KR101062386B1 - Power Factor Correction Method of Matrix Converter and Its System - Google Patents

Abstract

A method of improving the power factor of a matrix converter is disclosed. The method for improving power factor of a matrix converter connected to an input filter may include calculating an actual phase difference between an input voltage and an input current input to the matrix converter by measuring an input voltage and an input current input to the input filter, based on the calculated actual phase difference. Determining whether compensation of the input power factor is necessary; calculating a target phase difference between the input voltage and the input current input to the matrix converter; and determining a space vector modulation using the calculated target phase difference. Performing a step. Accordingly, it is possible to improve the power factor reduction of the matrix converter by using the input filter.

SVM, Power Factor, Input Filter

Description

Power factor improving method of matrix converter and system

The present invention relates to a method and system for improving power factor of a matrix converter, and more particularly, to a method and system for improving power factor of a matrix converter using an input filter.

With the development of power devices, efforts have been made to improve the efficiency and performance of power converters. Among the power converters, the matrix converter does not convert commercial AC power to direct current, unlike conventional inverters, to supply variable frequency and voltage using a constant AC power source. It is attracting attention as a next-generation power converter that can be replaced.

1 is a circuit diagram showing the configuration of a conventional matrix converter.

Since the matrix converter 130 converts directly from the commercial AC power source 110 to variable AC, a rectifier and a smoothing capacitor are not required, which significantly reduces the volume and weight of the system, and also facilitates operation at high temperatures and conversion efficiency. It has the advantage of extending the life as well as improving. In addition, the matrix converter 130 can maintain the input / output current as a sine wave without any additional device, enable bidirectional power control, and set the input power factor to 1. In addition, when controlling the electric motor there is an advantage that the current is not concentrated in a specific power device, even if the operation for a long time at a low speed state does not give a burden to the device.

As shown in FIG. 1, the matrix converter 130 is composed of nine bidirectional power semiconductors having a self-extinguishing function. The matrix converter 130 eliminates the influence of harmonic components generated by high-speed switching of the power semiconductors on the main power stage. In order to install the input filter 110. The LC filter is mainly used for this purpose and it is very important to design the filter effectively to remove the desired harmonic components from the input current and to suppress the distortion of the input voltage waveform as much as possible.

However, despite the optimal design of the input filter 120, due to the inductance or capacitance components included in the filter, the conventional power vector power factor for controlling the phase of the voltage and current input to the matrix converter 130 is the same as the power factor at the power source. It is difficult to keep it at 1. In addition, when the output frequency or load is changed, the input power factor is also changed at the same time, and when the operating range of the matrix converter is widened, it is difficult to maintain a high power factor in all operation sections.

An object of the present invention is to provide a method for improving the power factor of a matrix converter using an input filter and a system according to the above problem.

In order to achieve the above object, a method of improving power factor of a matrix converter connected to an input filter according to an embodiment of the present invention may include: an input voltage input to the matrix converter by measuring an input voltage and an input current input to the input filter; And calculating the actual phase difference of the input current, determining whether compensation of the input power factor is necessary based on the calculated actual phase difference, and when it is determined that compensation of the input power factor is necessary, the input voltage and the input to the matrix converter Calculating a target phase difference of the current, and performing spatial vector modulation by using the calculated target phase difference.

The performing of the spatial vector modulation may include: dividing a state space for spatial vector modulation into a plurality of regions, and determining a reference output voltage vector, which is an adjacent vector component of an output voltage vector located in a predetermined section, among the plurality of sections. Detecting, selecting switching vectors capable of generating the reference output voltage vector, and calculating switching application time of each switching vector by using the switching vectors and the calculated target phase difference. have.

In this case, the reference output voltage vector includes a first reference output voltage vector and a second reference output voltage vector, which are adjacent vector components of the output voltage vector, and the switching vectors capable of generating the first reference output voltage vector include: The switching vectors that are the first and second switching vectors and capable of generating the second reference output voltage vector may be third and fourth switching vectors.

Meanwhile, when the input voltage vector and the input current vector inputted to the matrix converter are located in the same section, the switching application times d1 and d2 for the first and second switching vectors may be calculated by the following equations, respectively. .

Where α ₀ is the phase of the output voltage vector, α _i is the phase of the input voltage vector, δ is the phase difference between the input voltage vector and the input current vector, k _v is the section in which the output voltage vector is located, and k _i is the input current vector ( And an interval in which the input voltage vector) is located, q is a voltage conversion rate ( V ₀ / V _m) .

In addition, when the input voltage vector and the input current vector input to the matrix converter are located in the same section, the switching application times d3 and d4 for the third and fourth switching vectors may be calculated by the following equations, respectively.

Where α ₀ is the phase of the output voltage vector, α _i is the phase of the input voltage vector, δ is the phase difference between the input voltage vector and the input current vector, k _v is the section in which the output voltage vector is located, and k _i is the input current vector ( And an interval in which the input voltage vector) is located, q is a voltage conversion rate ( V ₀ / V _m) .

