

In the control device of the power conversion device according to the third aspect of the present invention, in the control device according to the second aspect, the first and fourth one-way switch elements are turned on in the first half of each cycle, and the second half of each cycle. It is a control device that periodically controls the AC / DC converter in the first cycle so that the second and third unidirectional switch elements are turned on.

The control device of the power conversion device according to the fourth aspect of the present invention is the control device according to the third aspect, and the duration of the first mode is such that the switching of the AC / DC converter and the matrix converter is a zero voltage. It is a control device determined to be switching.

In the control device of the power conversion device according to the fifth aspect of the present invention, in the control device according to the third aspect, the value of the current i _L flowing through the reactor at the start timing of the second mode is _IL0 , and the current i. The value at the timing when the first and fourth one-way switch elements of _L are turned on is _IL3 , the duration of the first mode is _dz / 2, the control cycle of the matrix converter and the AC / DC converter. Assuming that the phase difference of the control period is δ, when δ <d _z / 2, the first relationship between the d _z and the δ obtained by setting _{−IL0 = IL3} . It is a control device determined according to.

The control device of the power conversion device according to the sixth aspect of the present invention is the control device according to the fifth aspect, and when the dz is δ> _dz / 2, the first relationship and _{dz =} . A control device determined according to the second relationship between the d _z and the δ passing through the intersection with the 2 δ (δ ₁ , d _z 1) and the point where the current _IL is maximized (δ ₂ , 0). Is.

The control device of the power conversion device according to the seventh aspect of the present invention is a control device in which the _dz and the δ are determined by a dichotomy based on the first and second relationships.

According to the first to third aspects of the present invention, it is possible to suppress an increase in the circulating current when the output voltage on the AC / DC converter side (power supply voltage on the secondary side) drops, so that the MC type DAB type bidirectional When operating the isolated AC / DC converter in a wide voltage range, it is possible to suppress a decrease in the power factor.

According to the fourth to seventh aspects of the present invention, when the MC type DAB type bidirectional isolated AC / DC converter is operated in a wide voltage range, in addition to suppressing the decrease in power factor, the increase in switching loss is increased. It becomes possible to suppress.

It is a figure which shows the structure of the power conversion device 1 and the control device 4 by this embodiment. It is a figure which shows the high frequency equivalent circuit of the power conversion apparatus 1 shown in FIG. Command values I _dc ^* (= P ^* /) of voltage e _u , ev, e _w , _{v MC} , v _INV , current i _eu , i _ev , i _ew , I _dc , i _L , and current I _dc at the time of power running It is a signal waveform diagram which shows the simulation result of each waveform of V _dc ). It is an enlarged view of a part of FIG. 3 (for 0.01 seconds to 0.00014 seconds). It is a signal waveform diagram which shows typically the waveform at the time of power running of voltage v _MC , nv _INV and reactor current i _L in the case of nV _dc / e _M = 0.25. It is a signal waveform diagram which shows typically the waveform at the time of power running of voltage v _MC , nv _INV and reactor current IL in the case of nV _dc / e _{M =} 0.5. It is a figure which shows the detail of the control of a matrix converter 10. It is a figure which shows the detail of the control of a matrix converter 10. It is a figure which shows the detail of the control of a matrix converter 10. It is a figure which shows the detail of the control of a matrix converter 10. It is a figure which shows the detail of the control of a matrix converter 10. The figure which shows the state of the _{switch element Sw} Is. The figure which shows the state of the _{switch element Sw} Is. (A) is a diagram showing the relationship between δ and d _z in the case of nV _dc / e _M = 0.5, and (b) is a diagram showing the relationship between δ and d _z in the case of nV _dc / e _M = 0.25. It is a figure which shows the relationship of. (A) is a diagram showing the relationship between the current I _dc and the high frequency link power factor λ in the case of nV _dc / e _M = 0.5, and (b) is a diagram showing the relationship of nV _dc / e _M = 0.25. It is a figure which shows the relationship between the current I _dc and the high frequency link power factor λ in the case. It is a schematic block diagram which shows the functional block of a control device 4. It is a processing flow diagram which shows the calculation process of δ, d _m , d _z performed by the arithmetic unit 40. (A) is a diagram showing the power factor characteristics (Examples) for V _dc and I _dc obtained by the simulations shown in FIGS. 3 and 4, and (b) is shown in FIGS. 19 and 20. It is a figure which shows the power factor characteristic (comparative example) with respect to V _dc , I _dc obtained by the simulation. Results of simulating waveforms of voltage e eu, _ev , _ew and current i _eu , i _ev , i _ew , i _L , _{i u at} the time of power running under the same conditions as in FIG. 3 except that d _z = 0. It is a signal waveform diagram which shows. Results of simulating waveforms of voltage e eu, _ev , _ew and current i _eu , i _ev , i _ew , i _L , _{i u at} the time of power running under the same conditions as in FIG. 4 except that d _z = 0. It is a signal waveform diagram which shows.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 is a diagram showing a configuration of a power conversion device 1 and a control device 4 thereof according to the present embodiment. As shown in the figure, the power conversion device 1 according to the present embodiment is an MC type having a three-phase single-phase matrix converter 10, a transformer 20, a full-bridge type AC / DC converter 30, and a protection circuit 35. It is a DAB type bidirectional isolated AC / DC converter. The matrix converter 10, the transformer 20, and the AC / DC converter 30 are connected between the system power supply 2 and the load 3 in this order. In the following, the three ends of the matrix converter 10 connected to the grid power supply 2 are referred to as a _seventh node n7, an _eighth node n8, and a _ninth node n9, respectively, and a matrix connected to the transformer 20. The two ends of the converter 10 are referred to as the _first node n1 and the _second node n2, respectively, and the two ends of the AC / DC converter 30 connected to the transformer 20 are referred to as the _third node n3, respectively. It is referred to as a _fourth node n4, and the two ends of the AC / DC converter 30 connected to the load 3 are referred to as a _fifth node n5 and a _sixth node n6, respectively. Further, the voltage of the first node n ₁ with respect to the voltage of the second node n ₂ is referred to as a voltage v _MC , and the voltage of the third node n ₃ with respect to the voltage of the fourth node n ₄ is referred to as a voltage v _INV .

The transformer 20 is configured to have two coils 20a and 20b (first and second coils) that are magnetically coupled to each other. The turns of the coils 20a and 20b are _NP and _NS , respectively, and the turns ratio n of the coils 20a and 20b is n = N _P / _NS . One end of the coil 20a is connected to the matrix converter 10 at the _first node n1 via the reactor L, and the other end of the coil 20a is connected to the matrix converter 10 at the _second node n2. One end of the coil 20b is connected to the AC / DC converter 30 at the _third node n3, and the other end of the coil 20b is connected to the AC / DC converter 30 at the _fourth node n4. Hereinafter, the current flowing through the reactor L is referred to as a reactor current i _L. Regarding the direction of the reactor current i _L , the direction from the _first node n1 to the _second node n2 is the positive direction, and the opposite is the negative direction. The reactor current i _L is expressed as a function of time t as in the following equation (1) when the voltage v _MC and nv _INV are used.

