CN101674046B

CN101674046B - Electric current reconstructing and over-modulating device of air conditioning frequency converter and method thereof

Info

Publication number: CN101674046B
Application number: CN2009100923453A
Authority: CN
Inventors: 孙凯; 魏庆; 黄立培; 马鸿雁; 史宇超
Original assignee: Tsinghua University
Current assignee: Tsinghua University
Priority date: 2009-09-07
Filing date: 2009-09-07
Publication date: 2011-09-14
Anticipated expiration: 2029-09-07
Also published as: CN101674046A

Abstract

The invention relates to a current reconstruction overmodulation device of an air conditioner frequency converter, which is characterized in that it includes a current reconstruction device, a vector controller, a pulse width modulation (PWM) frequency converter, a DC side current sensor, a motor and an encoder; The current reconstruction device includes a SVPWM overmodulation module, a pulse width modulation (PWM) signal adjustment module, a DC current sampling module and an AC current reconstruction module; Carry out two samplings, and then calculate the action time of the adjacent vector switch that meets the minimum sampling time requirement according to the two sampling values, and sample the DC side current twice again within the action time of the adjacent vector switch, according to the sampling of the two DC currents The value reconstructs the three-phase AC current, so as to realize the closed-loop control of the pulse width modulation (PWM) frequency converter. The modulation method of the invention can be widely used in the field of current reconstruction over-modulation of frequency converters in air conditioners and electric vehicles.

Description

A current reconstruction overmodulation device and method for an air conditioner inverter

技术领域technical field

本发明涉及一种电机及变频器调制装置及方法，尤其是一种空调变频器的电流重构过调制装置及方法。The invention relates to a modulation device and method for a motor and a frequency converter, in particular to a current reconstruction overmodulation device and method for an air conditioner frequency converter.

背景技术Background technique

随着世界范围内节能降耗技术的积极推广，变频空调因其节省耗电量的特点正在逐渐受到市场的关注。现阶段变频空调的电机主要采用永磁同步电机，永磁同步电机具有结构简单、体积小、重量轻、损耗小和效率高等特点。然而，在传统的永磁同步电机的矢量控制系统中，由于需要测量交流电流信号作为反馈，实现电流的闭环控制，因此变频器的交流侧需要3个电流传感器。另外，为保障系统的安全性，还需要过载保护和直流短路保护，因此在母线上还需要一个电流传感器测量直流母线电流。这样，整个系统总共需要4个电流传感器，造成整个变频装置的成本高、结构复杂、体积大，不利于集成化。With the active promotion of energy-saving and consumption-reducing technologies worldwide, inverter air conditioners are gradually attracting the attention of the market because of their power-saving features. At present, the motors of frequency conversion air conditioners mainly use permanent magnet synchronous motors, which have the characteristics of simple structure, small size, light weight, low loss and high efficiency. However, in the traditional vector control system of the permanent magnet synchronous motor, since the AC current signal needs to be measured as feedback to realize the closed-loop control of the current, three current sensors are required on the AC side of the inverter. In addition, in order to ensure the safety of the system, overload protection and DC short-circuit protection are also required, so a current sensor is required on the bus to measure the DC bus current. In this way, the whole system needs 4 current sensors in total, resulting in high cost, complex structure and large volume of the whole frequency conversion device, which is not conducive to integration.

除此之外，由于绝大多数空调压缩机都需要运行在中高速区，而压缩机调速系统的性能又受变频器输出电压的影响，目前三相桥式电压型变频器采用SPWM(正弦电压脉宽调制，Sine Pulse Width Modulation)和SVPWM(空间矢量电压脉宽调制，Space Vector Pulse Width Modulation)来产生PWM(脉宽调制，Pulse WidthModulation)信号，而SPWM算法虽然结构简单，实现方便，但是变频器输出相电压的基波幅值较低，相比之下SVPWM算法具有电压利用率较高的特点，即最大输出电压的基波幅值较高。但是SVPWM算法仍然存在未充分利用直流母线电压的问题，因此要对变频器进行过调制控制，以提高电源电压利用率。In addition, since most air-conditioning compressors need to operate in the medium-high speed area, and the performance of the compressor speed control system is affected by the output voltage of the inverter, the current three-phase bridge voltage inverter adopts SPWM (sinusoidal Voltage Pulse Width Modulation, Sine Pulse Width Modulation) and SVPWM (Space Vector Voltage Pulse Width Modulation, Space Vector Pulse Width Modulation) to generate PWM (Pulse Width Modulation, Pulse Width Modulation) signal, and although the SPWM algorithm is simple in structure and easy to implement, but The fundamental wave amplitude of the output phase voltage of the frequency converter is relatively low. In contrast, the SVPWM algorithm has the characteristics of high voltage utilization, that is, the fundamental wave amplitude of the maximum output voltage is relatively high. However, the SVPWM algorithm still has the problem of not fully utilizing the DC bus voltage, so it is necessary to over-modulate the frequency converter to improve the utilization of the power supply voltage.

发明内容Contents of the invention

针对上述问题，本发明的目的是提供一种结构简单、可以提高电源电压利用率的空调变频器的电流重构过调制装置及方法。In view of the above problems, the object of the present invention is to provide a current reconstruction overmodulation device and method for an air conditioner inverter with a simple structure and which can improve the utilization rate of the power supply voltage.

为实现上述目的，本发明采取以下技术方案：一种空调变频器的电流重构过调制装置，其特征在于：它包括电流重构装置、矢量控制器、脉宽调制(PWM)变频器、直流侧电流传感器、电机和编码器；其中，所述电流重构装置包括正弦电压脉宽调制(SVPWM)过调制模块、脉宽调制(PWM)信号调节模块、直流电流采样模块和交流电流重构模块；所述电流重构装置与矢量控制器共同对脉宽调制(PWM)变频器进行闭环控制的步骤为：首先，所述直流电流采样模块在所述脉宽调制(PWM)信号调节模块输入的采样点控制信号的作用下，通过所述直流侧电流传感器对所述脉宽调制(PWM)变频器的直流侧进行两次直流电流采样，将所述两次直流电流采样值输入到所述交流电流重构模块中，所述交流电流重构模块结合所述正弦电压脉宽调制(SVPWM)过调制模块输入的扇区号信号重构出三相交流电流，将所述三相交流电流值输入所述矢量控制器，结合所述编码器输出的转子位置角，计算得到所述参考电压矢量V_r的值；In order to achieve the above object, the present invention adopts the following technical solutions: a current reconstruction overmodulation device for an air conditioner frequency converter, which is characterized in that it includes a current reconstruction device, a vector controller, a pulse width modulation (PWM) frequency converter, a DC A side current sensor, a motor and an encoder; wherein the current reconstruction device includes a sinusoidal voltage pulse width modulation (SVPWM) overmodulation module, a pulse width modulation (PWM) signal adjustment module, a DC current sampling module and an AC current reconstruction module The step that the current reconstruction device and the vector controller jointly carry out closed-loop control of the pulse width modulation (PWM) frequency converter is: first, the DC current sampling module is input by the pulse width modulation (PWM) signal adjustment module Under the action of the control signal of the sampling point, the DC side of the pulse width modulation (PWM) frequency converter is subjected to two DC current samples through the DC side current sensor, and the two DC current sample values are input to the AC In the current reconstruction module, the AC current reconstruction module reconstructs a three-phase AC current in combination with the sector number signal input by the sinusoidal voltage pulse width modulation (SVPWM) overmodulation module, and inputs the three-phase AC current value into the Described vector controller, in conjunction with the rotor position angle that described encoder outputs, calculates the value of described reference voltage vector V _r ;

所述正弦电压脉宽调制(SVPWM)过调制模块对所述参考电压矢量V_r进行矢量分解，若经过矢量分解得到临近有效矢量的作用时间不满足最小采样时间的要求，则调整所述临近有效矢量的作用时间；将调整后的所述临近有效矢量的作用时间发送给所述脉宽调制(PWM)信号调节模块，所述脉宽调制(PWM)信号调节模块根据所述各个调整后的临近矢量的作用时间控制所述脉宽调制(PWM)变频器中各个开关的状态的持续时间，并向所述直流电流采样模块输出采样点控制信号，所述直流电流采样模块在采样点控制信号的作用下通过所述直流侧电流传感器对所述脉宽调制(PWM)变频器的直流侧进行两次直流电流采样，将所述两次直流电流采样值输入到所述交流电流重构模块中，所述交流电流重构模块结合所述正弦电压脉宽调制(SVPWM)过调制模块输入的扇区号信号重构出三相交流电流。The sinusoidal voltage pulse width modulation (SVPWM) overmodulation module performs vector decomposition on the reference voltage vector V _r , if the action time of the adjacent effective vector obtained through vector decomposition does not meet the minimum sampling time requirement, then adjust the adjacent effective vector The action time of the vector; the adjusted action time of the adjacent effective vector is sent to the pulse width modulation (PWM) signal adjustment module, and the pulse width modulation (PWM) signal adjustment module is based on each adjusted approach The action time of the vector controls the duration of the state of each switch in the pulse width modulation (PWM) frequency converter, and outputs the sampling point control signal to the direct current sampling module, and the direct current sampling module is at the sampling point control signal Under the action, the DC side current sensor of the pulse width modulation (PWM) frequency converter is used to perform two DC current samples on the DC side, and the two DC current sampled values are input into the AC current reconstruction module, The alternating current reconstruction module reconstructs a three-phase alternating current by combining the sector number signal input by the sinusoidal voltage pulse width modulation (SVPWM) overmodulation module.