In addition, when the input voltage vector and the input current vector input to the matrix converter are located in the same section, the switching application time d5 to which the zero voltage vector is applied may be calculated by the following equation.

Where α ₀ is the phase of the output voltage vector, α _i is the phase of the input voltage vector, δ is the phase difference between the input voltage vector and the input current vector, k _v is the section in which the output voltage vector is located, and k _i is the input current vector ( And an interval in which the input voltage vector) is located, q is a voltage conversion rate ( V ₀ / V _m) .

Meanwhile, when the input voltage vector and the input current vector inputted to the matrix converter are located in different sections, the switching application times d1 and d2 for the first and second switching vectors may be calculated by the following equations, respectively.

Where α ₀ is the phase of the output voltage, α _i is the phase of the input voltage vector, δ is the phase difference between the input voltage vector and the input current vector, k _v is the section in which the output voltage vector is located, and k _i is the input current vector. Where q is the voltage conversion rate ( V ₀ / V _m) .

In addition, when the input voltage vector and the input current vector input to the matrix converter are located in different sections, the switching application times d3 and d4 for the third and fourth switching vectors may be calculated by the following equations, respectively. .

Where α ₀ is the phase of the output voltage vector, α _i is the phase of the input voltage vector, δ is the phase difference between the input voltage vector and the input current vector, k _v is the section where the output voltage vector is located, and k _i is the input current vector. Q is the voltage conversion rate ( V ₀ / V _m) .

In addition, when the input voltage vector and the input current vector input to the matrix converter are located in different sections, the switching application time d5 to which the zero voltage vector is applied may be calculated by the following equation.

Where α ₀ is the phase of the output voltage vector, α _i is the phase of the input voltage vector, δ is the phase difference between the input voltage vector and the input current vector, k _v is the section where the output voltage vector is located, and k _i is the input current vector. Q is the voltage conversion rate ( V ₀ / V _m) .

The matrix converter system including an input filter may include a detector configured to measure an input voltage and an input current input to the input filter and calculate an actual phase difference between the input voltage and the input current input to the matrix converter, and the calculated actual value. If it is determined that compensation of the input power factor is necessary based on the phase difference, and it is determined that compensation of the input power factor is necessary, an operation unit for calculating the target phase difference between the input voltage and the input current input to the matrix converter, and the calculated target phase difference And a spatial vector modulator for performing spatial vector modulation.

The spatial vector modulator may be configured to detect a reference output voltage vector, which is an adjacent vector component of an output voltage vector located in a predetermined section, among the plurality of sections divided for spatial vector modulation, and generate the reference output voltage vector. By selecting the vectors, the switching application time of each switching vector may be calculated using the selected switching vectors and the calculated target phase difference.

In this case, the reference output voltage vector includes a first reference output voltage vector and a second reference output voltage vector, which are adjacent vector components of the output voltage vector, and the switching vectors capable of generating the first reference output voltage vector include: The switching vectors that are the first and second switching vectors and capable of generating the second reference output voltage vector may be third and fourth switching vectors.

Meanwhile, when the input voltage vector and the input current vector input to the matrix converter are located in the same section, the switching application time d1 and d2 for the first and second switching vectors may be calculated by the following equations, respectively.

Where α ₀ is the phase of the output voltage vector, α _i is the phase of the input voltage vector, δ is the phase difference between the input voltage vector and the input current vector, k _v is the section where the output voltage vector is located, and k _i is the input current vector. (And the input voltage vector) is located, q is the voltage conversion rate ( V ₀ / V _m) .

In addition, when the input voltage vector and the input current vector input to the matrix converter is located in the same section,

The switching application times d3 and d4 for the third and fourth switching vectors may be calculated by the following equations, respectively.

Where α ₀ is the phase of the output voltage vector, α _i is the phase of the input voltage vector, δ is the phase difference between the input voltage vector and the input current vector, k _v is the section in which the output voltage vector is located, and k _i is the input current vector ( And an interval in which the input voltage vector) is located, q is a voltage conversion rate ( V ₀ / V _m) .

In addition, when the input voltage vector and the input current vector input to the matrix converter are located in the same section, the switching application time d5 to which the zero voltage vector is applied may be calculated by the following equation.

Where α ₀ is the phase of the output voltage vector, α _i is the phase of the input voltage vector, δ is the phase difference between the input voltage vector and the input current vector, k _v is the section in which the output voltage vector is located, and k _i is the input current vector ( And an interval in which the input voltage vector) is located, q is a voltage conversion rate ( V ₀ / V _m) .

Meanwhile, when the input voltage vector and the input current vector inputted to the matrix converter are located in different sections, the switching application times d1 and d2 for the first and second switching vectors may be calculated by the following equations, respectively.

Where α ₀ is the phase of the output voltage vector, α _i is the phase of the input voltage vector, δ is the phase difference between the input voltage vector and the input current vector, k _v is the section where the output voltage vector is located, and k _i is the input current vector. Q is the voltage conversion rate ( V ₀ / V _m) .