The AC / DC converter 30 is a device that mutually converts a single-phase AC voltage (coil 20b side) and a DC voltage (load 3 side). The load 3 is, for example, a power converter for driving a storage battery or a motor of a hybrid car, and is operated (charged) by DC power supplied from the power conversion device 1 (power running), and conversely, power conversion. There is a case where DC power is supplied to the device 1 (regeneration). One end of the load 3 is connected to the AC / DC converter 30 at the _fifth node n5, and the other end of the load 3 is connected to the AC / DC converter 30 at the _sixth node n6. Hereinafter, the voltage of the fifth node n ₅ with respect to the voltage of the sixth node n ₆ is referred to as a voltage V _dc , and the current flowing from the fifth node n ₅ toward the load 3 is referred to as a current I _dc .

The AC / DC converter 30 has a switch element G1 ( _first one-way switch element) having one end connected to the _fifth node n5 and the other end connected to the _third node n3, and one end having a first. A switch element G ₂ (second one-way switch element) connected to node n ₆ of 6 and the other end connected to a third node n ₃ , and one end connected to a fifth node n ₅ and others. A switch element G3 ( _third one-way switch element) whose end is connected to the _fourth node n4, one end connected to the _sixth node n6, and the other end connected to the _fourth node n4. It has a switch element G ₄ (fourth one-way switch element) and a capacitor C1 having one end connected to the fifth node n ₅ and the other end connected to the sixth node n ₆ . Will be done.

_The switch elements G1 to G4 are semiconductor elements such as MOSFET (Metal-Oxide-Semiconductor Field-Effect Transistor) and IGBT (Insulated Gate Bipolar Transistor), _respectively , and diodes and snubber capacitors connected in parallel with the semiconductor elements. It is a one-way switch configured to include and. The snubber capacitor may be a parasitic capacitance of a semiconductor element or an independent capacitive element. The switch element G ₁ is incorporated into the circuit so that the diode anode is connected to the third node n ₃ , and the switch element G ₂ is circuit so that the diode cathode is connected to the third node n ₃ . The switch element G3 is incorporated into the circuit such that the diode anode is connected to the _fourth node n4, and the switch element _G4 is connected to the diode cathode to the _fourth node _n4 . It is incorporated in the circuit so as to be.

An individual control signal is supplied from the control device ₄ to each of the control electrodes of the semiconductor elements constituting the switch elements G1 to _G4 . Each control signal is a signal that takes either a high value or a low value. The control device ₄ individually controls the on / off states of the switch elements G1 to _G4 by individually controlling the values of these control signals.

The matrix converter 10 is a device that mutually converts a three-phase AC voltage (system power supply 2 side) and a single-phase AC voltage (coil 20a side). The grid power supply 2 is a three-phase AC power supply that generates AC voltages _eu , _ev , and _ew represented by sinusoidal signals that are out of phase by 2π / 3, and is, for example, a commercial power supply. When the AC voltage e _u , _ev , and _ew are expressed by a mathematical formula, the following equation (2) is obtained. However, E is the effective value (constant) of the voltage between the input lines, and ωt is the phase angle (constant) of the input voltage.

As shown in FIG. 1, the output end of the system power supply 2 corresponding to the AC voltage e _u corresponding to the u phase (first phase) is connected to the matrix converter 10 at the _seventh node n7, and the v phase (second phase). The output end of the system power supply 2 corresponding to the AC voltage _ev corresponding to the phase) is connected to the matrix converter 10 at the _eighth node n8, and corresponds to the AC voltage ew corresponding to the _w phase (third phase). The output end of the grid power supply 2 is connected to the matrix converter 10 at the _ninth node n9. Hereinafter, the currents flowing through the _seventh to _ninth nodes n7 to _n9 are referred to as currents _ieu , _iev , and iew, respectively.

The matrix converter 10 has a switch element _Sup (first bidirectional switch element) having one end connected to the _seventh node n7 and the other end connected to the first node n1 and _one end having an eighth node. A switch element _Svp (second bidirectional switch element) connected to node _n8 and the other end connected to the first node n1 and _one end connected to the _ninth node n9 and the other end A switch element _Swp (third bidirectional switch element) connected to the first node n ₁ and one end connected to the seventh node n ₇ and the other end connected to the second node n ₂ . Switch element Sun (fourth bidirectional switch element) and switch element S _vn (fifth bidirectional) with one end connected to the _eighth node n8 and the other end connected to the _second node _n2 . Switch element), switch element _Swn (sixth bidirectional switch element) having one end connected to the _ninth node n9 and the other end connected to the _second node n2, and an AC reactor Lf. It is configured to have a damping resistor Rf and an input capacitor Cf.

The AC reactor Lf includes an inductor inserted between the seventh node n ₇ and the tenth node n ₁₀ which is a connection point between the switch elements _Up and _Sun , and the eighth node n ₈ and the switch element S. The inductor inserted between the eleventh node n ₁₁ which is the connection point of _vp and S _vn , and the twelfth node n ₁₂ which is the connection point between the ninth node n ₉ and the switch elements _Swp and _Swn . It is composed of an inductor inserted between and. The damping resistor Rf is composed of three resistance elements connected in parallel with each of these inductors. The input capacitor Cf is composed of three capacitors connected to each other by delta connection or star connection, and the three connection points are connected to the _tenth to _twelfth nodes n10 to n12, respectively. Hereinafter, the currents flowing through the _tenth to _twelfth nodes n10 to n12 are referred to as currents i _u , _iv , and i _w , respectively.

The switch elements _{Sup, Svp , Swp} , _Sun , _Svn , and _Swn are bidirectional switches including two switch elements connected in series, respectively. Specifically, the switch element _Sup is composed of switch elements _Gpu and _Gup connected in series and a snubber capacitor connected in parallel with these. Further, the switch element S _vp is composed of switch elements G _pv , G _vp connected in series and a snubber capacitor connected in parallel with these, and the switch element _{Sw p} is a switch element G connected in series. It is composed of _pw , G _pp and a snubber capacitor connected in parallel with them, and the switch element _Sun is a switch element G _nu , _Gun connected in series and a snubber capacitor connected in parallel with them. The switch element S _vn is composed of switch elements G _nv , G _vn connected in series and a snubber capacitor connected in parallel with these, and the switch element S _vn is a switch connected in series. It is composed of elements G _nw and G _wn and a snubber capacitor connected in parallel with them. These switch elements G _pu , G _up , G _pv , G v p, G _{p w, G pp, G nu, Gun, G nv, G vn , G nw , and G n are} semiconductor devices such as MOSFETs and IGBTs, respectively. This is a one-way switch including a semiconductor element and a diode connected in parallel.