一种实现所述装置的空调变频器的电流重构过调制方法，其特征在于：首先设置空调变频器电流重构过调制的装置，所述空调变频器电流重构装置包括电流重构装置、矢量控制器、脉宽调制(PWM)变频器、直流侧电流传感器；其中，所述电流重构装置包括正弦电压脉宽调制(SVPWM)过调制模块、脉宽调制(PWM)信号调节模块、直流电流采样模块和交流电流重构模块；所述矢量控制器根据所述交流电流重构模块输入的直流电流采样值计算得出参考电压矢量V_r，然后将所述参考电压矢量V_r输入到所述正弦电压脉宽调制(SVPWM)过调制模块，所述正弦电压脉宽调制(SVPWM)过调制模块根据方程(1)计算调制比M，A current reconstruction and over-modulation method for an air-conditioning frequency converter that implements the device, characterized in that: firstly, a device for current reconstruction and over-modulation of the air-conditioning frequency converter is provided, and the current reconstruction device for the air-conditioning frequency converter includes a current reconstruction device, A vector controller, a pulse width modulation (PWM) frequency converter, and a DC side current sensor; wherein, the current reconstruction device includes a sinusoidal voltage pulse width modulation (SVPWM) overmodulation module, a pulse width modulation (PWM) signal adjustment module, a DC A current sampling module and an AC current reconstruction module; the vector controller calculates a reference voltage vector V _r according to the DC current sampling value input by the AC current reconstruction module, and then inputs the reference voltage vector V _r to the Described sinusoidal voltage pulse width modulation (SVPWM) overmodulation module, described sinusoidal voltage pulse width modulation (SVPWM) overmodulation module calculates modulation ratio M according to equation (1),

$M m = = \frac{| | {V V}_{r r} | |}{{U u}_{dc dc} / / \sqrt{33}} - - - - - - ((11))$

其中U_dc为直流母线电压；然后计算最大线性调制下的电压利用率η_linear、一般过调制方式下的电压利用率η_edge和最大过调制方式下的电压利用率η_limit；Wherein U _dc is the DC bus voltage; Then calculate the voltage utilization rate η _linear under the maximum linear modulation, the voltage utilization rate η _edge under the general over-modulation mode and the voltage utilization rate η _limit under the maximum over-modulation mode;

若调制比M过低，或者所述参考电压矢量V_r接近六个有效矢量中的任一个时，则所述正弦电压脉宽调制(SVPWM)过调制模块选择修改开关状态法对各邻近有效矢量的作用时间进行调整；若所述参考电压矢量V_r落在无效区域，则选择电流重构过调制法对各临近矢量的作用时间进行调整；使所述脉宽调制(PWM)变频器中的作用时间满足所述最小采样时间要求；If the modulation ratio M is too low, or when the reference voltage vector V _r is close to any one of the six effective vectors, the sinusoidal voltage pulse width modulation (SVPWM) overmodulation module selects the method of modifying the switch state for each adjacent effective vector The action time of each adjacent vector is adjusted; if the reference voltage vector V _r falls in an invalid region, then select the current reconstruction over-modulation method to adjust the action time of each adjacent vector; make the pulse width modulation (PWM) in the frequency converter The action time meets the minimum sampling time requirement;

所述电流重构过调制方法步骤包括：所述正弦电压脉宽调制(SVPWM)过调制模块将所述调制比M与最大线性调制下的电压利用率η_linear、一般过调制方式下的电压利用率η_edge和最大过调制方式下的电压利用率η_limit比较：The steps of the current reconstruction overmodulation method include: the sinusoidal voltage pulse width modulation (SVPWM) overmodulation module compares the modulation ratio M with the voltage utilization ratio η _linear under the maximum linear modulation, and the voltage utilization ratio under the general overmodulation mode Comparison between the rate η _edge and the voltage utilization rate η _limit under the maximum overmodulation mode:

设应用所述电流重构过调制方法后得到的调制后参考电压矢量为V′_r，Let the modulated reference voltage vector obtained after applying the current reconstruction overmodulation method be V′ _r ,

A)若调制比M≤最大线性调制下的电压利用率η_linear，此时不需要对所述参考电压矢量V_r做调整，也即调制后参考电压矢量V′_r＝V_r；A) If the modulation ratio M≤the voltage utilization rate η _linear under the maximum linear modulation, at this time there is no need to adjust the reference voltage vector V _r , that is, the modulated reference voltage vector V′ _r =V _r ;

B)若最大线性调制下的电压利用率η_linear＜调制比M≤一般过调制方式下的电压利用率η_edge，此时令系数B) If the voltage utilization rate η _linear under the maximum linear modulation < modulation ratio M ≤ the voltage utilization rate η _edge under the general over-modulation mode, the seasonal coefficient

${k k}_{11} = = \frac{M m - - {η η}_{linear linear}}{{η η}_{edge edge} - - {η η}_{linear linear}}$

调制后参考电压矢量V′_r＝k₁V_edge+(1-k₁)V_linear；Modulated reference voltage vector V′ _r =k ₁ V _edge +(1-k ₁ )V _linear ;

C)若一般过调制方式下的电压利用率η_edge＜调制比M≤最大过调制方式下的电压利用率η_limit，此时令系数C) If the voltage utilization rate η _edge under the general over-modulation mode < modulation ratio M ≤ the voltage utilization rate η _limit under the maximum over-modulation mode, the seasonal coefficient

${k k}_{22} = = \frac{M m - - {η η}_{edge edge}}{{η η}_{limit limit} - - {η η}_{edge edge}}$

调制后参考电压矢量V′_r＝k₂V_limit+(1-k₂)V_edge；Modulated reference voltage vector V′ _r =k ₂ V _limit +(1-k ₂ )V _edge ;

所述正弦电压脉宽调制(SVPWM)过调制模块利用正弦电压脉宽调制(SVPWM)算法计算两临近有效矢量作用时间T₁、T₂和零矢量的作用时间T₀：The sinusoidal voltage pulse width modulation (SVPWM) overmodulation module utilizes the sinusoidal voltage pulse width modulation (SVPWM) algorithm to calculate the action time T ₀ of two adjacent effective vector action times T ₁ , T ₂ and the zero vector:

$\{\begin{matrix} {T T}_{11} = = \sqrt{33} {T T}_{s the s} | | {V V}_{r r}^{' '} | | sin sin {θ θ}_{r r}^{' '} / / {U u}_{dc dc} \\ {T T}_{22} = = \sqrt{33} {T T}_{s the s} | | {V V}_{r r}^{' '} | | sin sin ((π π / / 33 - - {θ θ}_{r r}^{' '})) / / {U u}_{dc dc} \\ {T T}_{00} = = {T T}_{s the s} - - {T T}_{11} - - {T T}_{22} \end{matrix}$

其中，θ′_r为调制后参考电压矢量为V′_r的相角，所述正弦电压脉宽调制(SVPWM)过调制模块将所述两临近有效矢量作用时间T₁、T₂和零矢量的作用时间T₀发送到所述脉宽调制(PWM)信号调节模块中，所述脉宽调制(PWM)信号调节模块根据所述两临近有效矢量作用时间T₁、T₂和零矢量的作用时间T₀的大小，采用七段式对所述脉宽调制(PWM)变频器开关进行控制；同时所述脉宽调制(PWM)信号调节模块输出采样点控制信号到所述直流电流采样模块；所述直流电流采样模块将两次采样得到的直流电流采样值发送到所述交流电流重构模块中，所述交流电流重构模块结合所述正弦电压脉宽调制(SVPWM)模块发送的扇区号信号重构出三相交流电流。Wherein, θ' _r is the phase angle of the modulated reference voltage vector V' _r , and the sinusoidal voltage pulse width modulation (SVPWM) overmodulation module uses the two adjacent effective vector action times T ₁ , T ₂ and the phase angle of the zero vector The action time T ₀ is sent to the pulse width modulation (PWM) signal adjustment module, and the pulse width modulation (PWM) signal adjustment module is based on the action time T ₁ , T ₂ of the two adjacent effective vectors and the action time of the zero vector The size of T ₀ , adopts seven-segment type to control the switch of the pulse width modulation (PWM) frequency converter; at the same time, the pulse width modulation (PWM) signal adjustment module outputs the sampling point control signal to the direct current sampling module; The DC current sampling module sends the DC current sampling value obtained by two samplings to the AC current reconstruction module, and the AC current reconstruction module combines the sector number signal sent by the sinusoidal voltage pulse width modulation (SVPWM) module A three-phase alternating current is reconstructed.