In addition, when the input voltage vector and the input current vector input to the matrix converter are located in the same section, the switching application times d3 and d4 for the third and fourth switching vectors may be calculated by the following equations, respectively.

Where α ₀ is the phase of the output voltage vector, α _i is the phase of the input voltage vector, δ is the phase difference between the input voltage vector and the input current vector, k _v is the section where the output voltage vector is located, and k _i is the input current vector. Q is the voltage conversion rate ( V ₀ / V _m) .

In addition, when the input voltage vector and the input current vector input to the matrix converter are located in the same section, the switching application time d5 to which the zero voltage vector is applied may be calculated by the following equation.

Where α ₀ is the phase of the output voltage vector, α _i is the phase of the input voltage vector, δ is the phase difference between the input voltage vector and the input current vector, k _v is the section where the output voltage vector is located, and k _i is the input current vector. Q is the voltage conversion rate ( V ₀ / V _m) .

Accordingly, when the input filter is used in the matrix converter, the input power factor can be maintained at almost 1 in almost all load regions even when the load or output frequency is changed.

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

2 is a block diagram illustrating a configuration of a matrix converter system according to an embodiment of the present invention. According to FIG. 2, the matrix converter system 200 includes a power supply unit 210, an input filter unit 220, a matrix converter unit 230, a detection unit 240, an operation unit 250, and a spatial vector modulator 260. do. Meanwhile, since the circuit structures of the power supply unit 210, the input filter unit 220, and the matrix converter unit 230 according to the present invention are the same as those shown in FIG. 1, with reference to the circuit diagram shown in FIG. Explain.

The power supply unit 210 is composed of an AC power supply, and supplies a variable frequency and voltage. Specifically, the power supply unit 210 may be configured as a three-phase AC power supply. Three-phase input currents from the three-phase alternating current source are respectively input from the three-phase alternating current source through the input filter unit 220 to the matrix converter unit 230.

The input filter unit 220 removes the influence of the harmonic component noise generated by the high-speed switching of the nine bidirectional power semiconductors provided in the matrix converter 230 on the main power supply terminal.

The input filter unit 220 may be designed to reduce the ripple of the input current while minimizing the reactive energy of the device. For example, the input filter unit 220 may be implemented as an LC low pass filter, such as the input filter 120 shown in FIG. 1, and in some cases, may be implemented as a multistage filter when a higher attenuation rate is required. .

The cutoff frequency of the input filter unit 220 should be lower than the switching frequency. Even when the output of the matrix converter unit 230 is small, the input power factor should be maintained at maximum, and the input filter may have a small voltage drop in the inductor when the rated current flows. Part 220 should be designed.

As shown in FIG. 1, the matrix converter 230 includes nine bidirectional switches, and each bidirectional switch connects an input and an output. The matrix converter 230 may supply a variable voltage and a variable frequency by controlling gates of nine bidirectional switches, that is, on / off control.

The detector 240 measures an input power voltage and an input power current input to the input filter 220 to calculate an actual phase difference between the input voltage and the input current input to the matrix converter 230. Here, the detector 240 may periodically measure the input power current input to the input filter 220.

The calculator 250 determines whether compensation of the input power factor is necessary based on the actual phase difference provided by the detector 240.

In addition, when it is determined that compensation of the input power factor is necessary, the calculator 250 calculates a target phase difference between the input voltage and the input current input to the matrix converter 230.

The space vector modulator 260 performs a space vector modulation for controlling the switching of the matrix converter 230 by using the target phase difference calculated by the calculator 250.

In detail, the space vector modulator 260 may divide the state space for spatial vector modulation into a plurality of regions, and detect a reference output voltage vector that is an adjacent vector component of the output voltage vector located in a predetermined section among the plurality of sections. have. In addition, by selecting the switching vectors capable of generating the detected reference output voltage vector, it is possible to calculate the switching application time of each switching vector using the selected switching vectors and the target phase difference calculated by the calculator 240.

Hereinafter, a detailed configuration of the matrix converter system 200 will be described based on the functions of the calculator 250 and the space vector modulator 260.

FIG. 3 is a circuit diagram illustrating a per phase equivalent circuit of the input filter illustrated in FIGS. 1 and 2.

Voltage and current relational expressions between the input and the output of the input filter unit 120 shown in FIG. 3 are represented by Equations 1 to 3 below.

If the steady state values of the power supply voltage and current are assumed to be sinusoidal waves having the same phase angle as shown in Equation 4 below, the steady state fundamental wave voltage values and current values applied to the matrix converter 220 in Equations 1 to 3 are given by Equation 1 below. Equations 5 and 6 are the same.

Where ω is the input power frequency. In addition, since the voltage drop due to the filter inductor L of the input filter unit 220 is generally negligible in Equation 5, Equation 7 below holds.

Therefore, in Equation 6, the phase difference φ between the input voltage V _i and the current I _i of the matrix converter 230 is equal to the following Equation 8, and the input voltage and the input current input to the matrix converter 230 are used. It is possible to calculate the target phase difference δ of.