The switch elements G _pu and G _up are connected between the first node n ₁ and the tenth node n ₁₀ in this order and in the direction in which the anodes of the respective diodes are connected to each other. The switch elements G _pv and G _vp are connected between the first node n ₁ and the eleventh node n ₁₁ in this order and in the direction in which the anodes of the respective diodes are connected to each other. The switch elements G _pw and G _wp are connected between the first node n ₁ and the twelfth node n ₁₂ in this order and in the direction in which the anodes of the respective diodes are connected to each other. The switch elements G _nu and _Gun are connected between the second node n ₂ and the tenth node n ₁₀ in this order, and in the direction in which the anodes of the respective diodes are connected to each other. The switch elements G _nv and G _vn are connected between the second node n ₂ and the eleventh node n ₁₁ in this order and in the direction in which the anodes of the respective diodes are connected to each other. The switch elements G _nw and G _wn are connected between the second node n ₂ and the twelfth node n ₁₂ in this order and in the direction in which the anodes of the respective diodes are connected to each other. Regarding the connection of the two switch elements that make up the bidirectional switch, here it is assumed that the anodes of the respective diodes are connected in the direction in which they are connected to each other, but the direction in which the cathodes of the respective diodes are connected to each other. It may be connected by.

The control electrodes of the semiconductor elements constituting the switch elements G _pu , G _up , G _pv , G _vp , G _pw , G _wp , G _nu , _{Gun, G nv ,} G _vn , G _nw , and G _{w n} are Individual control signals are supplied from the control device 4. Each control signal is a signal that takes either a high value or a low value. The control device 4 individually controls the values of these control signals to switch elements G _pu , G _up , G _pv , G _vp , G _{p w} , G _wp , G _n u, _{Gun, G nv ,} G _vn . , G _nw , G _wn Each on / off state is controlled individually.

The protection circuit 35 is a circuit for absorbing the reactor current _IL at the time of commutation failure, and is a rectifier circuit consisting of four diodes connected in full bridge, a smoothing capacitor and a load connected in parallel with this rectifier circuit. It is composed of and. The voltage across the load is V _p . The two input ends of the rectifier circuit are connected to the _first node n1 and the _second node n2, respectively.

FIG. 2 is a diagram showing a high frequency equivalent circuit of the power conversion device 1. As shown in the figure, when the power conversion device 1 operates at a high frequency, a power supply circuit that outputs a voltage v _MC and a power supply circuit that outputs a voltage nv _INV are connected in series with a reactor L interposed therebetween. It is equivalent to the circuit that was made.

Returning to FIG. 1, the control device 4 switches based on the power command value P ^* and the power factor angle command value α ^* supplied from the outside and the respective values of the voltages e _u , _{ev, e w ,} and V _dc . This is a device that controls the _on / off state of _each of the elements Sup, _Svp , _Swp , _{Sun, Svn , Swn} , and G1 to _G4 . By performing this control, the control device 4 applies the power specified by the power command value P ^* to the load 3 from the system power supply 2 at the power factor angle specified by the power factor angle command value α ^* (in the case of power running). ) Or, transmission from the load 3 to the system power supply 2 (in the case of regeneration) is realized. The operation of the control device 4 during power running and regeneration is basically the same except that the signs of the phase difference δ, which will be described later, are exchanged. Therefore, in the following, the explanation will be continued focusing on the time of powering.

FIG. 3 shows the command values I _dc ^* of the voltage e _u , e _v , e _w , v _MC , v _INV , the current i _eu , i _ev , i _ew , I _dc , i _L , and the current I _dc at the time of power running. It is a signal waveform diagram which shows the simulation result of each waveform of = P ^* / V _dc ). The detailed simulation conditions are as shown in Table 1. FIG. 3 shows one cycle of three-phase alternating current represented by the voltages e _u , _ev , and _ew . FIG. 4 is an enlarged view of a part of FIG. 3 (for 0.01 to 0.00014 seconds).

One cycle of three-phase alternating current can be divided into 12 spaces I to XII as shown in FIG. 3 according to the phases of the voltages e _u , _ev , and _ew . Specifically, the space I whose phase of the voltage e _u is 0 or more and less than π / 6, the space II where π / 6 or more and less than π / 3, and the space III and π / where π / 3 or more and less than π / 2 are present. Space IV that is 2 or more and less than 2π / 3, space V that is 2π / 3 or more and less than 5π / 6, space VI that is 5π / 6 or more and less than π, space VII that is π or more and less than 7π / 6, 7π / 6 or more Space VIII that is less than 4π / 3, space IX that is 4π / 3 or more and less than 3π / 2, space X that is 3π / 2 or more and less than 5π / 3, space XI that is 5π / 3 or more and less than 11π / 6, It can be divided into spaces XII which are 6 or more and less than 2π. The waveform shown in FIG. 4 corresponds to a part of the space IV.

In the operation of the control device 4 in each of the spaces I to XII, the switch elements to be controlled are switched according to the magnitude relation of the voltages e _{u, ev, and ew, and the voltage e u , e v , and ew are} operated. It is basically the same except that the output voltage may change depending on the sign of the one that takes an intermediate value. Therefore, in the following, the explanation will be continued focusing on the space IV ( _{eu>ev> e w ) .}

5 and 6 are signal waveform diagrams schematically showing the waveforms of the voltage v _MC , nv _INV and the reactor current i _L at the time of power running, respectively. FIG. 5 shows an example in which the value of the voltage V _dc shown in FIG. 1 is relatively small (specifically, nV _dc / _eM = 0.25), and FIG. 6 shows an example shown in FIG. An example in which the value of the voltage V _dc is relatively large (specifically, nV _dc / _eM = 0.5) is shown. The voltages _em and _em shown in FIGS. 5 and 6 are represented by the following equations (3) and (4), respectively. However, the current _iv ^* shown in the equation (3) is a command value of the current _iv . The command values i _u ^* , i _v ^* , i _w ^* of the currents i _u , i _v , i _w are expressed by the equation (5).

Explaining the outline of the operation of the control device 4 with reference to FIG. 5, first, the control device 4 is configured to periodically control the matrix converter 10 in the cycle T (first cycle). The time t ₁ shown in FIG. 5 corresponds to the beginning of this control cycle, and the time t ₁₀ corresponds to the end. Details of the control of the matrix converter 10 will be described later.

Further, the control device 4 is configured to periodically control the AC / DC converter 30 with the same period T (first period). The time t ₂ shown in FIG. 5 corresponds to the beginning of this control cycle, and the time t ₁₁ corresponds to the end. Specifically, the control device 4 turns on the switches G1 and G4 in the first half of each cycle, turns on the switches G2 and G3, turns on the switches G2 and G3 in the latter half of each cycle, and turns on the switches G1 and G4. It is configured to do. Therefore, as shown in FIG. 5, the value of the voltage nv _inv becomes nV _dc in the first half of each cycle and becomes −nV _dc in the latter half of each cycle. In the following description, the elapsed time length from the start of the control cycle of the matrix converter 10 to the start of the control cycle of the AC / DC converter 30 that arrives for the first time after that is referred to as a phase difference δ.

