JP2005269768A - Three-phase inverter - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To ensure high performance while reducing the cost by eliminating the need for judging the relation between a DC input current and the load current of each phase, and deriving the component of a load current vector required for inverter control through simple processing thereby shortening the processing time. <P>SOLUTION: The controller of a three-phase inverter comprises means 26, 29 and 27 for sample holding a DC input current flowing between a DC voltage part and a load through an arm part, and a current conversion means 30 for determining the component of a load current vector using two DC input current values sample held in two kinds of mode and the phase angle information of a voltage command vector in the two kinds of mode being specified every predetermined phase period of the voltage command vector out of a plurality of modes where a load terminal takes a state substantially conducting with the positive pole or the negative pole of the DC voltage part only at one of three arm parts. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a three-phase inverter device that indirectly detects a load current vector component of an inverter using a DC input current detection value of the three-phase inverter and uses the load current vector component for controlling the inverter.

FIG. 5 is a block diagram showing the first prior art, showing the basic configuration of a three-phase inverter and its control device.
In FIG. 5, reference numeral 10 denotes a three-phase inverter provided with semiconductor switching elements 11 to 16, and as is well known, a u-phase arm portion including switching elements 11 and 14 connected in series and a v-phase including 12 and 15. An arm part and a w-phase arm part composed of 13 and 16 are connected in parallel to a DC voltage part (not shown), and a connection part between the switching elements 11 and 14 and a connection part between the 12 and 15. The load is connected to the output terminal which is the connection part between the 13 and 16.

Among the three-phase outputs of the inverter 10, load currents for two phases (for example, u phase and w phase) are detected by current detection means 21, 22, and these detected currents i _{u det} , i _{w det} are converted into coordinate conversion means 23. Has been entered.
The coordinate conversion means 23 uses the phase angle θ synchronized with the phase of the voltage command vector to convert the three-phase load current into a component on the biaxial rotation coordinate, and derives the effective current amplitude i _δ and the reactive current amplitude i _γ . To do. The coordinate conversion means 23 derives another one-phase (v-phase) current that is not detected by the current detection means 21 and 22 by utilizing the fact that the sum of the three-phase currents becomes zero.

Here, when the active current amplitude i _δ and the reactive current amplitude i _γ are obtained, the phase angle θ is nothing but the phase of the voltage command vector. For example, when using a synchronous motor as a load, the magnetic pole position of the rotor coordinate transformation in accordance with, i.e. q-axis component i _q is the component of the no-load induced voltage in _phase, and, it is necessary to disassemble the d-axis component i _d perpendicular thereto. In that case, a phase angle representing the magnetic pole position may be used as θ.

The voltage command calculation unit 24 determines the amplitude V and the phase angle θ of the voltage command vector based on the derived biaxial currents i _δ and i _γ . For this purpose, for example, adjusting means for calculating the deviation V between the current command value and the biaxial currents i _δ and i _γ and outputting the amplitude V and the phase angle θ of the voltage command so as to minimize it, typically A proportional integral controller is used.
The command pulse generating means 25 generates on / off signals to be given to the switching elements 11 to 16 of the three-phase inverter 10 based on the amplitude V and the phase angle θ of the voltage command.
In the three-phase inverter 10, the switching elements 11 to 16 are repeatedly turned on and off according to a predetermined rule, so that a voltage proportional to the voltage command is output and applied to the load.

Next, FIG. 6 is a block diagram showing the second prior art, in which the load current detecting means 21 and 22 of FIG. 5 are removed for the purpose of cost reduction, and instead the DC input current of the inverter 10 is detected. The effective current amplitude i _δ and the reactive current amplitude i _γ are derived.
The DC input current i _dc of the inverter 10 always matches the load current of any one of the three phases depending on the on / off states of the switching elements 11 to 16. In addition, when only the switching elements 11 to 13 of all the upper arms of the inverter 10 are turned on or only the switching elements 14 to 16 of all the lower arms are turned on, the load current circulates through the switching elements in the on state. _dc is zero.