Meanwhile, the matrix converter 230 includes nine bidirectional switches as shown in FIG. 1. For convenience of description, the voltage and current at the input terminal of the matrix converter 230 are displayed using lowercase letters a, b, and c, and the output terminal is displayed using uppercase letters A, B, and C.

There are 27 switching modes of the nine bidirectional switches of the matrix converter 230, and the output voltage vector and input current vector for each case are shown in Table 1 (switching mode of the matrix converter MC).

In Table 1, the type of switching of the matrix converter 230 can be divided into three groups (I to III). In the group I, two output terminals are connected to one input terminal, and the group II is one output terminal. Group III indicates that all the output terminals are connected to different input terminals.

On the other hand, in Group III, the voltage between each output line is directly connected to the voltage between each input line, the output line voltage varies according to the input line voltage, and the input line current is also affected by the output line current. Therefore, the spatial vector position α ₀ of the output line voltage depends on the spatial vector position α _i of the input line voltage, and the spatial vector position β _i of the input line current is also the spatial vector position β of the output line current. _o ). Since the purpose of the spatial vector modulation is to control the output line voltage and the input line current, group III is not suitable for the space vector method and will be omitted.

In Group I, two output line voltages are connected to one input line voltage, and one output line voltage is zero. Unlike group III, 18 switching arrays in group I determine the positions of six space vectors of output line voltage and input line current independently of input line voltage and output line voltage. Therefore, the group I becomes an effective switching arrangement for the spatial vector modulation method.

In Group II, all three output line voltages are directly connected to one input line voltage, so all output line voltages are zero. The input currents are all zero during the zero vector period, and the load current is refluxed through the switching of the matrix converter 230. Therefore, group II becomes a switching arrangement effective for the spatial vector modulation method.

As a result, 18 non-zero spatial vectors belonging to group I (± 1, ± 2, ..., ± 9) and three zero spatial vectors belonging to group II (0 _a , 0 _b , 0 _c ) can be used for spatial vector modulation according to the present invention.

4A and 4B show a relationship between an output phase voltage vector and an input current vector according to a switching form when the input voltage of the matrix converter 220 is as shown in Equation 9 below, and the state space is divided into six sections. The voltage vector and the current vector existing in each section can be expressed using adjacent space vectors.

Therefore, the input voltage vector and the output voltage vector for spatial vector modulation of the matrix converter 230 can be expressed as in Equation 10.

In the same way, the input current vector and output current vector for spatial vector modulation of the matrix converter 230 can be expressed as in Equation (11).

4A and 4B are diagrams for describing a space vector modulation method of the space vector modulator 260 when the voltage vector and the current vector input to the matrix converter 230 are in the same section in the state space.

When the input voltage vector and the input current vector input to the matrix converter 230 are in the same section in the state space defined as shown in FIGS. 4A and 4B, a suitable switching pattern is obtained. For convenience of analysis, assuming that the input voltage vector and the output voltage vector are in interval 1 (-π / 6≤α _i ≤π / 6, 0≤α _o ≤π / 3), a reference output voltage v ₀ 'is generated. Suitable switching vectors can be obtained from Table 1 -7 and +9. If the output voltage vector corresponding to the switching type (-9≤j≤ + 9) of Table 1 is V _j , the voltage vectors V _-7 and Each switching application time (switching cycle) d ₁ , d ₂ for V _{+ 9} satisfies the condition of Equations 12 and 13.

If the phase difference between the desired input voltage and the input current from Equation 12 and 13 is set to δ as shown in FIG. 4B and generalized to the case where the output voltage vector and the input voltage vector exist in an arbitrary section, d ₁ and d ₂ are represented by , 15

Where α ₀ is the phase of the output voltage, α _i is the phase of the input voltage vector, δ is the phase difference between the input voltage vector and the input current vector, k _v is the section in which the output voltage vector is located, and k _i is the input current vector (and Where the input voltage vector) is located, q is the voltage conversion rate ( V ₀ / V _m ), and the magnitude thereof is expressed by Equation 16 below.

In a similar manner, the switching vectors suitable for generating the reference output voltage v ₀ '' are +1, -3, and the respective voltage vectors V ₊₁ and Each switching application time (switching cycle) for V- ₃ can be obtained, and the voltage and current vectors are k _v , When the switching application time of each switching vector is obtained by generalizing to the case of k _i , the following equations (17) and (18) are used.

In addition, since the switching application time d _{5 to} which the zero voltage vector is applied must satisfy Equation 19, Equation 20 is obtained.

Table 2 (a switching pattern when the input voltage vector and the input current vector are in the same section) indicates possible switching states according to the input voltage vector and the output voltage vector when the input voltage vector and the input current vector are in the same section.

5A and 5B are diagrams for describing a space vector modulation method of the space vector modulator 260 when the voltage vector and the current vector input to the matrix converter are in different sections in the state space.