7 to 11 are diagrams showing details of control of the matrix converter 10. The dashed line with the arrow shown in the figure represents the path and direction of the current _IL . Hereinafter, the control of the matrix converter 10 will be described in detail with reference to these figures together with FIG. Actually, the commutation operation is performed when switching the on / off of each switch element, but here, the commutation operation will be ignored. The commutation operation will be described in detail later with reference to FIGS. 12 and 13.

First, as shown in FIG. 5, the control device 4 is configured to control the matrix converter 10 by sequentially executing seven modes from the first mode MODE1 to the seventh mode MODE7 in each cycle. To. The durations (duties) of the first mode MODE1 and the seventh mode _MODE7 are the same as each other, and in the following, this duration is referred to as dz / 2. The duration of the fourth mode _MODE4 is twice (= dz) the duration of the first mode MODE1 and the seventh mode MODE7. The durations of the third mode MODE3 and the sixth mode _MODE7 are also the same as each other, and in the following, this duration is referred to as dm. The durations of the second mode MODE2 and the fifth mode MODE5 are also the same (= T / 2-d _{z −dm} ). In the example of FIG. 5, δ < _dz / 2, and the start of the control cycle of the AC / DC converter 30 arrives during the execution of the first mode MODE1.

FIG. 7A shows the states of the switch elements _Sup , _Svp , _Swp , _Sun , _Svn , and _Swn at the time t1 which is the start of the _first mode MODE1. Further, FIG. _7B shows the states of the switch elements _Sup , _Swp , _Swp , _Sun , _Svn , and _Swn at time t2, which is the start of the control cycle of the AC / DC converter 30. ing. As shown in these figures, in the first mode MODE1, the control device 4 turns on the switch elements _Swp and _Swn , and turns off the other switch elements _Swp , _Swp , _Sun and _Sven . do. As a result, the first node n ₁ and the second node n ₂ are short-circuited, so that the reactor L is in a state where only the voltage nv _INV is applied. Since nv _INV = −nV _dc at time t ₁ to time t ₂ , the current i _L changes in an increasing direction as shown in FIG. On the other hand, since nv _INV = nV _dc at time t ₂ to time t ₃ , the current i _L changes in a decreasing direction as shown in FIG.

FIG. _8A shows the states of the switch elements _Sup , _Svp , _Swp , _Sun , _Svn , and _Swn at time t3, which is the start of the second mode MODE2. As shown in the figure, in the second mode MODE2, the control device 4 turns on the switch elements _{Sup and Swn, and turns off the other switch elements Svp , Swp , Sun} and _Sven . .. As a result, the voltage e _M is applied between the first node n ₁ and the second node n ₂ , so that the reactor L is in a state where the voltage e _M −nv _INV is applied. Then, from time t ₃ to time t ₄ , nv _INV = nV _dc , but the voltage e _M is generally set to a value larger than the voltage nV _dc (e _M > nV _dc ), so as shown in FIG. The current i _L changes in an increasing direction.

FIG. _8B shows the states of the switch elements _Sup , _Svp , _Swp , _Sun , _Svn , and _Swn at time t4, which is the start of the third mode MODE3. As shown in the figure, in the third mode MODE3, the control device 4 turns on the switch elements S _{bp and Swn ,} and turns off the other switch elements _Sup , _{Sw p} , _Sun and S _vn . .. As a result, the voltage em is applied between the first node _{n 1} and the second node n ₂ , so that the reactor L is in a state where the voltage em _{−nv INV} is applied. Then, from time t ₄ to time _{t 5} , nv _INV = nV _dc , but the voltage em is generally set to a value larger than the voltage nV _dc ( _em > nV _dc ), so as shown in FIG. The current i _L changes in an increasing direction. However, since the voltage em is smaller than the voltage _em (e _M > _{em )} , the rate of increase of the current i _L is higher than the rate of increase from time t ₃ to time t ₄ as shown in FIG. Becomes smaller.

FIG. _9A shows the states of the switch elements _Sup , _Svp , _Swp , Sun, _Svn , and _Swn at time t5, which is the start of the fourth mode _MODE4 . Further, in FIG. 9B, each switch element _Sup , _Svp , _Swp , S at the time t ₆ (t ₆ −t ₂ = T / ₂ ), which is the switching timing of the switch elements G1 to _G4 , is shown. The states of _un , S _vn , and _Swn are shown. As shown in these figures, in the fourth mode _MODE4 , the control device 4 turns on the switch elements _Swp and _Swn , and turns off the other switch elements _Swp , _Swp , Sun and _Sven . do. This is the same state as the first mode MODE1. Therefore, since the first node n ₁ and the second node n ₂ are short-circuited, the reactor L is in a state where only the voltage nv _INV is applied. Since nv _INV = nV _dc from time t ₅ to time t ₆ , the current i _L changes in a decreasing direction as shown in FIG. On the other hand, since nv _INV = −nV _dc at time t ₆ to time t ₇ , the current i _L changes in an increasing direction as shown in FIG.

FIG. 10A shows the states of the switch elements _Sup , _Svp , _Swp , _Sun , _Svn , and _Swn at time _t7 , which is the start of the fifth mode MODE5. As shown in the figure, in the fifth mode MODE5, the control device 4 turns on the switch elements _Swp and _Sun , and turns off the other switch elements _Swp , _Svp , _Svn and _Swn . .. As a result, the voltage- _eM is applied between the _first node n1 and the _second node n2, so that the reactor _L is in a state where the voltage-eM-nv _INV is applied. Then, since nv _INV = −nV _dc from time t ₇ to time t ₈ , assuming that e _M > nV _dc holds as described above, the current i _L changes in a decreasing direction as shown in FIG. do.

FIG. _10B shows the states of the switch elements _Sup , _Svp , _Swp , Sun, _Svn , and _Swn at time t8, which is the start of the sixth mode _MODE6 . As shown in the figure, in the sixth mode _MODE6 , the control device 4 turns on the switch elements _Swp and _Swn , and turns off the other switch elements _Swp , _Swp , Sun and _Swn . .. As a result, the voltage −em is applied between the first node _{n 1} and the second node n ₂ , so that the reactor L is in a state where the voltage _−m −nv _INV is applied. Then, since nv _INV = −nV _dc at time t ₈ to time _{t 9} , assuming that em> nV _dc holds as described above, the current i _L changes in a decreasing direction as shown in FIG. do. However, since e _M > _em holds as described above, the reduction rate of the current _IL is smaller than the reduction rate at time _t7 to time t8 as shown _in FIG.

FIG. 11 _shows the states of the switch elements _Sup , _Svp , _Swp , Sun, _Svn , and _Swn at time t9, which is the start of the seventh mode _MODE7 . As shown in the figure, in the seventh mode _MODE7 , the control device 4 turns on the switch elements _Swp and _Swn , and turns off the other switch elements _Swp , _Swp , Sun and _Sven . .. This is the same state as the first mode MODE1. Therefore, since the first node n ₁ and the second node n ₂ are short-circuited, the reactor L is in a state where only the voltage nv _INV is applied. Since nv _INV = −nV _dc from time t ₉ to time t ₁₀ , the current i _L changes in an increasing direction as shown in FIG.