Therefore, if i _dc is sampled and held in the state of a certain switching element and taken into the control system, it can be determined from the state of the switching element which phase the load current corresponds to i _dc. Is performed in a very short period (a period that can be regarded as almost simultaneous) for two phases of the three phases, a two-phase load current at substantially the same time can be obtained.

The prior art of FIG. 6 is based on the above principle, and the DC input current i _dc detected by the DC input current detecting means 26 is sampled and held by a sample and hold circuit 27 for two phases in a very short time to convert the current. The data is taken into the means 28 and input to the coordinate conversion means 23 as detected current equivalent values i _u ′ and i _w ′ for two phases. The subsequent operation is the same as that of the prior art of FIG.
In FIG. 6, reference numeral 29 denotes a sample / hold signal generating means for generating the sample / hold signal a based on the output of the voltage command calculating means 24.
According to this prior art, the load current of the inverter 10 can be detected by the single inexpensive DC input current detection means 26 and the inverter 10 can be controlled, so that the cost of the apparatus can be reduced.

In Patent Document 1 described later,
a. Similar to FIG. 6, the principle of deriving the load current from the DC input current b. The voltage vector defined by the three-phase voltage command value is defined, and the instantaneous voltage vector (that is, the switching mode) output at every 60 ° of the phase angle is determined in two types. Therefore, the load current that matches the DC input current is also 2 Determined by the type, this being a detectable current if the voltage vector depends on the 60 ° phase region c. It is disclosed to determine which phase load current corresponds to the detected DC input current, and to obtain the load current for three phases from the two-phase load current.
Patent Document 2 discloses a technique substantially similar to the above-mentioned a, and Patent Document 3 discloses a technique substantially similar to the above-described a and b.

Japanese Patent No. 2563226 (Claim 1, [0014] to [0016], [0019] to [0024], FIGS. 1 to 3, FIG. 5, etc.) JP-A-8-19263 (Claim 1, [0021], [0036] to [0038], FIGS. 2 to 4 etc.) JP 2001-314090 A ([0016] to [0019], FIG. 2, FIG. 3, FIG. 5 etc.)

  As apparent from a comparison between FIG. 5 and FIG. 6, the current detection means is simplified in the configuration of FIG. 6, but it is necessary to determine which phase load current the DC input current corresponds to. is there. When this processing is realized by software using a microprocessor, the processing time increases, and the inherent performance of the control device such as voltage command calculation may be sacrificed. Further, when the determination process is realized by hardware, inconveniences such as an increase in cost and an increase in hardware mounting area are expected. These problems are common to the prior arts of Patent Documents 1 to 3.

  Note that claim 10 of Patent Document 3 and the corresponding embodiments ([0027], [0028]) describe that the effective current and the reactive current are derived directly from the DC input current. However, since the calculation formula for obtaining both currents is actually changed for each 60 ° region of the voltage command vector, the load current is substantially changed for each region of the voltage vector as in the prior art of FIG. As in the case of determining the two phases capable of detecting the above, the complexity of the arithmetic processing and the processing time are not so improved.

  Therefore, the problem to be solved by the present invention is that it is not necessary to determine the relationship between the DC input current and the load current and the attribute region of the voltage command vector, and the load current vector component necessary for inverter control can be derived by simple processing. The object is to provide a three-phase inverter device.

In order to solve the above-mentioned problem, the invention described in claim 1 is characterized in that three arm portions formed by connecting at least two semiconductor switching elements in series are connected in parallel to the DC voltage portion, and these three arm portions are In a three-phase inverter device comprising a three-phase inverter in which a load is connected to a connection portion between semiconductor switching elements in and a control device for controlling the inverter,
The control device includes:
Three types of modes (switching mode) in which the terminal of the load is in a substantially conductive state with the positive electrode of the DC voltage unit in only one arm unit among the three arm units, or only one arm unit In the three types of modes specified for each predetermined phase period of the voltage command vector of the three-phase inverter, among the three types of modes in which the terminal of the load is substantially conductive with the negative electrode of the DC voltage unit Means for sampling and holding a DC input current flowing between the DC voltage unit and the load via the arm unit;
For each region obtained by dividing the phase angle that the voltage command vector can take by the predetermined phase period, two DC input current values sampled and held in the two types of modes by the means, and the phase angle of the voltage command vector Means for determining the component of the load current vector using the information,
It is equipped with.
In other words, the two types of modes are none other than switching modes for outputting two voltage vectors that determine the region where the voltage command vector exists.