所述最大线性调制下的电压利用率η_linesr为The voltage utilization ratio η _linesr under the maximum linear modulation is

${η η}_{linear linear} = = min min ((11,, \frac{22 ((11 - - ρ ρ))}{\sqrt{33}}))$

所述最大线性调制下的参考电压矢量V_linear为The reference voltage vector V _linear under the maximum linear modulation is

${V V}_{linear linear} = = \frac{{u u}_{dc dc}}{\sqrt{33}} {η η}_{linear linear} \cdot &Center Dot; {e e}^{j j {θ θ}_{r r}}$

其中，θ_r为所述最大线性调制下的参考电压矢量V_linear的相角。Wherein, θ _r is the phase angle of the reference voltage vector V _linear under the maximum linear modulation.

所述一般过调制方式下的电压利用率η_edge为The voltage utilization ratio _ηedge under the general overmodulation mode is

所述一般过调制方式下的参考电压矢量V_edge为The reference voltage vector V _edge under the general overmodulation mode is

其中θ_r为所述一般过调制方式下的参考电压矢量V_edge的相角；

为矢量分解图的六边形中心到无效区域的一顶点的矢量与矢量V₁之间的夹角：Wherein θ _r is the phase angle of the reference voltage vector V _edge under the general overmodulation mode;

is the angle between the vector from the center of the hexagon of the vector decomposition diagram to a vertex of the invalid area and the vector V ₁ :

所述最大过调制方式下的电压利用率η_limit为The voltage utilization rate η _limit under the maximum overmodulation mode is

${η η}_{limit limit} = = \frac{22 \sqrt{33}}{π π} [[11 - - ((22 - - \sqrt{33})) ρ ρ]]$

所述最大过调制方式下的参考电压矢量V_limit为The reference voltage vector V _limit under the maximum overmodulation mode is

${V V}_{limit limit} ((θ θ)) = = \{\begin{matrix} ρ ρ \frac{{U u}_{dc dc}}{\sqrt{33}} & 00 \leq \leq {θ θ}_{r r} < < \frac{π π}{66} \\ ((11 - - ρ ρ)) \frac{{U u}_{dc dc}}{\sqrt{33}} & \frac{π π}{66} \leq \leq {θ θ}_{r r} < < \frac{π π}{33} \\ \frac{{U u}_{dc dc}}{\sqrt{33}} & \frac{π π}{33} \leq \leq {θ θ}_{r r} < < \frac{π π}{22} \end{matrix}$

其中，θ_r为所述最大过调制方式下的参考电压矢量V_limit的相角。Wherein, θ _r is the phase angle of the reference voltage vector V _limit in the maximum overmodulation mode.

所述ρ为所述最小采样时间T_min与控制周期T_s的比值：The ρ is the ratio of the minimum sampling time T _min to the control period T _s :

$ρ ρ = = \frac{{T T}_{min min}}{{T T}_{s the s}} . .$

所述无效区域为六边形矢量分解图中六个顶点处的菱形区域，即落在所述无限区域内的参考电压矢量V_r经过分解后，得到两个所述临近的有效矢量中的任一个的作用时间小于所述最小采样时间T_min。The invalid area is a diamond-shaped area at six vertices in the hexagonal vector decomposition diagram, that is, after the reference voltage vector V _r falling in the infinite area is decomposed, any of the two adjacent valid vectors is obtained. The action time of one is less than the minimum sampling time T _min .

本发明由于采取以上技术方案，其具有以下优点：1、本发明采用三相桥式变频器以180°导通六拍方式工作，在一个周期内每相桥臂上、下两开关管各导通半个周期，有效提高变频器直流母线电压利用率。2、本发明采用的电流重构过调制方法使得直流母线电压利用率得到提高，因此可以降低变频装置中的电力电子器件的容量，以及系统成本和装置损耗，具有节能的优点。3、本发明的过调制方法具有输出电压中的基波成分与调制比呈线性，谐波含量低，易于数字化等优点。本发明的调制方法可广泛用于空调和电动汽车中的变频器的电流重构过调制领域。Because the present invention adopts the above technical scheme, it has the following advantages: 1. The present invention adopts a three-phase bridge frequency converter to work in a 180° conduction six-beat mode, and each phase bridge arm upper and lower switch tubes each conduct in one cycle. Through half a cycle, effectively improve the frequency converter DC bus voltage utilization. 2. The current reconstruction overmodulation method adopted in the present invention improves the utilization rate of the DC bus voltage, thereby reducing the capacity of power electronic devices in the frequency conversion device, as well as system cost and device loss, and has the advantage of energy saving. 3. The overmodulation method of the present invention has the advantages that the fundamental wave component in the output voltage is linear to the modulation ratio, the harmonic content is low, and it is easy to digitize. The modulation method of the invention can be widely used in the field of current reconstruction over-modulation of frequency converters in air conditioners and electric vehicles.

附图说明Description of drawings

图1是本发明应用于电流重构过调制方法的永磁同步电机电流重构过调制装置结构示意图Fig. 1 is a structural schematic diagram of a permanent magnet synchronous motor current reconstruction overmodulation device applied to the current reconstruction overmodulation method of the present invention

图2是只使用一个直流侧电流传感器的变频器拓扑结构示意图Figure 2 is a schematic diagram of the inverter topology using only one DC side current sensor

图3是本发明变频器中开关的状态为V₁(100)时的电流流通路径示意图Fig. 3 is the schematic diagram of the current flow path when the state of the switch in the frequency converter of the present invention is V ₁ (100)

图4是本发明变频器中开关的状态为V₀(000)时的电流流通路径示意图Fig. 4 is a schematic diagram of the current flow path when the state of the switch in the frequency converter of the present invention is V ₀ (000)

图5是本发明七段式开关状态分布及直流电流采样点示意图Fig. 5 is a schematic diagram of seven-segment switch state distribution and DC current sampling points of the present invention

图6是本发明当调制比过低时的参考电压矢量分解示意图Figure 6 is a schematic diagram of the vector decomposition of the reference voltage when the modulation ratio is too low in the present invention

图7是本发明接近有效矢量的参考电压矢量分解示意图Fig. 7 is a schematic diagram of the decomposition of the reference voltage vector approaching the effective vector in the present invention

图8是本发明调制比过低时开关状态调整方法示意图Fig. 8 is a schematic diagram of the switching state adjustment method when the modulation ratio is too low in the present invention

图9是本发明低调制比时调整后的开关状态分布示意图Fig. 9 is a schematic diagram of the adjusted switch state distribution when the modulation ratio is low in the present invention

图10是本发明参考电压矢量接近有效矢量时的开关状态调整方法示意图Fig. 10 is a schematic diagram of the switch state adjustment method when the reference voltage vector of the present invention is close to the effective vector

图11是本发明参考电压矢量接近有效矢量时调整后的开关状态分布示意图Fig. 11 is a schematic diagram of the adjusted switch state distribution when the reference voltage vector of the present invention is close to the effective vector

图12是本发明修改开关状态方法无法作用的无效区域示意图Figure 12 is a schematic diagram of an invalid area where the method of modifying the switch state of the present invention cannot work

图13是本发明当T_min较小时的最大线性调制情况下的电压矢量轨迹示意图Fig. 13 is a schematic diagram of the voltage vector trajectory in the case of maximum linear modulation when T _min is small in the present invention

图14是本发明当T_min较大时的最大线性调制情况下的电压矢量轨迹示意图Fig. 14 is a schematic diagram of the voltage vector trajectory in the case of maximum linear modulation when T _min is larger in the present invention

图15是本发明一般过调制方式下的电压矢量轨迹示意图Fig. 15 is a schematic diagram of the voltage vector trajectory under the general overmodulation mode of the present invention

图16是本发明一般过调制方式时电压矢量轨迹及输出电压的时域波形Figure 16 is the time-domain waveform of the voltage vector track and the output voltage during the general overmodulation mode of the present invention

图17是本发明最大过调制方式下的电压矢量示意图Fig. 17 is a schematic diagram of the voltage vector under the maximum overmodulation mode of the present invention

图18是本发明最大过调制方式时电压矢量轨迹及输出电压的时域波形Fig. 18 is the time-domain waveform of the voltage vector track and the output voltage during the maximum overmodulation mode of the present invention

图19本发明采用SVPWM方式下的速度仿真曲线Figure 19 The present invention adopts the speed simulation curve under the SVPWM mode

图20本发明电流重构过调制方式下的速度仿真曲线Fig. 20 The speed simulation curve under the current reconstruction overmodulation mode of the present invention

具体实施方式Detailed ways

下面结合附图和实施例对本发明进行详细的描述。The present invention will be described in detail below in conjunction with the accompanying drawings and embodiments.