As described above, in the present embodiment, the mode in which the switch elements _Swp and _Swn are turned on at the same time (that is, the mode in which _VMC = 0. Specifically, the first, fourth, and fourth modes are set. Since the mode MODE1, MODE4, MODE7) of 7 is performed once or more in each cycle, it is possible to suppress an increase in the circulating current when the output voltage nV _dc on the AC / DC converter 30 side decreases. Therefore, when the power conversion device 1 is operated in a wide voltage range, it is possible to suppress a decrease in the power factor.

The arithmetic mean values of the transmission power P and the current _iv obtained as a result of the control device 4 performing the above control can be obtained by using the equation (1) and the equivalent circuit of FIG. (6) and (7). However, P ₁ in the formula (6) is as shown in the formula (8). Further, f is a high frequency link frequency.

Next, the commutation operation performed when the switch elements _{Sw, Swp , Swp} , _Sun , _Svn , and _Swn are switched on and off will be described.

Switch element _Swp , _Swp , _Swp , _Sun , _Swn , _Swn on / off switching is always one of the switch elements _Swp , _Swp , _Swp , _Sun , _Swn , _Swn . This is executed with one of the switch elements _Sup , _Svp , _Swp , _Sun , _Svn , and _Swn as the commutation source, and any other one as the commutation destination. Then, the control device 4 turns off one of the two switch elements constituting the bidirectional switch element of the commutation source, and one of the two switch elements constituting the bidirectional switch element of the commutation destination. On, turn off the remaining one of the two switch elements that make up the bidirectional switch element of the commutation source, and turn on the remaining one of the two switch elements that make up the bidirectional switch element of the commutation destination. , The commutation operation is executed in this order. Of the two switch elements that make up the bidirectional switch element of the commutation source, the switch element that is turned off first flows current _IL through a diode arranged in parallel with the switch element even after it is turned off. It is selected according to the direction of the current _IL at the start of the commutation operation so that the commutation operation can be started. Similarly, of the two switch elements constituting the bidirectional switch element of the commutation destination, the switch element that is turned on first is in parallel with the other switch element without turning on the other switch element. This is also selected according to the direction of the current i _L at the start of commutation so that the current i _L can flow through the arranged diode.

12 and 13 are diagrams showing the states of the switch elements _{Sup and Swp when} switching from the first mode MODE1 to the second mode MODE2 described above, respectively. In this case, the switch element _Swp is the commutation source, and the switch element _Sup is the commutation destination. Other switch elements _Svp , _{Sun, Svn ,} and _Swn that are not particularly related to the commutation operation are not shown.

FIG. 12 shows a case where the current IL in the state immediately before the start of the commutation operation (the state shown in FIG. _12A ) is negative. In this case, the control device 4 first turns off the switch element _GPw of the commutation source as shown in FIG. 12 (b). The current i _L then passes through the diode in the switch element _GPw and flows as before.

Next, the control device 4 _turns on the switch element Group of the commutation destination as shown in FIG. 12 (c). Here, in the snubber capacitor in the switch element _Swp , a charge having a negative electrode on the reactor L side is accumulated in advance. Since no voltage is applied to both ends of the switch element _Gup due to the blocking action due to this electric charge, the on operation of the switch element _Gup becomes a so-called soft turn-on, and ZVS is established.

Next, the control device 4 turns off the switch element G _wp of the commutation source as shown in FIG. 12 (d). In this case, since the voltage rise across the switch element G _wp is alleviated by the _snubber capacitor in the switch element _Swp and the Switch, the off operation of the switch element G _wp becomes a so-called soft turn-off, and ZVS is established. After the switch element G _wp is turned off, the electric charge stored in the snubber capacitor is discharged, and when the discharge is completed, a current starts to flow in the switch element G _up as shown in FIG. 12 (e).

Finally, the control device 4 turns on the switch element G _pu of the commutation destination as shown in FIG. 12 (f). This completes the commutation operation.

FIG. 13 shows a case where the current IL in the state immediately before the start of the commutation operation (the state shown in FIG. _13A ) is positive. In this case, the control device 4 first turns off the switch element G _wp of the commutation source as shown in FIG. 13 (b). The reason why the switch element G _wp is turned off instead of the switch element G _pw is that the current does not flow in the switch element _Swp otherwise. The current i _L after the switch element G _wp is turned off passes through the diode in the switch element G _wp and flows in the same manner as before.

Next, the control device 4 turns on the switch element G _pu at the commutation destination as shown in FIG. 13 (c). In this case, since the commutation of the current i _L occurs at the same time as the switch element G _pu is turned on, the on operation of the switch element G _pu becomes a so-called hard turn-on and ZVS is not established.

Next, the control device 4 turns off the switch element _GPw of the commutation source as shown in FIG. 13 (d). Since the current i _L has already been commutated at this point, this off operation becomes a so-called soft turn-on and ZVS is established.

Finally, the control device 4 _turns on the switch element Group of the commutation destination as shown in FIG. 13 (e). This completes the commutation operation.

The above is the commutation operation in the matrix converter 10. However, since it takes a certain amount of time to switch each switch element, such commutation operation is performed, so that the power supply is short-circuited or the inductor current is cut off during commutation. Is likely to occur. By carrying out the above commutation operation, it is possible to prevent the occurrence of such a situation. Since the same situation can occur in the AC / DC converter 30, the control device 4 also performs a commutation operation in the AC / DC converter 30. Specifically, when the switch elements G1 to G4 are switched _on and off, _a time (dead time) for turning off all of the switch elements G1 to _G4 is provided for _a certain period of time.

As described with reference to FIGS. 12 and 13, _ZVS may or may not be established depending on the direction of the current IL during the commutation operation. If the power conversion device 1 is operated in a state where ZVS is not established, the switching loss becomes large, so it is preferable that ZVS is always established. Here, the direction of the current _IL during the commutation operation can be controlled by the above-mentioned phase difference δ and the values of the duty d _m and d _z . Therefore, in the control device 4 according to the present embodiment, the values of δ, dm, d _z are determined so that ZVS is established and the power factor as large as possible can be obtained, and the determined δ, _dm , _{d z} . The matrix converter 10 and the AC / DC converter 30 are controlled according to the above. Specifically, it is configured to determine the values of δ, dm, and _{d z} so as to satisfy the following equation (9).

X, Y, and _Z in the formula (9) are values represented by the following formula (10) or the formula (11) according to the relationship between δ and dz. However, M = e _M / nV _dc and _m = em / nV _dc . Further, d _z1 is a value expressed as in the equation (12) using d _m1 obtained by the equation (14). δ ₁ is a value of 1/2 of d _z 1 as shown in the equation (13). A, B, and C in the equation (14) are expressed as in the equation (15).