In the state space defined in FIGS. 5A and 5B, when a voltage vector and a current vector input to the matrix converter are in different sections, a suitable switching pattern is obtained. For convenience of analysis, the input voltage vector and the output voltage vector are in interval 1 (-π / 6≤α _i ≤π / 6, 0≤α _o ≤π / 3), and the input current vector is in interval 6 (0≤β Assuming that _i ≤ 11π / 6), the switching type that can generate the reference output voltage v ₀ 'is -7, +8 from Table 1, and corresponding voltage vectors V _-7 and The switching application time (switching cycle) d ₁ and d ₂ to which V _{+ 8} is applied satisfies the conditions of Equations 21 and 22.

When using Equation 21 and 22 put as δ, as the phase difference between the desired input voltage and input current and Figure 4b, the output voltage vector is generalized to k _v interval, when the input voltage vector is present in the k _i interval, d _1, d ₂ is represented by Equations 23 and 24, respectively.

Where α ₀ is the phase of the output voltage, α _i is the phase of the input voltage vector, δ is the phase difference between the input voltage vector and the input current vector, k _v is the section in which the output voltage vector is located, and k _i is the input current vector (and Where q is the voltage conversion factor ( V ₀ / V _m) .

In the same way, the switching forms suitable for generating the reference output voltage v ₀ '' are +1, -2, and the respective voltage vectors V ₊₁ and Each switching application time (switching cycle) for V _-2 can be obtained, and the voltage and current vectors are k _v , When the switching application time is obtained by generalizing to the case where k _i is present, Equations 25 and 26 are as follows.

In addition, the switching application time d ₅ for the zero voltage vector satisfying Equation 19 is expressed by Equation 27.

Table 3 below (switching pattern when input voltage vector and input current vector are in different sections) shows possible switching states according to input voltage vector and output voltage vector when input voltage vector and input current vector are in different sections. Indicates.

가 되고, 역률이 1인 경우에서

가 된다. 본 발명에서 매트릭스 컨버터의 입력전압과 입력전류의 위상차를 보상할 수 있는 최대값 δ_max는 출력전압에 따라 다르며, 그 크기는 아래의 수학식 28과 같다. On the other hand, as can be seen from the equation representing the switching cycle, the maximum value of the voltage conversion factor ( V ₀ / V _m ) q is

Where the power factor is 1

Becomes In the present invention, the maximum value δ _max that can compensate the phase difference between the input voltage and the input current of the matrix converter depends on the output voltage, and the magnitude thereof is expressed by Equation 28 below.

이면,

If so,

이면,

If so,

6 is a flowchart illustrating a method for improving power factor according to an embodiment of the present invention.

According to the power factor improvement method of FIG. 6, the actual phase difference φ between the input voltage v _s and the input current i _s is calculated by measuring the input voltage and the input current input to the input filter unit 220 (S610). Specifically, since the voltage drop caused by the inductor in FIG. 1B is small enough to be neglected by the input voltage, the voltage applied to the matrix converter 230 may be regarded as the input power supply voltage. Therefore, it is possible to know the phase difference between the input voltage and the input current applied to the matrix converter by measuring the input power current.

On the basis of the actual phase difference φ calculated in step S610 it is determined whether the input power factor compensation is necessary (S620). Specifically, when the power factor value by the actual phase difference φ calculated in step S510 is smaller than the predetermined threshold value (or when | sin (φ) | is greater than the predetermined value), it may be determined that power factor correction is necessary. . For example, when the threshold value is set to 0.995, it may be determined that power factor correction is necessary when the power factor value due to the actual phase difference φ is smaller than 0.995 (or when {sin (φ) |> 0.1).

If it is determined in step S620 that compensation of the input power factor is necessary, a target phase difference between the input voltage and the input current is calculated (S630). Specifically, the target phase difference between the input voltage and the input current may be obtained using Equation 8 described above, and the calculated target phase difference should satisfy Equation 28. That is, if the calculated target phase difference δ is smaller than δ _max of equation (28) has a desired phase difference δ used, and the calculated target phase difference δ is larger than the maximum value δ _max of equation (28) the output has the maximum value δ _max The spatial vector modulation method can be applied.

The spatial vector modulation is performed using the target phase difference calculated in step S630 (S640).

FIG. 7 is a flowchart illustrating the spatial vector modulation method of FIG. 6 in detail.

According to the spatial vector modulation method of FIG. 7, first, a state space for spatial vector modulation is divided into a plurality of regions (S710). For example, as shown in FIGS. 3A, 3B, 4A, and 4B, the state spaces of the output voltage vector and the input voltage (current) vector may be divided into six regions.

The reference output voltage vector, which is an adjacent vector component of the output voltage vector located in the predetermined section, is detected among the plurality of sections divided in operation S710 (S720). Here, the reference output voltage vector is the output voltage vector v ₀ Of the adjacent vector components of a first reference output voltage vector v ₀ and a second reference output voltage vector v '' includes a _0, the first reference output voltage vector v switching vector capable of producing a _zero are the first and the The switching vectors, which are two switching vectors and can generate the second reference output voltage vector v '' ₀ , can be the third and fourth switching vectors.