The invention described in claim 2 is the invention according to claim 1,
The predetermined phase period is set to 60 °, and the phase angle region that the voltage command vector can take is divided into six.

The invention described in claim 3 is, in claim 2,
The phase angle information of the voltage command vector is a value between −60 ° and 60 ° centered on a three-phase reference axis distributed at 120 ° intervals.

The invention described in claim 4 is, in claim 3,
The component of the load current vector is obtained only by the four arithmetic operations using the DC input current value, the voltage command vector phase angle information and the coefficient in the two types of modes, and the polarity determination of the phase angle of the voltage command vector.

The invention described in claim 5 is any one of claims 1 to 4,
The biaxial component (effective current component, reactive current component) on the rotation coordinate of the load current vector is obtained.

  According to the present invention, there is no need to determine the relationship between the DC input current and each phase load current and the voltage vector attribution area, which is unavoidable in conventional inverter control based on DC input current detection, and the inverter can be easily processed. It is possible to derive a component (biaxial component or uniaxial component) on the rotational coordinate of the load current vector necessary for controlling the current. As a result, it is possible to provide an inverter device that reduces the burden on software and hardware, shortens the processing time, and satisfies both low cost and high performance.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 shows a configuration of a three-phase inverter device according to the embodiment, and the same reference numerals are given to the same components as those in FIG. In this embodiment, it is possible to derive a component on the rotational coordinate of the load current vector (hereinafter also simply referred to as a biaxial current) without performing a process of determining which phase of the load current the DC input current corresponds to. Is.

In FIG. 1, the difference from FIG. 6 will be mainly described. The sample and hold signal generation means 29 uses the information on the amplitude V and the phase angle θ of the voltage vector output from the voltage command calculation means 24, and the inverter 10 The sample and hold means 27 generates the sample and hold signal a so that the sample and hold means 27 takes in the DC input current i _dc in two types of modes that output a predetermined voltage vector.

In addition to the configuration shown in FIG. 1, the sample-and-hold signal a can be used to determine which voltage vector is selected by the command pulse output from the command pulse generation means 25 to the inverter 10, for example. Alternatively, the timing at which the command pulse is output may be determined in advance, and the sample and hold may be performed at a timing at which the predetermined command pulse is generated and the DC input current i _dc can be detected. Thus, it may be output by timer processing.
In addition, the sample and hold signal a is obtained from the sample and hold means 27 in such a short time that the DC input current i _dc for two phases that can be detected can be regarded as almost the same time for each 60 ° region where the voltage command vector exists. Is output so that it can be captured.

The current conversion unit 30 does not determine which phase of the obtained two DC input current detection values i _{dc det} corresponds to the two-axis component (effective current) on the rotation coordinate of the load current vector. Amplitude i _δ and reactive current amplitude i _γ ) are derived directly. As a result, the calculation processing amount of the control device is reduced, so that the original performance of the microprocessor or the like is not impaired.
The above contents will be described in detail below.