Note that d _m1 shown in the equation (14) is a value that satisfies the following equation (16) when 2δ ₌ dz, and the equations (6), (7), and (13) are substituted into the equation (16). It is obtained by further substituting the equation (12) into the equation (17) obtained by and solving the result for _dm1 . However, | _iv (δ ₁ , d _m1 , d _z1 ) | and P (δ ₁ , d _m1 , d _z1 ) in the equation (16) are δ = δ in the equations (7) and (6), respectively. Represents a value obtained by substituting ₁ , d _m = d _m1 and d _z = d _z1 .

Hereinafter, the derivation method of the equation (9) will be described in detail.

When the condition that ZVS is established in the power conversion device 1 is generalized, when the potential of the _first node n1 rises after switching (in the example of FIG. 5, for example, time t3; in the example of FIG. ₆ , for example, time t). In ₂ ), it is a condition that ZVS is established that a current is flowing into the first node n ₁ , and the potential of the first node n ₁ drops after switching (in the example of FIG. 5, in the example of FIG. 5). For example, at times t ₄ and t _5. In the example of FIG. 6, for example, at times t ₄ and t ₅ ), it is a condition for establishing ZVS that a current is flowing into the first node n ₁ . Therefore, using the current values _IL0 to _IL3 shown in FIGS. 5 and 6, it can be said that ZVS is established when the following equation (18) is satisfied. However, the current value _IZVS is the minimum current value at which the snubber capacitor can be completely charged and discharged during the commutation operation of the matrix converter 10 or the dead time of the AC / DC converter 30.

The current value _IL0 is the value of the current i _L when switching from the first mode MODE 1 to the second mode MODE 2, and the current value IL ₁ is the current i when switching from the second mode MODE 2 to the third mode MODE 3. The value of _L , the current value IL2 is the value of the current iL when switching from the third mode MODE3 to the _fourth mode _MODE4 , and the current value _IL3 is _a voltage controlled by the switch elements G1 to _G4 . It is the value of the current i _L when the value of nv _inv changes to nV _dc . When the current values _IL0 to _IL3 are specifically expressed by a mathematical formula, the following equation (19) or equation (20) is obtained depending on the relationship between δ and _dz . However, the values shown in these equations are the values standardized by I _BASE = V _dc / 4fL.

Hereinafter, it is considered to find the relationship of δ, dm, and _{d z} such that the equation (18) is satisfied.

FIG. 14 is a diagram showing the relationship between δ and _dz . FIG. 14 (a) shows the case of the same nV _dc / e _M = 0.5 as in FIG. 6, and FIG. 14 (b) shows the case of the same nV _dc / e _M = 0.25 as in FIG. However, assuming that the influence of _dm is small, _dm = 0 is set here.

The "optimal relationship" shown in FIG. 14 is the result of finding the relationship of δ, dm (= 0), and _{d z} that can obtain a high power factor by simulation using each parameter shown in Table 1 above. Is. Therefore, the purpose here is to find the relationship of δ, dm, _{d z} that is as close as possible to this “optimal relationship” and that satisfies Eq. (18). Hereinafter, it will be described in detail.

First, focusing on the region of δ <dz / 2, it can be considered that the equation (18) always _holds from the equation (19) for the current values _IL1 and _IL2 in the range of the normal _IZVS . On the other hand, for the current values _IL0 and _IL3 , from equation (19), _IL3 - _IL0 = d _z -2δ, and this value is always positive ((d _z ) under the condition of δ <d _z / 2]. Since -2δ)> 0), as can be understood from FIG. 5, for example, if _-IL0 = _IL3 , the equation (18) holds. Therefore, by solving the equation (19) obtained by substituting the equation (19) into −I _{L0 = IL3} for dz, the equation (9) (X, Y, Z are shown in the equation (10)) is obtained.

In each of FIGS. 14 (a) and 14 (b), the upper left half is a region corresponding to δ <d _z / 2. As shown in these figures, at least in this region, the relationship of δ, _dm , _dz obtained by Eq. (9) and the “optimal relationship” are in good agreement. Therefore, it is understood that equation (9) may be used as the relationship (first relationship) of δ, dm, and _{d z} when at least δ <d _z / 2 holds.

Next, focusing on the region of δ> d _z / 2, as can be understood from FIG. 14, in the “optimal relationship”, d _z decreases monotonically with the increase of δ. Further, the current i _L becomes maximum when (δ, d _z ) = (δ ₂ , 0) (however, δ ₂ = 0.5), so that this point can be passed through δ, d. It is preferable to determine the relationship between _m and _dz . Further, it is preferable that the relationship of δ, d _m , and d _z does not become discontinuous at the boundary between the region of δ <d _z / 2 and the region of δ> d _z / 2.

Therefore, by a straight line passing through the intersection (δ ₁ , d _z 1) between the equation (9) and d _{z =} 2 δ and the point (0, 0.5) where the current IL becomes maximum, δ, d _m , d. The relationship of _z will be determined. The equation (9) obtained by substituting X, Y, and _Z shown in the equation (11) represents the relationship (second relationship) of δ, _dm , dz thus determined. .. By doing so, as can be understood from FIG. 14, even in the region corresponding to δ> _dz / 2, the point (0,0) where the “optimal relationship” is almost the same and the current _IL is maximized. A relationship is obtained that passes through the case of .5) and the point (d _z1 , δ ₁ ) for continuity with the case of δ <d _z / 2.

In addition, although detailed calculation is omitted, when the current values _IL0 to _IL3 are calculated based on the equation (9) obtained by substituting X, Y, Z shown in the equation (11), the equation ( A result satisfying 18) is obtained. Therefore, even when δ> d _z / 2 holds, it can be said that the equation (9) may be used as the relationship between δ, dm, and _{d z} .

The derivation method of the equation (9) has been described above. Next, the validity of equation (9) will be examined from the viewpoint of power factor.

FIG. 15 is a diagram showing the relationship between the current I _dc and the high frequency link power factor λ of the power conversion device 1. FIG. 15A shows the same case of nV _dc / _eM = 0.5 as in FIG. 6, and FIG. 15B shows the same case of nV _dc / _eM = 0.25 as in FIG. However, assuming that the influence of _dm is small, _dm = 0 is set here. Further, the high frequency link power factor λ according to the equation (9) was obtained by the following equation (21). However, i _Lrms is an effective value of the current i _L calculated from the equations (19) and (20). Further, the current I _dc was obtained by dividing the transmission power P obtained from the equation (6) by the voltage V _dc .

The "optimal relationship" shown in FIG. 15 corresponds to the "optimal relationship" shown in FIG. The figure also shows the high frequency link power factor λ in the case of _{dz =} 0 (that is, when the control cycle of the matrix converter 10 does not include the period in which _VMC = 0 = background technique). .. As shown in the figure, in the background technology, the high frequency link power factor λ is a value far from the “optimal relationship”, but δ, dm (= 0), _{d z} that satisfy the equation (9) are used. If so, a high frequency link power factor λ similar to the “optimal relationship” can be obtained. From this result, it is understood that it is preferable to use the equation (9) from the viewpoint of power factor.

Next, regarding the specific configuration of the control device 4 for specifically determining each value of δ, _dm , and _dz based on the equation (9) and controlling the power conversion device 1 by the determined value. explain.