如图1所示，本发明的电流重构过调制装置包括电流重构装置1、矢量控制器2、PWM变频器3、直流侧电流传感器4、电机5和编码器6。电机5运行在中高速区，编码器6对电机5进行测速，输出转子位置角θ到矢量控制器2中计算电压矢量，电流重构装置1与矢量控制器2共同对PWM变频器进行闭环控制。其中，电流重构装置1包括SVPWM过调制模块11、PWM信号调节模块12、直流电流采样模块13和交流电流重构模块14。电流重构装置1的作用是重构出三相交流电流作为反馈以实现电流的闭环控制。As shown in FIG. 1 , the current reconstruction overmodulation device of the present invention includes a current reconstruction device 1 , a vector controller 2 , a PWM converter 3 , a DC side current sensor 4 , a motor 5 and an encoder 6 . The motor 5 runs in the medium-high speed area, the encoder 6 measures the speed of the motor 5, and outputs the rotor position angle θ to the vector controller 2 to calculate the voltage vector, and the current reconstruction device 1 and the vector controller 2 jointly perform closed-loop control on the PWM inverter . Wherein, the current reconstruction device 1 includes a SVPWM overmodulation module 11 , a PWM signal adjustment module 12 , a DC current sampling module 13 and an AC current reconstruction module 14 . The function of the current reconstruction device 1 is to reconstruct the three-phase alternating current as feedback to realize the closed-loop control of the current.

首先，直流电流采样模块13在PWM信号调节模块12输入的采样点控制信号的作用下，通过直流侧电流传感器4对PWM变频器3的直流侧进行两次直流电流采样。将两次直流电流采样值i_dc输入到交流电流重构模块14中，交流电流重构模块14结合SVPWM过调制模块11输入的扇区号信号重构出三相交流电流i_a、i_b和i_c，将三相交流电流值i_a、i_b和i_c输入矢量控制器2，结合编码器6输出的转子位置角θ，计算得到参考电压矢量V_r的实部μ_α和虚部μ_β，SVPWM过调制模块11根据V_r＝μ_α+jμ_β计算得到参考电压矢量V_r的值。其中，直流电流采样模块13要在有效矢量作用期间内通过直流侧电流传感器4来对直流电流进行采样，那么该有效矢量必须持续一个最小采样时间T_min。First, the DC current sampling module 13 samples the DC current of the DC side of the PWM inverter 3 twice through the DC side current sensor 4 under the action of the sampling point control signal input by the PWM signal adjustment module 12 . Input the two DC current sampling values i _dc into the AC current reconstruction module 14, and the AC current reconstruction module 14 reconstructs the three-phase AC currents i _a , i _b and i in combination with the sector number signal input by the SVPWM overmodulation module 11 _c , input the three-phase AC current values _ia , _ib and _ic into the vector controller 2, and combine the rotor position angle θ output by the encoder 6 to calculate the real part μ _α and imaginary part μ _β of the reference voltage vector V _r , the SVPWM overmodulation module 11 calculates the value of the reference voltage vector V _r according to V _r =μ _α +j μ _β . Wherein, if the DC current sampling module 13 is to sample the DC current through the DC side current sensor 4 during the effective vector action period, the effective vector must last for a minimum sampling time T _min .

若参考电压矢量V_r经过矢量分解后，得到的临近有效矢量的作用时间不满足最小采样时间要求，则SVPWM过调制模块11对各临近有效矢量的作用时间进行调整，将得到的各临近有效矢量的作用时间T₁、T₂及零矢量T₀作用时间信号输入到PWM信号调节模块12中。PWM信号调节模块12进而控制PWM变频器3中的各个开关状态的持续时间，使各个临近有效矢量的作用时间T₁、T₂满足最小采样时间T_min的要求。同时PWM信号调节模块12输出采样点控制信号到直流电流采样模块13，直流电流采样模块13在采样点控制信号的作用下通过直流侧电流传感器4再一次对PWM变频器3的直流侧电流进行两次采样。交流电流重构模块14利用两次直流电流的采样值i_dc及SVPWM过调制模块11输入的扇区序号信号，根据三相电流和为零这一特点，就可以计算出第三相电流，从而重构出三相交流电流。If the reference voltage vector V _r undergoes vector decomposition, and the action time of the obtained adjacent effective vectors does not meet the minimum sampling time requirement, then the SVPWM overmodulation module 11 adjusts the action time of each adjacent effective vector, and the obtained adjacent effective vectors The action time T ₁ , T ₂ and the zero vector T ₀ action time signal are input to the PWM signal adjustment module 12 . The PWM signal adjustment module 12 further controls the duration of each switching state in the PWM inverter 3 so that the action time T ₁ and T ₂ of each adjacent effective vector meets the requirement of the minimum sampling time T _min . Simultaneously, the PWM signal adjustment module 12 outputs the sampling point control signal to the DC current sampling module 13, and the DC current sampling module 13 double-checks the DC side current of the PWM frequency converter 3 through the DC side current sensor 4 under the action of the sampling point control signal. subsampling. The AC current reconstruction module 14 uses the sampling value i _dc of the two DC currents and the sector number signal input by the SVPWM overmodulation module 11, and according to the characteristic that the sum of the three-phase currents is zero, the third phase current can be calculated, thereby A three-phase alternating current is reconstructed.

下面对SVPWM过调制模块11对不满足最小采样时间T_min要求的各个临近有效矢量进行调整的方法作具体说明。如图2所示，当三相桥式PWM变频器3采用180°导电制工作模式时，共有8种开关状态，其中V₁、V₂、V₃、V₄、V₅和V₆为有效矢量，V₁代表开关状态为100，即第一相桥臂的上半部分的开关导通，第2、第3相桥臂的下半部分的开关导通，V₂代表开关状态为000，即全部三相桥臂的下半部分开关导通，其他开关状态以此类推。有效矢量作用时间分别为T₁、T₂、T₃、T₄、T₅和T₆，其中V₀和V₇为零矢量，其作用时间分别为T₀、T₇。参考电压矢量V_r在空间矢量图中进行分解后得到的临近有效矢量可以为V₁、V₂、V₃、V₄、V₅或V₆。每一种有效矢量对应一种不同的开关状态，将两个临近有效矢量的作用时间发送给PWM变频器3，以控制其内部6个开关的状态。因为在不同的开关状态作用下，电流的流通路径是确定的，因而直流侧电流和交流侧电流存在确定的关系。The method for adjusting the adjacent effective vectors that do not meet the requirement of the minimum sampling time T _min by the SVPWM overmodulation module 11 will be described in detail below. As shown in Figure 2, when the three-phase bridge PWM inverter 3 adopts the 180° conduction mode, there are 8 switching states in total, among which V ₁ , V ₂ , V ₃ , V ₄ , V ₅ and V ₆ are valid Vector, V ₁ means that the switch state is 100, that is, the switch in the upper half of the first phase bridge arm is turned on, and the switches in the lower half of the second and third phase bridge arms are turned on, V ₂ means that the switch state is 000, That is, the switches in the lower half of all three-phase bridge arms are turned on, and the state of other switches can be deduced by analogy. The effective vector action times are T ₁ , T ₂ , T ₃ , T ₄ , T ₅ and T ₆ respectively, among which V ₀ and V ₇ are zero vectors, and their action times are T ₀ and T ₇ respectively. The adjacent effective vector obtained after decomposing the reference voltage vector V _r in the space vector diagram can be V ₁ , V ₂ , V ₃ , V ₄ , V ₅ or V ₆ . Each effective vector corresponds to a different switch state, and the action time of two adjacent effective vectors is sent to the PWM converter 3 to control the states of the six internal switches. Because under different switch states, the current flow path is definite, so there is a definite relationship between the DC side current and the AC side current.

如图3所示，当开关状态为V₁(100)时，此时直流侧电流采样值i_dc等于电机5的a相电流i_a，直流侧电流传感器4通过检测直流侧电流值采样值i_dc就可以获得a相电流值。如图4所示，当开关状态为V₀(000)时直流侧电流在电机5内部流通，此时直流侧电流相当于为0。逐一分析可得，PWM变频器3的8种开关状态下直流电流采样模块13检测到的直流侧电流采样值i_dc与交流侧电流关系如表1所示。通过表1即可得知在不同的开关状态下直流侧电流采样值i_dc为三相交流电中的哪一相交流电流。As shown in Figure 3, when the switch state is V ₁ (100), the DC side current sampling value i _dc is equal to the a-phase current i _a of the motor 5, and the DC side current sensor 4 detects the DC side current value sampling value i _dc can get a phase current value. As shown in FIG. 4 , when the switch state is V ₀ (000), the DC side current flows inside the motor 5 , and the DC side current is equivalent to 0 at this time. By analyzing one by one, the relationship between the DC side current sampling value i _dc detected by the DC current sampling module 13 and the AC side current in the eight switching states of the PWM inverter 3 is shown in Table 1. From Table 1, it can be known which phase of the three-phase alternating current the DC side current sampling value i _dc is in different switching states.