2 shows the voltage command vector v of the three-phase inverter 10 and six types of voltage vectors (U, U ′, V, V ′, W, W ′) that can be output by the inverter 10 along the voltage reference axis. , Phase current vector i, three-phase current vectors i _u , i _v , i _w constituting the current vector i, phase angle θ of the voltage command vector v, and a region divided by the six types of voltage vectors <I> to <VI> are shown.
In addition, (0, 1, 0), (1, 0, 0), etc. written together with the symbol of the voltage vector indicate the on / off states of the (u-phase, v-phase, w-phase) switching elements, “1” indicates that the upper arm switching element is ON, and “0” indicates that the lower arm switching element is ON. For example, (0, 1, 0) indicates that the u-phase lower arm switching element 14 in the inverter 10 of FIG. 1 is on, the v-phase upper arm switching element 12 is on, and the w-phase lower arm switching element 16 is on. State. (0, 0, 0) and (1, 1, 1) all indicate zero voltage vectors.
Also, below the vector diagram, there is a change in each phase voltage command according to the phase angle θ (0 to 360 °) that the voltage command vector can take, and the voltage command vector v exists at the phase angle θ 6 The types of areas <I> to <VI> (corresponding to the areas of the vector diagram) are shown as waveforms.

First, the current vector i is expressed as Equation 1 using a complex vector.

  The relationship of Formula 1 is generally established in three-phase alternating current. Further, it is assumed that the relationship shown in Formula 2 is established for the three-phase current.

Now, when the voltage command vector v is in the region <I> in FIG. 2, the inverter 10 outputs three types of voltage vectors U (1, 0, 0), W ′ (1, 1, 0), and zero voltage vector. Since the output is divided in terms of time, the load currents that can be detected by the DC input current i _{dc in} FIG. 1 are u u i _u and w phase −i _w . Similarly, when the voltage command vector v is in the region <II>, the v-phase _iv and the w-phase -i _w can be detected.
Here, among the load current of the two phases can be detected, the current of the phase polarity is positive and i _s1, detects the phase of the current is negative as i _s2, further, the phase current polarity is positive Assume that the current vector is defined according to Equation 2 as the vector reference phase. Table 1 _summarizes the relationship between the voltage command vector v existing region, i _s1 , i _s2 , and the current vector reference phase at this time.

  For each region of <I> to <VI>, formulas 3 and 8 are obtained by rearranging the relations of formulas 1 and 2 using the phase information that can detect the load current shown in Table 1.

As described above, it can be seen that the real part X and the imaginary part jY of the current vector calculation formula are common in each region, and only the sign of the imaginary part jY is alternately replaced.
Here, the phase angle θ of the voltage command vector v is also defined with reference to the current vector reference phase shown in Table 1 for each region where the voltage command vector v exists, and the phase angle is newly set to θ _v. For example, the relationship of Equation 9 and Equation 10 is established.

From the above discussion, the effective component i _δ and the invalid component i _γ of the current vector i based on the DC input current detection value i _{dc det} are derived from the DC input current value and voltage at two points of time by the calculation block shown in FIG. It can be seen that this can be realized only by the four arithmetic operations using the phase angle information and coefficient of the command vector v and the polarity determination of the phase angle.
In FIG. 3, 31 is a current detection value attribute determining means, 32, 34 and 36 are coefficient multiplying means, 33 is a 60 ° mask processing means, 35 is an adding means, 37 is a sign determining function, 38 is a multiplying means, and 39 is a phase. It is an angular function matrix.
The configuration of FIG. 3, the X is output from the coefficient multiplying means 34, Y is output from the coefficient multiplying means 36, the Y is converted into Y _s in multiplication means 38 by the equation 9. By calculating these X, Y _s and the matrix 39, the effective component (effective current amplitude) i _δ and the reactive component (reactive current amplitude) i _γ of the current vector i are obtained.

The current detection value attribute determination unit 31, _i u as a positive polarity of the current _{i s1,} as shown in Table _1, i v, and can output either _{i w,} as a negative polarity of the current _{i s1} - Any of i _u , −i _v , and −i _w can be output.
Further, the 60 ° mask processing means 33 assumes a case where conversion from θ to θ _v is realized by software processing, so that θ expressed in a signed binary number is carried at θ = 60 °, By performing mask processing for the digits higher than the sign bit, it can be realized without any additional processing. In this connection, theta _v code determination function sign (θ _v) is, theta _v since can be realized only by a reference sign bit when expressed in a signed binary number, the software processing time is very small, The effect on the whole can be ignored.