The switching application time of each switching vector is calculated using the switching vectors selected in step S720 and the target phase difference calculated in step S630. Specifically, the switching application time for the first and second switching vectors is calculated using the first and second switching vectors for the first reference output voltage vector v ' ₀ and the target phase difference δ calculated in step S630. The switching application time for the first and second switching vectors is calculated using the third and fourth switching vectors for the second reference output voltage vector v '' ₀ and the target phase difference δ calculated in step S630. Since a detailed calculation method of the switching application time for the switching vectors has been described above with reference to FIGS. 1 to 5, Tables 1 to 3, and Equations 1 to 28, a detailed description thereof will be omitted.

Meanwhile, the above-described power factor correction method may be applied to an open loop method or a closed loop method. When the power factor is less than or equal to the threshold value, the phase difference δ for obtaining a desired input power factor is obtained and the power factor correction method shown in FIG. 6 is performed only once, whereas the closing method performs feedback control to suppress the desired input power factor. In other words, if the power factor is measured at regular intervals and out of a certain range, the power factor correction method shown in FIG. 6 is automatically performed. In general, the phase difference δ between the matrix converter input voltage and the input current for setting the input power factor to 1 can be obtained by using Equation 8. However, when the value is compensated, the magnitude of the input current changes greatly due to the change of the power factor. As a result, the power factor may change again, making it difficult to obtain a desired power factor. In addition, the power factor value may change depending on the load variation. Therefore, the open compensation method is preferably used when the system characteristics are well known, and in general, the closed compensation method can be applied.

    On the other hand, the validity of the proposed spatial vector modulation method is verified using Matlab / Simulink, and the loads and constants used in the simulation are as follows.

ㆍ Input Voltage: 3 Phase 220V / 60Hz

ㆍ 3 phase RL load: R = 10Ω, 5mH

ㆍ Input Filter: L = 2mH, C = 50μF

8A and 8B are diagrams for comparing input power factor values with respect to the spatial vector modulation method and the conventional spatial vector modulation method according to the present invention under given conditions when there is an input filter in the matrix converter.

8A shows the change of the input power factor according to the load current variation. In this case, the output voltage was kept constant at 60V and 40Hz. As can be seen in FIG. 8A, unlike the conventional switching method, the switching method according to the present invention can be seen that the power factor can be maintained at almost all regions except for a very small load.

8b shows the change in power factor according to the change of the output frequency at a given three-phase RL load. In this case, the ratio of the output voltage and the output frequency was kept constant.

9a to 9c show the power factor correction when the output frequency is 30 Hz, 40 Hz, and 60 Hz, respectively, when the input power factor is controlled by the conventional spatial vector modulation method and the opening and closing methods of the spatial vector modulation method according to the present invention. The change in input power factor is shown. As shown in FIGS. 9A to 9C, it is difficult to set the input power factor to 1 by the opening method. This is because the phase angle δ value calculated to make the input power factor 1 in the open-circuit method is not optimal because the input current is changed through power factor correction, and the phase angle suitable for this current is different from the previous one. Because. In particular, at low frequencies such as 30 Hz and 40 Hz, the current change due to power factor correction is large, which means that it is difficult to set the power factor to 1 by the open method.

FIG. 10A illustrates the input / output voltage and current waveforms of the matrix converter 230 according to the conventional spatial vector modulation method when the output voltage is 60 V and the output frequency is 40 Hz. If the input filter 220 controls the input voltage and the input current of the matrix converter 230 to be in phase, it can be seen that the power current phase is ahead of the power voltage phase. FIG. 10B illustrates an input / output voltage and a current waveform of the matrix converter 230 when the closed power control is performed using the spatial vector modulation method according to the present invention to set the input power factor to 1 under the same conditions as in FIG. 10A. As can be seen in Figure 10b the power supply voltage and current is in phase, it can be seen that the input current is significantly reduced compared to the conventional spatial vector modulation method due to the power factor improvement.

11A and 11B show an output voltage of 75 V and an output frequency of 60 Hz, the contents of which correspond to FIGS. 10A and 10B, respectively.

FIG. 12 shows the output voltage and current of the matrix converter 230, the power factor correction characteristic, the main power supply voltage and the current by changing the output voltage stepwise from 90V, 60Hz to 60V, and 40Hz in order to examine the transient characteristics of the proposed closure compensation scheme. And the relationship between the voltage and current input to the matrix converter 230 through the filter is shown at the same time. As can be seen in FIG. 12, not only the power factor correction is well performed by the proposed algorithm but also the electrical characteristics of various voltages and currents are excellent.

The matrix converter (the DSP TI 20F2812 and the ALTERA EPM7128 FPGA are used to drive the matrix converter switching element for the implementation of the space vector algorithm) used in the embodiments according to FIGS. Phosphorus low voltage input was used, but the voltage conversion ratio q of the matrix converter 230 was 0.47 and 0.71, so that the output voltage and output frequency such as the output voltage of 60V and 90V were simulated.

13A and 14A illustrate an input voltage and an input current of a power source and a voltage and a current applied to a matrix converter when the output frequency is 40 Hz or 60 Hz and controlled by a conventional space vector control method. 13B and 14B are cases where the spatial vector control method according to the present invention is used when the output frequencies are 40 Hz and 60 Hz, respectively. 13C and 14C show output voltages and currents of a matrix converter applied to a load under the same experimental conditions when the output frequencies are 40 Hz and 60 Hz, respectively.