表1直流侧电流采样值i_dc与交流电流关系Table 1 Relationship between DC side current sampling value i _dc and AC current

电压矢量Voltage vector 开关状态 switch status 直流侧电流采样值i_dc DC side current sampling value i _dc V₀ V ₀ 000000 00 V₁ V ₁ 100100 i_a i _a V₂ V ₂ 110110 -i_c -i _c V₃ V ₃ 010010 -i_b -i _b V₄ V ₄ 011011 i_a i _a V₅ V ₅ 001001 -i_c -i _c

V₆ V ₆ 101101 -i_b -i _b V₇ V ₇ 111111 00

当参考电压矢量V_r分解为实部μ_α和虚部μ_β后，在空间矢量图中得到两个临近有效矢量，即两种不同的开关状态。当参考电压矢量V_r分布在空间矢量图中的不同扇区时，一个控制周期T_s内的两次直流电流采样值i_dc与交流侧电流的关系是不同的。记一个控制周期T_s内的前一个直流电流采样值为i_dc1，后一个直流电流采样值为i_dc2，则各扇区下三相交流电流值与两次直流电流采样值的关系如表2所示。通过表2在交流电流重构模块14中，只要知道同一控制周期T_s内的两次直流电流的采样值和参考电压矢量V_r所处的扇区，就可重构出三相交流电流。When the reference voltage vector V _r is decomposed into the real part μ _α and the imaginary part μ _β , two adjacent effective vectors are obtained in the space vector diagram, that is, two different switching states. When the reference voltage vector V _r is distributed in different sectors of the space vector diagram, the relationship between the two DC current sampling values i _dc and the AC side current in a control period T _s is different. Note that the previous DC current sampling value is i _dc1 and the next DC current sampling value is i _dc2 within a control cycle T _s , then the relationship between the three-phase AC current value and the two DC current sampling values in each sector is shown in Table 2 shown. According to Table 2, in the AC current reconstruction module 14, the three-phase AC current can be reconstructed as long as the sampling values of the two DC currents in the same control period T _s and the sector where the reference voltage vector V _r is located are known.

表2三相交流电流值与两次直流电流采样值关系Table 2 The relationship between the three-phase AC current value and the two DC current sampling values

扇区号sector number i_a i _a i_b i _b i_c i _c 11 i_dc2 i _dc2 i_dc1-i_dc2 i _dc1 -i _dc2 -i_dc1 -i _dc1 22 i_dc1-i_dc2 i _dc1 -i _dc2 i_dc2 i _dc2 -i_dc1 -i _dc1 33 -i_dc1 -i _dc1 i_dc2 i _dc2 i_dc1-i_dc2 i _dc1 -i _dc2 44 -i_dc1 -i _dc1 i_dc1-i_dc2 i _dc1 -i _dc2 i_dc2 i _dc2 55 i_dc1-i_dc2 i _dc1 -i _dc2 -i_dc1 -i _dc1 i_dc2 i _dc2 66 i_dc2 i _dc2 -i_dc1 -i _dc1 i_dc1-i_dc2 i _dc1 -i _dc2

在SVPWM过调制模块11中计算调制比M，其中调制比M为参考电压矢量幅值与空间矢量图的六边形内切圆半径的比值，即Calculate the modulation ratio M in the SVPWM overmodulation module 11, wherein the modulation ratio M is the ratio of the hexagonal inscribed circle radius of the reference voltage vector magnitude and the space vector diagram, that is

其中U_dc为直流侧母线电压的值。Among them, U _dc is the value of the DC side bus voltage.

如图5所示，参考电压矢量V_r经矢量分解后，所得的两个临近的有效矢量的作用时间若满足最小采样时间T_min要求时，此时PWM信号调节模块12不需要对PWM变频器3中的开关状态进行修改，采用现有的七段式矢量开关状态即可采样。此时采样点P的位置可以在控制周期T_s的后半部分的有效矢量V₁、V₂上，也可以在控制周期T_s的前半部分的有效矢量V₁、V₂上，本发明对此不作要求。As shown in Figure 5, after the reference voltage vector V _r is vector decomposed, if the action time of the two adjacent effective vectors obtained meets the minimum sampling time T _min requirement, the PWM signal adjustment module 12 does not need to control the PWM frequency converter at this time. The switch state in 3 is modified, and the existing seven-segment vector switch state can be used for sampling. At this time, the position of the sampling point P can be on the effective vectors V ₁ and V ₂ in the second half of the control period T _s , or on the effective vectors V ₁ and V ₂ in the first half of the control period T _s . This is not required.

当临近的有效矢量的作用时间不满足最小采样时间T_min的要求时，具体又可以分为两种情况：如图6所示，当调制比M过低时，即当参考电压矢量V_r在六边形阴影内时，两个临近有效矢量的作用时间T₁、T₂都小于最小采样时间T_min，则在两个临近的有效矢量作用期间均无法完成对交流电流的采样。如图7所示，当参考电压矢量V_r接近六个有效矢量中的某一个时，此时其中一个临近的有效矢量的作用时间大于T_min，在那个临近的有效矢量的作用时间内可检测到一相交流电流；而另一个临近的有效矢量的作用时间小于T_min，在那个临近的有效矢量的作用时间内无法检测到直流电流；由于只能获得一相电流，因此无法重构出三相交流电流。When the action time of the adjacent effective vector does not meet the minimum sampling time T _min , it can be divided into two cases: as shown in Figure 6, when the modulation ratio M is too low, that is, when the reference voltage vector V _r is at When within the hexagonal shadow, the action time T ₁ and T ₂ of the two adjacent effective vectors are both less than the minimum sampling time T _min , and the AC current cannot be sampled during the action period of the two adjacent effective vectors. As shown in Figure 7, when the reference voltage vector V _r is close to one of the six effective vectors, the action time of one of the adjacent effective vectors is greater than T _min , and it can be detected within the action time of the adjacent effective vector and the action time of another adjacent effective vector is less than T _min , and the DC current cannot be detected during the action time of that adjacent effective vector; since only one phase current can be obtained, three phases cannot be reconstructed phase alternating current.

为了保证在PWM变频器3的直流侧采样到两相直流电流，可以通过现有技术中的修改开关状态方法来确保在每个控制周期T_s内，都可以通过直流侧电流传感器4检测到直流电流值i_dc，进而重构出三相电流。In order to ensure that the two-phase DC current is sampled on the DC side of the PWM inverter 3, the method of modifying the switch state in the prior art can be used to ensure that the DC can be detected by the DC side current sensor 4 in each control cycle T _s The current value i _dc is used to reconstruct the three-phase current.

以临近的有效矢量V₁、V₂在第I扇区为例对修改开关状态法进行说明，其他扇区情况与之类似。记T₁，T₂，T₀分别为利用SVPWM算法计算得到的有效矢量V₁，V₂和零矢量V₀的作用时间，即T_s＝T₁+T₂+T₀。由于有效矢量V₁、V₂在第I扇区内，因此SVPWM过调制模块11将扇区号I发送到交流电流重构模块14中。若有效矢量V₁、V₂的作用时间T₁，T₂不满足最小采样时间T_min的要求；即调制比M过低或者参考电压矢量V_r接近有效矢量V₁，V₂，此时SVPWM过调制模块11对有有效矢量的作用时间T₁，T₂进行修改。具体修改方法如下：Take the adjacent effective vectors V ₁ and V ₂ in sector I as an example to illustrate the method of modifying the switch state, and the situation in other sectors is similar. Note that T ₁ , T ₂ , T ₀ are the action time of effective vector V ₁ , V ₂ and zero vector V ₀ calculated by using SVPWM algorithm respectively, that is, T _s =T ₁ +T ₂ +T ₀ . Since the effective vectors V ₁ and V ₂ are in sector I, the SVPWM overmodulation module 11 sends the sector number I to the AC current reconstruction module 14 . If the action time T ₁ and T ₂ of the effective vectors V ₁ and V ₂ do not meet the requirements of the minimum sampling time T _min ; that is, the modulation ratio M is too low or the reference voltage vector V _r is close to the effective vectors V ₁ and V ₂ , then SVPWM The overmodulation module 11 modifies the action times T ₁ , T ₂ with active vectors. The specific modification method is as follows:

①如图8所示，当调制比M过低时，以T₁＜T_min，T₂＜T_min为例来说明对于有效矢量V₁、V₂的作用时间T₁，T₂的修改：由于作用时间T₁，T₂均小于T_min，无法满足采样要求，因此令ΔT₁＝T_min-T₁，ΔT₂＝T_min-T₂。首先将作用时间T₁，T₂修改为T′₁＝ΔT₁+T₁＝T_min，T′₂＝ΔT₂+T₂＝T_min，以保证采样的最短时间要求；为了维持参考电压矢量V_r不变，在有效矢量V₁、V₂的反方向分别加入一个补偿矢量V₄、V₅，其作用时间为T₄＝ΔT₁，T₅＝ΔT₂，因此得到各个矢量的持续时间归纳如下：① As shown in Figure 8, when the modulation ratio M is too low, take T ₁ <T _min , T ₂ <T _min as an example to illustrate the modification of the action time T ₁ and T ₂ of the effective vectors V ₁ and V ₂ : Since the action time T ₁ and T ₂ are both less than T _min , which cannot meet the sampling requirements, ΔT ₁ =T _min -T ₁ , ΔT ₂ =T _min -T ₂ . First, modify the action time T ₁ and T ₂ to T′ ₁ =ΔT ₁ +T ₁ =T _min , T′ ₂ =ΔT ₂ +T ₂ =T _min to ensure the minimum sampling time requirement; in order to maintain the reference voltage vector V _r remains unchanged, and a compensation vector V ₄ , V ₅ is respectively added to the opposite direction of the effective vector V ₁ , V ₂ , and its action time is T ₄ = ΔT ₁ , T ₅ = ΔT ₂ , so the duration of each vector is obtained Summarized as follows:

$\{\begin{matrix} {T T}_{11}^{' '} = = {T T}_{11} + + Δ Δ {T T}_{11} \\ {T T}_{22}^{' '} = = {T T}_{11} + + Δ Δ {T T}_{22} \\ {T T}_{44} = = Δ Δ {T T}_{11} \\ {T T}_{55} = = Δ Δ {T T}_{22} \\ {T T}_{00}^{' '} = = {T T}_{s the s} - - {T T}_{11}^{' '} - - {T T}_{22}^{' '} - - {T T}_{44} - - {T T}_{55} \end{matrix} - - - - - - ((22))$

如图9所示，此时PWM信号调节模块12根据方程(2)输出PWM信号控制PWM变频器3的各个开关状态，采样点控制信号控制直流电流采样模块3通过电流传感器4对两相交流电流进行采样。当开关状态分别为有效矢量V₁、V₂时进行采样，根据表1，可知检测到的电流为i_a和-i_c，此时采样点P的位置可以在控制周期T_s的后半部分的有效矢量V₁、V₂上。As shown in Figure 9, at this time, the PWM signal adjustment module 12 outputs the PWM signal according to the equation (2) to control each switch state of the PWM frequency converter 3, and the sampling point control signal controls the DC current sampling module 3 to pass the current sensor 4 to the two-phase AC current Take a sample. Sampling is performed when the switching states are the effective vectors V ₁ and V ₂ respectively. According to Table 1, it can be known that the detected currents are i _a and -i _c . At this time, the position of the sampling point P can be in the second half of the control period T _s on the effective vectors V ₁ and V ₂ .

②当参考电压矢量V_r接近有效矢量V₁、V₂时可以分为两种情况：第一种情况是参考电压矢量V_r接近有效矢量V₁，第二种情况为参考电压矢量V_r接近有效矢量V₂，这两种情况类似。下面以参考电压矢量V_r接近有效矢量V₂为例进行说明，这种情况用数学关系式表示为：②When the reference voltage vector V _r is close to the effective vector V ₁ and V ₂ , it can be divided into two cases: the first case is that the reference voltage vector V _r is close to the effective vector V ₁ , and the second case is that the reference voltage vector V _r is close to Effective vector V ₂ , the two cases are similar. Let's take the reference voltage vector V _r close to the effective vector V ₂ as an example to illustrate, this situation is expressed as:

T₁/2＜T_min，T₂/2≥T_min (3)T ₁ /2<T _min , T ₂ /2≥T _min (3)

此时又可以细分为以下两种情况：T₁＜T_min，T₂≥2T_min和T_min≤T₁＜2T_min，T₂≥2T_min。如图10所示，当T₁＜T_min，T₂≥2T_min时，此时有效矢量V₁的作用时间T₁需要修改，而有效矢量V₂的作用时间T₂不需要修改。令ΔT₁＝T_min-T₁，首先将作用时间T₁修改为T′₁＝ΔT₁+T₁＝T_min，同时将作用时间T₂减小为T′₂＝T₂-ΔT₁＝T₂+T₁-T_min。为了维持修改后参考电压矢量V_r的大小和幅值不变，加入补偿矢量V₃，作用时间为T₃＝ΔT₁。各矢量作用时间如下式所示：At this time, it can be subdivided into the following two situations: T ₁ <T _min , T ₂ ≥2T _min and T _min ≤T ₁ <2T _min , T ₂ ≥2T _min . As shown in Figure 10, when T ₁ <T _min and T ₂ ≥ 2T _min , the action time T ₁ of the effective vector V ₁ needs to be modified, while the action time T ₂ of the effective vector V ₂ does not need to be modified. Set ΔT ₁ =T _min -T ₁ , first modify the action time T ₁ to T′ ₁ =ΔT ₁ +T ₁ =T _min , and at the same time reduce the action time T ₂ to T′ ₂ =T ₂ -ΔT ₁ = T ₂ +T ₁ -T _min . In order to maintain the size and amplitude of the modified reference voltage vector V _r unchanged, a compensation vector V ₃ is added, and the action time is T ₃ =ΔT ₁ . The action time of each vector is shown in the following formula:

$\{\begin{matrix} {T T}_{11}^{' '} = = {T T}_{11} + + Δ Δ {T T}_{11} \\ {T T}_{22}^{' '} = = {T T}_{22} - - Δ Δ {T T}_{11} \\ {T T}_{33} = = Δ Δ {T T}_{11} \\ {T T}_{00}^{' '} = = {T T}_{s the s} - - {T T}_{11}^{' '} - - {T T}_{22}^{' '} - - {T T}_{33} \end{matrix} - - - - - - ((44))$

如图11所示，此时PWM信号调节模块12根据方程(4)输出PWM信号控制PWM变频器3的各个开关状态，采样点控制信号控制直流电流采样模块3通过电流传感器4对两相交流电流进行采样。当开关状态分别为有效矢量V₁、V₂时进行采样，检测到电流i_a和-i_c。此时采样点P的位置可以在控制周期T_s的后半部分的有效矢量V₁、V₂上。As shown in Figure 11, at this time, the PWM signal adjustment module 12 outputs the PWM signal according to the equation (4) to control each switch state of the PWM frequency converter 3, and the sampling point control signal controls the DC current sampling module 3 to pass the current sensor 4 to the two-phase AC current Take a sample. Sampling is carried out when the switching states are the effective vectors V ₁ and V ₂ respectively, and the currents _ia and _-ic are detected. At this time, the position of the sampling point P may be on the effective vectors V ₁ and V ₂ in the second half of the control period T _s .

在上述修改开关状态的方法中，修改后所有非零矢量的作用时间比修改前有所增加。这些增加的作用时间需要通过减少零矢量V₀的作用时间T₀来获得。在过调制情况下，随着调制比M上升，零矢量V₀的作用时间T₀越来越短从而无法满足修改参考电压矢量V_r的要求。如图12所示，当参考电压矢量V_r落在空间矢量图中的阴影部分时，参考电压矢量V_r经过分解后，得到两个临近的有效矢量。其中一个临近的有效矢量的作用时间小于最小采样时间T_min，此时无法在保持参考电压矢量V_r不变的情况下通过修改开关状态分布来确保采样要求，因此将阴影部分称之为无效区域，无效区域的作用时间是最小采样时间T_min，此时也可以看作是如图7所示的特殊情况。六边形内除无效区域外的非阴影区域称之为有效区域，在无效区域中SVPWM过调制模块11使用本发明的电流重构的过调制方法对参考电压矢量V_r进行调整。In the above method of modifying the switch state, the action time of all non-zero vectors after modification is increased compared with that before modification. These increased action times need to be obtained by reducing the action time T ₀ of the zero vector V ₀ . In the case of over-modulation, as the modulation ratio M increases, the action time T ₀ of the zero vector V ₀ becomes shorter and shorter, which cannot meet the requirement of modifying the reference voltage vector V _r . As shown in Figure 12, when the reference voltage vector V _r falls in the shaded part of the space vector diagram, the reference voltage vector V _r is decomposed to obtain two adjacent effective vectors. The action time of one of the adjacent effective vectors is less than the minimum sampling time T _min . At this time, it is impossible to ensure the sampling requirements by modifying the switch state distribution while keeping the reference voltage vector V _r unchanged, so the shaded part is called an invalid area , the action time of the invalid area is the minimum sampling time T _min , which can also be regarded as a special case as shown in FIG. 7 . The non-shaded area in the hexagon except the invalid area is called the effective area. In the invalid area, the SVPWM overmodulation module 11 uses the current reconstruction overmodulation method of the present invention to adjust the reference voltage vector V _r .