As described above, according to the present embodiment, it is not necessary to determine which phase the load current of the DC input current detection value i _{dc deet} corresponds to as in the conventional technique, and the dynamics of the biaxial current calculation method are not required. The biaxial components i _δ and i _γ on the rotational coordinates of the load current vector can be derived by sequential processing that does not require any switching.
Since the control operation of the three-phase inverter 10 using the current components i _δ and i _γ is the same as that of the prior art, the description thereof is omitted here.

FIG. 4 shows X, Y, i _δ , i _γ , θ _v when the voltage command vector v exists in the region <II> and the current vector i has a delayed phase (the same case as in FIG. 2). This shows the relationship. Here, the current vector reference phase for a V-phase, a phase angle theta _v relative to the phase angle theta Again V-phase voltage command vector v according to Table 1.

As described above, the gist of the present invention is that the relationship between the phase in which the load current can be detected and the current vector to be derived is the same in each region divided by the phase angle of 60 ° of the voltage command vector. For this reason, the expression of each numerical formula and the difference in constants are included in the scope of the present invention.
Further, circulation of the regions <I> to <VI> is also included in the category of the present application. That is, in the example of FIGS. 3 and 4, the regions <I> to <VI> are determined based on the u phase, but the regions <I> to <VI> may be determined based on the v phase. )
Furthermore, since the current vector is obtained as a biaxial component, when only one of the uniaxial components (eg, effective current) is required, the calculation processing can be further simplified by eliminating the calculation of unnecessary components. it can.

It is a block diagram which shows embodiment of this invention. It is a voltage and current vector diagram of a three-phase inverter. It is a block diagram of the current conversion means in FIG. It is a voltage and current vector diagram of a three-phase inverter. It is a block diagram which shows 1st prior art. It is a block diagram which shows a 2nd prior art.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10: Three-phase inverter 11-16: Semiconductor switching element 24: Voltage command calculating means 25: Command pulse generation means 26: DC input current detection means 27: Sample hold means 29: Sample hold signal generation means 30: Current conversion means 31: Current detection value attribute determination means 32, 34, 36: Coefficient multiplication means 33: 60 ° mask processing means 35: Addition means 37: Sign determination function 38: Multiplication means 39: Phase angle function matrix

Claims (5)

Three-phase in which at least two semiconductor switching elements are connected in series with three arm parts connected in parallel to the DC voltage part, and a load is connected to the connection part between the semiconductor switching elements in these three arm parts In a three-phase inverter device comprising an inverter and a control device for controlling the inverter,
The control device includes:
Three types of modes in which the terminal of the load is substantially in conduction with the positive electrode of the DC voltage unit in only one arm unit among the three arm units, or the load terminal in only one arm unit. Of the three types of modes in which the terminal is substantially in a conductive state with the negative electrode of the DC voltage unit, in the two types of modes specified for each predetermined phase period of the voltage command vector of the three-phase inverter, the arm unit Means for sampling and holding a DC input current flowing between the DC voltage section and the load via
For each region obtained by dividing the phase angle that the voltage command vector can take by the predetermined phase period, two DC input current values sampled and held in the two types of modes by the means, and the phase angle of the voltage command vector Means for determining the component of the load current vector using the information,
A three-phase inverter device comprising: In the three-phase inverter device according to claim 1,
A three-phase inverter device, wherein the predetermined phase period is set to 60 °, and a region of a phase angle that can be taken by the voltage command vector is divided into six. In the three-phase inverter device according to claim 2,
The three-phase inverter device, wherein the phase angle information of the voltage command vector is a value of -60 ° to 60 ° centered on a three-phase reference axis distributed at 120 ° intervals. In the three-phase inverter device according to claim 3,
The load current vector component is obtained only by four arithmetic operations using the DC input current value, the voltage command vector phase angle information and the coefficient in the two types of modes, and the polarity determination of the phase angle of the voltage command vector. Three-phase inverter device. In the three-phase inverter device according to any one of claims 1 to 4,
A three-phase inverter device characterized by obtaining a biaxial component on a rotation coordinate of a load current vector.

Images