FIG. 16 is a schematic block diagram showing a functional block of the control device 4. As shown in the figure, the control device 4 is functionally configured to include a calculation unit 40, a duty cycle control unit 41, a phase shift control unit 42, and voltage detection units 43 and 44.

The voltage detection unit 43 is a functional unit that measures the voltages e _u , _ev , and _ew output from the system power supply 2 and supplies them to the arithmetic unit 40. Further, the voltage detection unit 44 is a functional unit that measures the voltage V _dc and supplies it to the calculation unit 40.

The calculation unit 40 supplies the power command value P ^* and the power factor angle command value α ^* supplied from the outside, the voltages e _u , _ev , e _w supplied from the voltage detection unit 43, and the voltage detection unit 44. A functional unit that calculates each value of δ, _dm , and _dz based on the voltage V _dc to be applied, and supplies _dm and _dz to the duty cycle control unit 41 and δ to the phase shift control unit 42, respectively. be.

The duty cycle control unit 41 is a functional unit that controls the matrix converter 10 according to _dm and _dz supplied from the calculation unit 40. Specifically, control signals for each of a plurality of switch elements G _pu , G _up , G _pv , G _vp , G _{p w} , G _wp , G _nu , _{Gun, G nv ,} G _vn , G _nw , and G _{w n} are generated. And is configured to supply. The phase shift control unit 42 is a functional unit that controls the AC / DC converter 30 according to δ supplied from the calculation unit 40. Specifically, it is configured to generate and supply control signals for _each of the plurality of switch elements G1 to _G4 .

In principle, the calculation unit 40 calculates δ, dm, _{d z} by combining the following equations (22), (23) and equation (9) into δ, dm, _{d z} ternary simultaneous equations. It can be executed by considering it as and solving it. However, P on the left side of the equation (22) is represented by the equation (6), and | _iV | on the left side of the equation (23) is represented by the equation (7). Further, | _iV ^* | on the right side of the equation (23) is a value obtained by substituting the power command value P ^* and the power factor angle command value α ^* into the equation (5).

However, since the equation (22), the equation (23), and the equation (9) include many variables and case classifications, the control cannot be made in time if the calculation is performed normally. Therefore, the arithmetic unit 40 according to the present embodiment is configured to calculate δ, _dm , and _dz by using the dichotomy method. Hereinafter, the details of the processing will be described in detail with reference to the processing flow chart.

FIG. 17 is a processing flow diagram showing the calculation processing of δ, _dm , and _dz performed by the calculation unit 40. In this process, the arithmetic unit 40 first performs the arithmetic of the equation (24), so that the current I _dc (δ ₁ , d _m1 ) when δ = δ ₁ , dm = _{d m1} and d _z = d _z1 , D _z1 ) (hereinafter referred to as current I _dc-1 ) (step S1). The final equation of the equation (24) is an equation obtained by substituting δ = δ ₁ , dm = _{d m1} , and d _z = d _z 1 into the equation (6) in the case of δ = d _z / 2. Is. Since δ ₁ , d _m1 , and d _z1 are numerical values that can be sequentially obtained by the equations (14), (12), and (13), the current I _dc-1 is also one numerical value.

Next, the calculation unit 40 compares the command value I _dc ^* (= P ^* / V _dc ) of the current I _dc with the current I _dc-1 obtained in step S1 (step S2). As a result, if the command value I _dc ^* is less than the current I _dc-1 , X, Y, Z in the equation (9) are set to the values shown in the equation (10) (step S3a), and further, δ. Substitute 0 for the section lower limit δ _L of δ and substitute δ ₁ for the section upper limit δ _H of δ (step S4a). On the other hand, if the command value I _dc ^* is the current I _dc-1 or more, X, Y, Z in the equation (9) are set to the values shown in the equation (11) (step S3b), and further, in δ. Substituting δ ₁ into the section lower limit δ _L and substituting 0.5 into the section upper limit δ _H of δ (step S4b).

Next, the arithmetic unit 40 sets the loop variable i to 0 (step S5), and then calculates the intermediate point δ _M between the interval upper limit δ _H and the interval lower limit δ _L by the following equation (25) (step S6).

Subsequently, the calculation unit 40 calculates d _z (hereinafter referred to as d _zM ) when δ = δ _M (step S7). Specifically, this calculation is performed by substituting δ = δ _M and dm = _{d m1} into Eq. (9). At this time, as the values of X, Y, and Z, the values set in steps S3a and S3b are used.

Next, the arithmetic unit 40 calculates the current I _dc (δ _M , d _m1 , d _zM ) _when δ = δ _M , d _m = d _{m 1} , and d _z = d z M (step S8). This calculation is performed by calculating the transmission power P by the equation (6) and further dividing the calculated transmission power P by V _dc . In that case, the case classification is performed based on δ _M and d _z M. Then, the calculated I _dc (δ _M , _{dm1, d zM )} is compared with the command value I _dc ^* of the current I _dc (step S9).

When it is determined in step S9 that the current I _dc (δ ₁ , d _m1 , d _zM ) is equal to the command value I _dc ^* (I _dc (δ _M , d _m1 , d _zM ) = I _dc ^* ), the calculation unit 40 Outputs δ _M , d _m1 , and d _zM at that time, and ends the process.

On the other hand, when it is determined in step S9 that the current I _dc (δ ₁ , d _m1 , d _zM ) is smaller than the command value I _dc ^* (I _dc (δ _M , d _m1 , d _zM ) <I _dc ^* ), the calculation is performed. Part 40 substitutes δ _M for the lower limit δ _L of the section (step S10). If it is determined in step S9 that the current I _dc (δ ₁ , d _m1 , d _zM ) is larger than the command value I _dc ^* (I _dc (δ _M , d _m1 , d _zM )> I _dc ^* ), the calculation is performed. Part 40 substitutes δ _M for the section upper limit δ _H (step S11).

After the end of step S10 or step S11, the arithmetic unit 40 adds 1 to the loop variable i (step S12), and then determines whether or not the loop variable i is below a predetermined value (for example, 5) (step S13). ). If it is determined that the value is lower than this, the process returns to step S6 and the process is repeated. On the other hand, if it is determined that the value is exceeded, δ _M , _dm1 and _dzM at that time are output, and the process is terminated.

Of the δ _M , _{dm1, and d zM output} as described above, the arithmetic unit 40 supplies _dm1 and d _zM to the duty cycle control unit 41, and supplies δ _M to the phase shift control unit 42. In this way, it is possible to transmit the power specified by the power command value P ^* at the power factor angle specified by the power factor angle command value α ^* , and to avoid an increase in switching loss while suppressing a decrease in the power factor. Will be done.

Hereinafter, the effects produced by the present embodiment will be described with reference to specific examples.

19 and 20 are the same conditions as those in FIGS. 3 and 4 except that _{dz = 0, and the voltages e u, ev, e w and the current} i _eu , i _ev , i _ew , i at the time of power running. It is a signal waveform diagram which shows the result of simulating the waveform of _L , i _u .