在本发明的电流重构的过调制方法中，首先定义电压利用率η为PWM变频器3输出的线电压基波幅值与直流母线电压的比值，即：In the overmodulation method of the current reconstruction of the present invention, first define the voltage utilization rate η as the ratio of the line voltage fundamental wave amplitude and the DC bus voltage output by the PWM frequency converter 3, that is:

$η η = = \frac{{U u}_{11__fd fd}}{{U u}_{dc dc}} - - - - - - ((55))$

其中U_{1_fd}为线电压基波幅值，U_dc为直流母线电压。定义最小采样时间T_min与控制周期T_s的比值为ρ，即：Among them, U _{1_fd} is the amplitude of the fundamental wave of the line voltage, and U _dc is the DC bus voltage. Define the ratio of the minimum sampling time T _min to the control period T _s as ρ, namely:

$ρ ρ = = \frac{{T T}_{min min}}{{T T}_{s the s}} - - - - - - ((66))$

以第I扇区为例，有效矢量V₁、V₂与其对应的作用时间T₁、T₂之间的关系如下：Taking sector I as an example, the relationship between effective vectors V ₁ , V ₂ and their corresponding action times T ₁ , T ₂ is as follows:

T_s＝T₁+T₂+T₀ (7)T _s =T ₁ +T ₂ +T ₀ (7)

T₁V₁+T₂V₂+T₀V₀＝T_sV_r (8)T ₁ V ₁ +T ₂ V ₂ +T ₀ V ₀ = T _s V _r (8)

其中，零矢量V₀的作用时间T₀；T_s为采样周期；V_r为参考电压矢量。Among them, the action time T ₀ of the zero vector V ₀ ; T _s is the sampling period; V _r is the reference voltage vector.

在SVPWM算法中，基本电压矢量作用时间为：In the SVPWM algorithm, the basic voltage vector action time is:

$\{\begin{matrix} \frac{{T T}_{11}}{{T T}_{s the s}} = = \frac{\sqrt{33} | | {V V}_{r r} | |}{{U u}_{dc dc}} sin sin ((π π / / 33 - - {θ θ}_{r r})) \\ \frac{{T T}_{22}}{{T T}_{s the s}} = = \frac{\sqrt{33} | | {V V}_{r r} | |}{{U u}_{dc dc}} sin sin (({θ θ}_{r r})) \\ {T T}_{00} = = {T T}_{s the s} - - {T T}_{11} - - {T T}_{22} \end{matrix} - - - - - - ((99))$

当有效矢量V₁、V₂处在第I扇区时，根据三相桥式PWM变频器3各相电压的特点可以计算得出：When the effective vectors V ₁ and V ₂ are in sector I, according to the characteristics of each phase voltage of the three-phase bridge PWM inverter 3, it can be calculated as follows:

$V_{1} = \frac{2}{3} U_{dc} e^{j 0},$

V_{1} = \frac{2}{3} u_{dc} e^{j} 0,

为了避免参考电压矢量V_r出现在无效区域，将参考电压矢量V_r分为如下三种临界情况：In order to avoid the reference voltage vector V _r appearing in the invalid region, the reference voltage vector V _r is divided into the following three critical situations:

a)最大线性调制。如图13所示，θ_r为参考电压矢量的相角，由于PWM变频器3所采用的电子器件不同，使得最小采样时间T_min的大小不同。a) Maximum linear modulation. As shown in FIG. 13 , θ _r is the phase angle of the reference voltage vector, and the minimum sampling time T _min is different due to the different electronic devices used in the PWM inverter 3 .

当最小采样时间T_min较小时，阴影处的无效区域处在空间矢量图的六边形内切圆之外，最大线性调制轨迹即为六边形内切圆。则最大可能输出的参考电压矢量 $V_{r} = 1 / \sqrt{3} U_{dc},$ 则线电压基波幅值 $U_{1_fd} = U_{dc} = \sqrt{3} V_{r} .$ 因此最大线性调制下的电压利用率为η_linear＝1。When the minimum sampling time T _min is small, the invalid area in the shadow is outside the hexagonal inscribed circle of the space vector diagram, and the maximum linear modulation trajectory is the hexagonal inscribed circle. Then the maximum possible output reference voltage vector $V_{r} = 1 / \sqrt{3} u_{dc},$ The amplitude of the fundamental wave of the line voltage $u$ $_{1_fd} = u_{dc} = \sqrt{3} V_{r} .$ Therefore, the voltage utilization ratio under the maximum linear modulation is η _linear =1.

如图14所示，当最小采样时间T_min较大时，阴影部分的无效区域进入六边形内切圆内部，最大线性调制轨迹为与无效区域正好相接的圆。由于不需要零矢量V₀进行调整，因此零矢量V₀的作用时间T₀＝0，当参考矢量V_r与有效矢量V₁重合时：V₁T₁＝V_rT_s，即：As shown in Fig. 14, when the minimum sampling time T _min is large, the invalid area in the shaded part enters the interior of the hexagonal inscribed circle, and the maximum linear modulation trajectory is a circle that just touches the invalid area. Since the zero vector V ₀ is not required to be adjusted, the action time T ₀ of the zero vector V ₀ =0, when the reference vector V _r coincides with the effective vector V ₁ : V ₁ T ₁ =V _r T _s , that is:

$\frac{22}{33} {U u}_{dc dc} ((11 - - ρ ρ)) = = | | {V V}_{r r} | | {e e}^{j j {θ θ}_{r r}} - - - - - - ((1111))$

则最大参考矢量V_r的幅值为：

此时对应的线电压基波幅值

U_{1_fd} = \frac{2}{\sqrt{3}} U_{dc} (1 - ρ),

根据方程(5)可知，电压利用率

η = \frac{U_{1_fd}}{U_{dc}} = \frac{2}{\sqrt{3}} (1 - ρ) .

Then the magnitude of the maximum reference vector V _r is:

The corresponding line voltage fundamental wave amplitude at this time

u

_{1_fd} = \frac{2}{\sqrt{3}} u_{dc} (1 - ρ),

According to equation (5), we can see that the voltage utilization

η = \frac{u_{1_fd}}{u_{dc}} = \frac{2}{\sqrt{3}} (1 - ρ) .

将最小采样时间T_min的两种情况下电压利用率统一表示为：The voltage utilization rate in the two cases of the minimum sampling time T _min is uniformly expressed as:

${η η}_{linear linear} = = min min ((11,, \frac{22 ((11 - - ρ ρ))}{\sqrt{33}})) - - - - - - ((1212))$

若用V_linear表示最大线性调制下的临界参考电压矢量，则由线电压基波幅值U_{1_fd}是V_linear的相电压的

倍的关系可知：If V _linear is used to represent the critical reference voltage vector under the maximum linear modulation, then the line voltage fundamental wave amplitude U _{1_fd} is the phase voltage of V _linear

The doubling relationship shows that:

${V V}_{linear linear} = = \frac{{u u}_{dc dc}}{\sqrt{33}} {η η}_{linear linear} \cdot \cdot {e e}^{j j {θ θ}_{r r}} - - - - - - ((1313))$

其中θ_r为临界参考电压矢量V_linear的相角。Where θ _r is the phase angle of the critical reference voltage vector V _linear .

b)一般过调制方式。如图15所示，图中加粗线表示一般过调制方式下参考电压矢量V_r的轨迹为有效区域的边界，V_edge、η_edge分别表示一般过调制方式下的临界参考电压矢量和电压利用率。以第I扇区为例，零矢量V₀的作用时间T₀＝0，按参考电压矢量V_r的相角θ_r大小，分为下列三种情况进行推导：b) General overmodulation mode. As shown in Figure 15, the bold line in the figure indicates that the trajectory of the reference voltage vector V _r in the general over-modulation mode is the boundary of the effective area, and V _edge and η _edge represent the critical reference voltage vector and voltage utilization in the general over-modulation mode, respectively. Rate. Taking sector I as an example, the action time T ₀ =0 of the zero vector V ₀ is divided into the following three situations for derivation according to the phase angle θ _r of the reference voltage vector V _r :

i)当

时，此时相邻的有效电压矢量的作用时间T₁＞T_min，T₂＜T_min，则需要将T₂调整为T_min，其中

为六边形中心到无效区域的两边顶点的幅值与矢量V₁之间的夹角。根据方程(8)，则有i) when

, when the action time of the adjacent effective voltage vector is T ₁ >T _min , T ₂ <T _min , then T ₂ needs to be adjusted to T _min , where

is the angle between the amplitude of the hexagon center and the vertices on both sides of the invalid area and the vector _V1 . According to equation (8), there are

V₁(T_s-T_min)+V₂T_min＝V_edgeT_s V ₁ (T _s -T _min )+V ₂ T _min ＝V _edge T _s

进而可知：V₁(1-ρ)+V₂ρ＝V_edge (14)Furthermore, it can be seen that: V ₁ (1-ρ)+V ₂ ρ=V _edge (14)

其中ρ为最小采样时间T_min与控制周期T_s的比值。将方程(10)带入方程(14)中，得：Among them, ρ is the ratio of the minimum sampling time T _min to the control period T _s . Substituting Equation (10) into Equation (14), we get:

$\frac{22}{33} {U u}_{dc dc} ((11 - - ρ ρ)) + + \frac{22}{33} {U u}_{dc dc} ρ ρ ((\frac{11}{22} + + j j \frac{\sqrt{33}}{22})) = = | | {V V}_{edge edge} | | {e e}^{j j {θ θ}_{r r}} = = | | {V V}_{edge edge} | | (({cos cos θ θ}_{r r} + + j j sin sin {θ θ}_{r r}))$