As can be seen from the experimental results, in the conventional method, the voltage and current phase of the matrix converter match, but the voltage and current of the power supply have a phase difference due to the input filter. However, in the case of using the space vector control method according to the present invention, the input power factor can be maintained at 1 by equalizing the power supply voltage and the current, and the power supply current becomes smaller under the same output voltage and load conditions.

Accordingly, when the input filter is used in the matrix converter, the input power factor can be maintained at almost 1 in almost all load regions even when the load or output frequency is changed. In particular, it is possible to compensate for the severe power factor drop at light loads.

1 is a circuit diagram showing the configuration of a conventional matrix converter.

2 is a block diagram illustrating a configuration of a matrix converter system according to an embodiment of the present invention.

FIG. 3 is a circuit diagram illustrating a per phase equivalent circuit of the input filter illustrated in FIGS. 1 and 2.

4A and 4B are diagrams for describing a space vector modulation method when the voltage vector and the current vector input to the matrix converter are in the same section in the state space.

5A and 5B are diagrams for describing a space vector modulation method when a voltage vector and a current vector input to a matrix converter are in different sections in a state space.

6 is a flowchart illustrating a method for improving power factor according to an embodiment of the present invention.

FIG. 7 is a flowchart illustrating the spatial vector modulation method of FIG. 6 in detail.

8 to 14 are diagrams for comparing and explaining a power factor value according to a spatial vector modulation method and a conventional spatial vector modulation method according to various embodiments of the present disclosure.

Explanation of symbols on the main parts of the drawing

210: power supply unit 220: input filter unit

230: matrix converter 240: detector

250: calculator 260: spatial vector modulator

Claims (18)

In the method of improving the power factor of the matrix converter connected to the input filter, Calculating an actual phase difference between an input voltage and an input current input to the matrix converter by measuring an input voltage and an input current input to the input filter; Determining whether an input power factor compensation is necessary based on the calculated actual phase difference; Calculating a target phase difference between an input voltage and an input current input to the matrix converter when it is determined that compensation of an input power factor is necessary; And And performing spatial vector modulation by using the calculated target phase difference. Performing the spatial vector modulation, Dividing a state space for spatial vector modulation into a plurality of sections; Detecting a reference output voltage vector, which is an adjacent vector component of an output voltage vector positioned in a predetermined section of the plurality of sections; Selecting switching vectors capable of generating the reference output voltage vector; And Calculating a switching application time of each switching vector by using the switching vectors and the calculated target phase difference; The reference output voltage vector includes a first reference output voltage vector and a second reference output voltage vector, which are adjacent vector components of the output voltage vector, and switching vectors capable of generating the first reference output voltage vector include first and second reference output voltage vectors. And a second switching vector, wherein the switching vectors capable of generating the second reference output voltage vector are third and fourth switching vectors. delete delete The method of claim 1, When the input voltage vector and the input current vector input to the matrix converter are located in the same section, The switching power application time d1 and d2 for the first and second switching vectors are respectively calculated by the following equations.

Where α ₀ is the phase of the output voltage vector, α _i is the phase of the input voltage vector, δ is the phase difference between the input voltage vector and the input current vector, k _v is the section in which the output voltage vector is located, and k _i is the input current vector ( And an interval in which the input voltage vector) is located, q is a voltage conversion rate ( V ₀ / V _m) . 5. The method of claim 4, When the input voltage vector and the input current vector input to the matrix converter are located in the same section, The switching application times d3 and d4 for the third and fourth switching vectors are respectively calculated by the following equations.

Where α ₀ is the phase of the output voltage vector, α _i is the phase of the input voltage vector, δ is the phase difference between the input voltage vector and the input current vector, k _v is the section in which the output voltage vector is located, and k _i is the input current vector ( And an interval in which the input voltage vector) is located, q is a voltage conversion rate ( V ₀ / V _m) . The method of claim 5, When the input voltage vector and the input current vector input to the matrix converter are located in the same section, The switching application time d5 to which the zero voltage vector is applied is calculated by the following equation.

Where α ₀ is the phase of the output voltage vector, α _i is the phase of the input voltage vector, δ is the phase difference between the input voltage vector and the input current vector, k _v is the section where the output voltage vector is located, and k _i is the input current vector ( And an interval in which the input voltage vector) is located, q is a voltage conversion rate ( V ₀ / V _m) . The method of claim 1, When the input voltage vector and the input current vector input to the matrix converter are located in different sections, The switching power application time d1 and d2 for the first and second switching vectors are respectively calculated by the following equations.

Where α ₀ is the phase of the output voltage vector, α _i is the phase of the input voltage vector, δ is the phase difference between the input voltage vector and the input current vector, k _v is the section where the output voltage vector is located, and k _i is the input current vector. Q is the voltage conversion rate ( V ₀ / V _m) . The method of claim 7, wherein When the input voltage vector and the input current vector input to the matrix converter are located in different sections, The switching application times d3 and d4 for the third and fourth switching vectors are respectively calculated by the following equations.