Looking at the reactor currents i _L of each of FIGS. 4 and 20, at least the current i _L3 does not satisfy the equation (18) in the example of FIG. 20, and ZVS is not realized, whereas in the example of FIG. All of the currents i _L0 to i _L3 satisfy the equation (18), and ZVS is realized. As described above, according to the present embodiment, ZVS is realized, and as a result, the avoidance of an increase in switching loss is realized. This effect is achieved by determining the values of δ, dm, and _{d z} so that the conditions of ZVS are satisfied.

Further, the amplitude of the reactor current IL in FIG. 4 is about half that of the _amplitude of the reactor current _IL in FIG. 20. This indicates that the above-mentioned circulating current is reduced in the present embodiment. The reduction of circulating current that does not contribute to power transmission results in an increase in power factor.

18 (a) is a diagram showing the power factor characteristics (examples) for V _dc and I _dc obtained by the simulations shown in FIGS. 3 and 4, and FIG. 18 (b) is a diagram showing FIGS. 19 and 4. It is a figure which shows the power factor characteristic (comparative example) with respect to V _dc , I _dc obtained by the simulation shown in 20. Looking at these figures, in the comparative example, the power factor decreases significantly with the decrease in V _dc , whereas in the example, the power factor does not depend much on V _dc , and a high power factor is achieved in a wide operating range. It is understood that there is. Therefore, according to the present embodiment, it can be said that it is possible to suppress a decrease in the power factor when operating in a wide voltage range. In other words, it can be said that this embodiment is effective for high-efficiency operation in a wide voltage range.

As described above, according to the present embodiment, it is possible to suppress an increase in the circulating current when the output voltage _Vdc on the AC / DC converter 30 side decreases, so that the MC type DAB type bidirectional isolated AC / DC can be suppressed. When the power conversion device 1 which is a converter is operated in a wide voltage range, it becomes possible to suppress a decrease in the power factor. Further, since the operation in ZVS becomes possible, when the power conversion device 1 is operated in a wide voltage range, it becomes possible to suppress an increase in switching loss in addition to suppressing a decrease in power factor.

Although the preferred embodiments of the present invention have been described above, the present invention is not limited to these embodiments, and the present invention can be implemented in various embodiments without departing from the gist thereof. Of course.

For example, in the above embodiment, the space IV ( _{eu>ev> e w ) shown} in FIG. 3 has been focused on, but the present invention can be applied to any of the spaces I to XII shown in FIG. Applicable. In this case, depending on the space to be applied, the mode in which _VMC = 0 is the mode in which the switch elements _{Sup and Sun are turned on at the same time, the mode in which the switch elements Svp and Svn are turned on at the same time, and the mode in which the switch elements Swp and S are turned} on at the same time. Of course, it is one of the modes in which _wn is turned on at the same time.

1 Power converter 2 System power supply 3 Load 4 Control device 10 Three-phase single-phase matrix converter 20 Transformer 20a, 20b Coil 30 AC / DC converter 35 Protection circuit 40 Calculation unit 41 Duty cycle control unit 42 Phase shift control unit 43, 44 Voltage Detection unit C1 Capacitor Cf Input capacitor G ₁ to G ₄ One-way switch element G _pu , G _up , G _pv , G _bp , G _pp , G _wp One-way switch element G _nu , _{Gun, G nv ,} G _vn , G _nw , G _{n One} -way switch element _Sup , S _bp , _Swp , _Sun , S _vn , _Swn Bidirectional switch element L Reactor Lf AC reactor Rf Damping resistance

Claims (3)

A transformer with first and second coils that are magnetically coupled to each other,
A first one-way switch element, one end connected to a fifth node constituting one end of a DC power supply, the other end connected to a third node constituting one end of the second coil, and one end connected to the DC. A second one-way switch element connected to a sixth node constituting the other end of the power supply, the other end connected to the third node, one end connected to the fifth node, and the other end connected to the fifth node. A third one-way switch element connected to a fourth node constituting the other end of the second coil, one end connected to the sixth node, and the other end connected to the fourth node. An AC / DC converter with a fourth one-way switch element,
A first bidirectional switch element, one end connected to a seventh node corresponding to the first phase of three-phase AC, and the other end connected to a first node constituting one end of the first coil. Is connected to the eighth node corresponding to the second phase of the three-phase AC, the other end is a second bidirectional switch element connected to the first node, and one end is the third phase of the three-phase AC. A third bidirectional switch element connected to a ninth node corresponding to the above, the other end of which is connected to the first node, one end of which is connected to the seventh node, and the other end of which is the first coil. A fourth bidirectional switch element connected to a second node constituting the other end, a fifth bidirectional switch having one end connected to the eighth node and the other end connected to the second node. A switch element and a matrix converter having a sixth bidirectional switch element having one end connected to the ninth node and the other end connected to the second node.
A reactor inserted between the first node and the first coil,
It is a control device of a power conversion device having
The mode in which the first and fourth bidirectional switch elements are turned on at the same time, the mode in which the second and fifth bidirectional switch elements are turned on at the same time, and the third and sixth bidirectional switch elements are The matrix converter is periodically controlled in the first cycle so that any one of the modes that are turned on at the same time is executed at least once in each cycle.
When the phase voltage of the first phase is larger than the phase voltage of the second phase and the phase voltage of the second phase is larger than the phase voltage of the third phase, the third and sixth phases are performed in each cycle. The first mode in which the bidirectional switch elements are turned on at the same time, the second mode in which the first and sixth bidirectional switch elements are turned on at the same time, and the second and sixth bidirectional switch elements are simultaneously turned on. A third mode in which the third and sixth bidirectional switch elements are turned on at the same time, a fourth mode in which the third and sixth bidirectional switch elements are turned on at the same time, and a fifth mode in which the third and fourth bidirectional switch elements are turned on at the same time. The mode, the sixth mode in which the third and fifth bidirectional switch elements are turned on at the same time, and the seventh mode in which the third and sixth bidirectional switch elements are turned on at the same time are in this order. Configured to run,
The AC / DC converter is used so that the first and fourth one-way switch elements are turned on in the first half of each cycle and the second and third one-way switch elements are turned on in the second half of each cycle. It is controlled periodically in 1 cycle,
Current i flowing through the reactor_LThe value at the start timing of the second mode of_L0, The current i_LThe value at the timing when the first and fourth one-way switch elements of are turned on is I._L3, The duration of the first mode is d_z/ 2, where the phase difference between the control cycle of the matrix converter and the control cycle of the AC / DC converter is δ, the d_zIs δ <d_zIf it is / 2, -I_L0= I_L3The above d obtained by_zIs determined according to the first relationship between,
SystemYour device. When d _z is δ> d _z / 2, the intersection of the first relationship and d _z = 2 δ (δ ₁ , d _z 1) and the point where the current i _L is maximized (δ ₂ ). , 0) and is determined according to the second relationship between the _dz and the δ.
The control device according to claim 1 . The _dz and the δ are determined by the dichotomy method based on the first and second relationships.
The control device according to claim 2 .

Images