Where α ₀ is the phase of the output voltage vector, α _i is the phase of the input voltage vector, δ is the phase difference between the input voltage vector and the input current vector, k _v is the section where the output voltage vector is located, and k _i is the input current vector. Q is the voltage conversion rate ( V ₀ / V _m) . The method of claim 8, When the input voltage vector and the input current vector input to the matrix converter are located in different intervals, the switching application time d5 to which the zero voltage vector is applied is calculated by the following equation.

Where α ₀ is the phase of the output voltage, α _i is the phase of the input voltage vector, δ is the phase difference between the input voltage vector and the input current vector, k _v is the section in which the output voltage vector is located, and k _i is the input current vector. Where q is the voltage conversion rate ( V ₀ / V _m) . A matrix converter system comprising an input filter and a matrix converter coupled to the input filter, A detector configured to measure an input voltage and an input current input to the input filter and calculate an actual phase difference between the input voltage and the input current input to the matrix converter; A calculator configured to determine whether compensation of an input power factor is necessary based on the calculated actual phase difference, and calculate a target phase difference between an input voltage and an input current input to the matrix converter when it is determined that compensation of an input power factor is necessary; And And a spatial vector modulator for performing spatial vector modulation using the calculated target phase difference. The space vector modulator, The selected switching by detecting a reference output voltage vector that is an adjacent vector component of an output voltage vector located in a predetermined section among a plurality of sections separated for spatial vector modulation, and selecting switching vectors capable of generating the reference output voltage vector; A switching application time of each switching vector is calculated using the vectors and the calculated target phase difference, The reference output voltage vector includes a first reference output voltage vector and a second reference output voltage vector, which are adjacent vector components of the output voltage vector, and switching vectors capable of generating the first reference output voltage vector include first and second reference output voltage vectors. And a second switching vector, wherein the switching vectors capable of generating the second reference output voltage vector are third and fourth switching vectors. delete delete The method of claim 10, When the input voltage vector and the input current vector input to the matrix converter are located in the same section, Matrix switching system, characterized in that the switching application time d1, d2 for the first and second switching vectors are respectively calculated by the following equation:

Where α ₀ is the phase of the output voltage vector, α _i is the phase of the input voltage vector, δ is the phase difference between the input voltage vector and the input current vector, k _v is the section in which the output voltage vector is located, and k _i is the input current vector ( And an interval in which the input voltage vector) is located, q is a voltage conversion rate ( V ₀ / V _m) . The method of claim 13, When the input voltage vector and the input current vector are located in the same section, Matrix switching system, characterized in that the switching application time d3, d4 for the third and fourth switching vectors are respectively calculated by the following equation:

Where α ₀ is the phase of the output voltage vector, α _i is the phase of the input voltage vector, δ is the phase difference between the input voltage vector and the input current vector, k _v is the section in which the output voltage vector is located, and k _i is the input current vector ( And an interval in which the input voltage vector) is located, q is a voltage conversion rate ( V ₀ / V _m) . The method of claim 14, When the input voltage vector and the input current vector are located in the same section, The switching converter time d5 to which the zero voltage vector is applied is calculated by the following equation:

Where α ₀ is the phase of the output voltage vector, α _i is the phase of the input voltage vector, δ is the phase difference between the input voltage vector and the input current vector, k _v is the section in which the output voltage vector is located, and k _i is the input current vector ( And an interval in which the input voltage vector) is located, q is a voltage conversion rate ( V ₀ / V _m) . The method of claim 10, When the input voltage vector and the input current vector are located in different intervals, Matrix switching system, characterized in that the switching application time d1, d2 for the first and second switching vectors are respectively calculated by the following equation:

Where α ₀ is the phase of the output voltage vector, α _i is the phase of the input voltage vector, δ is the phase difference between the input voltage vector and the input current vector, k _v is the section where the output voltage vector is located, and k _i is the input current vector. Q is the voltage conversion rate ( V ₀ / V _m) . The method of claim 16, When the input voltage vector and the input current vector are located in different intervals, Matrix switching system, characterized in that the switching application time d3, d4 for the third and fourth switching vectors are respectively calculated by the following equation:

Where α ₀ is the phase of the output voltage vector, α _i is the phase of the input voltage vector, δ is the phase difference between the input voltage vector and the input current vector, k _v is the section where the output voltage vector is located, and k _i is the input current vector. Q is the voltage conversion rate ( V ₀ / V _m) . The method of claim 17, When the input voltage vector and the input current vector are located in different intervals, The switching converter time d5 to which the zero voltage vector is applied is calculated by the following equation:

Where α ₀ is the phase of the output voltage vector, α _i is the phase of the input voltage vector, δ is the phase difference between the input voltage vector and the input current vector, k _v is the section where the output voltage vector is located, and k _i is the input current vector. Q is the voltage conversion rate ( V ₀ / V _m) .

Images