$\frac{22}{33} {U u}_{dc dc} ((11 - - ρ ρ)) + + \frac{22}{33} {U u}_{dc dc} \frac{sin sin {θ θ}_{r r}}{sin sin ((π π / / 33 - - {θ θ}_{r r}))} ((11 - - ρ ρ)) ((\frac{11}{22} + + j j \frac{\sqrt{33}}{22})) = = | | {V V}_{edge edge} | | {e e}^{j j {θ θ}_{r r}}$

$\frac{22}{33} {U u}_{dc dc} ((11 - - ρ ρ)) \frac{11}{sin sin ((π π / / 33 - - {θ θ}_{r r}))} [[((sin sin ((π π / / 33 - - {θ θ}_{r r})) + + \frac{11}{22} sin sin {θ θ}_{r r} + + j j \frac{\sqrt{33}}{22} sin sin {θ θ}_{r r}))]] = = | | {V V}_{edge edge} | | {e e}^{j j {θ θ}_{r r}}$

$\frac{11}{\sqrt{33}} {U u}_{dc dc} ((11 - - ρ ρ)) \frac{11}{sin sin ((π π / / 33 - - {θ θ}_{r r}))} {e e}^{j j {θ θ}_{r r}} = = | | {V V}_{edge edge} | | {e e}^{j j {θ θ}_{r r}}$

根据以上推导结果，可知：According to the above derivation results, it can be known that:

$| | {V V}_{edge edge} | | = = \frac{11}{\sqrt{33}} {U u}_{dc dc} ((11 - - ρ ρ)) \frac{11}{sin sin ((π π / / 33 - - {θ θ}_{r r}))} - - - - - - ((1515))$

ii)当

时，将T₀＝T_s-T₁-T₂＝0带入方程(9)和方程(15)中，得：ii) when

, taking T ₀ =T _s -T ₁ -T ₂ =0 into Equation (9) and Equation (15), we get:

$| | {V V}_{edge edge} | | = = \frac{11}{\sqrt{33}} \frac{{U u}_{dc dc}}{cos cos ((π π / / 66 - - {θ θ}_{r r}))} - - - - - - ((1616))$

iii)当

时，此时T₂＞T_min，T₁＜T_min，则需要将T₁调整为T_min。结合方程(8)，则有V₁T_min+V₂(T_s-T_min)＝V_edgeT_s，将方程(6)代入可知，iii) when

, when T ₂ >T _min , T ₁ <T _min , then T ₁ needs to be adjusted to T _min . Combined with Equation (8), V ₁ T _min +V ₂ (T _s -T _min )=V _edge T _s , substituting Equation (6) into

V₁ρ+V₂(1-ρ)＝V_edge (17)V ₁ ρ+V ₂ (1-ρ)=V _edge (17)

将方程(10)带入方程(17)，得Substituting Equation (10) into Equation (17), we get

$\frac{22}{33} {U u}_{dc dc} ρ ρ + + \frac{22}{33} {U u}_{dc dc} ((11 - - ρ ρ)) ((\frac{11}{22} + + j j \frac{\sqrt{33}}{22})) = = | | {V V}_{edge edge} | | {e e}^{j j {θ θ}_{r r}}$

$\frac{22}{33} {U u}_{dc dc} \frac{sin sin ((π π / / 33 - - {θ θ}_{r r}))}{sin sin {θ θ}_{r r}} ((11 - - ρ ρ)) + + \frac{22}{33} {U u}_{dc dc} ((11 - - ρ ρ)) ((\frac{11}{22} + + j j \frac{\sqrt{33}}{22})) = = | | {V V}_{edge edge} | | {e e}^{j j {θ θ}_{r r}}$

$\frac{22}{33} {U u}_{dc dc} ((11 - - ρ ρ)) \frac{11}{sin sin (({θ θ}_{r r}))} [[((sin sin ((π π / / 33 - - {θ θ}_{r r})) + + \frac{11}{22} sin sin {θ θ}_{r r} + + j j \frac{\sqrt{33}}{22} sin sin {θ θ}_{r r}))]] = = | | {V V}_{edge edge} | | {e e}^{j j {θ θ}_{r r}}$

$\frac{11}{\sqrt{33}} {U u}_{dc dc} ((11 - - ρ ρ)) \frac{11}{sin sin (({θ θ}_{r r}))} {e e}^{j j {θ θ}_{r r}} = = | | {V V}_{edge edge} | | {e e}^{j j {θ θ}_{r r}}$

$| | {V V}_{edge edge} | | = = \frac{11}{\sqrt{33}} {U u}_{dc dc} ((11 - - ρ ρ)) \frac{11}{sin sin (({θ θ}_{r r}))} - - - - - - ((1818))$

综合以上三种情况可得第I扇区的临界参考电压矢量V_edge为：Combining the above three situations, the critical reference voltage vector V _edge of sector I can be obtained as:

其中

为矢量分解图的六边形中心到无效区域的一顶点的矢量与矢量V₁之间的夹角。夹角的推导过程如下：in

is the angle between the vector from the hexagon center of the vector decomposition diagram to a vertex of the invalid area and the vector V ₁ . Angle The derivation process is as follows:

以第I扇区

为例进行推导 Sector I

Derivation as an example

由正弦定理：

By the law of sines:

(20)+(21)得：

(20)+(21) get:

则

but

如图16所示，若将空间矢量图第I扇区的参考电压矢量V_r的轨迹按时间变化映射到β轴，得到临界参考电压矢量V_edge在时域的波形。图左边的参考电压矢量V_r的β轴分量V_rsinθ_r为图右边的临界参考电压矢量V_edge在时域的形式。即临界参考电压矢量V_edge在[0，π/2]表达式如下：As shown in Figure 16, if the trajectory of the reference voltage vector V _r in sector I of the space vector diagram is mapped to the β axis according to the time change, the waveform of the critical reference voltage vector V _edge in the time domain is obtained. The β-axis component V _r sinθ _r of the reference voltage vector V _r on the left of the figure is the form of the critical reference voltage vector V _edge on the right of the figure in the time domain. That is, the expression of the critical reference voltage vector V _edge in [0, π/2] is as follows:

此时一般过调制方式下的电压利用率 $η_{edge} = \frac{\sqrt{3} V_{edge}}{U_{dc}},$ 即At this time, the voltage utilization rate under the general overmodulation mode $η_{edge} = \frac{\sqrt{3} V_{edge}}{u_{dc}},$ Right now

c)最大过调制方式。如图17所示，在最大过调制方式下，参考电压矢量V_r为图中以空间矢量图中心为起点呈放射状的12个电压矢量，由带箭头的粗线表示。用V_limit、η_limit分别表示最大过调制方式下，临界参考电压矢量和电压利用率。按参考电压矢量V_r的相角θ_r大小，分为下列三种情况进行推导：c) Maximum overmodulation mode. As shown in Figure 17, in the maximum overmodulation mode, the reference voltage vector V _r is 12 voltage vectors radially starting from the center of the space vector diagram in the figure, represented by thick lines with arrows. Use V _limit and η _limit to denote the critical reference voltage vector and voltage utilization ratio in the maximum overmodulation mode, respectively. According to the phase angle θ _r of the reference voltage vector V _r , it can be deduced in the following three situations:

I)当0＜θ_r＜π/6时，T₂＜T_min，故将T₂调整为T_min，由方程(9)得：I) When 0<θ _r <π/6, T ₂ <T _min , so adjust T ₂ to T _min , and get from equation (9):

${V V}_{limit limit} = = \frac{ρ ρ {U u}_{dc dc}}{\sqrt{33} sin sin {θ θ}_{r r}} {e e}^{j j {θ θ}_{r r}} - - - - - - ((2525))$

II)当π/6＜θ_r＜π/3时，T₁＜T_min，将T₁调整为T_min，由方程(9)得：II) When π/6<θ _r <π/3, T ₁ <T _min , adjust T ₁ to T _min , from equation (9):

${V V}_{limit limit} = = \frac{((11 - - ρ ρ)) {U u}_{dc dc}}{\sqrt{33} sin sin {θ θ}_{r r}} {e e}^{j j {θ θ}_{r r}} - - - - - - ((2626))$

综合以上两种情况得：Combining the above two situations:

${V V}_{limit limit} = = \{\begin{matrix} ((11 - - ρ ρ)) {V V}_{11} + + ρ ρ {V V}_{22},, {θ θ}_{r r} &Element; &Element; [[00,, π π / / 66)) \\ ρ ρ {V V}_{11} + + ((11 - - ρ ρ)) {V V}_{22},, {θ θ}_{r r} &Element; &Element; [[π π / / 66,, π π / / 33)) \end{matrix} - - - - - - ((2727))$

如图18所示，若将空间矢量图第I扇区的的参考电压矢量V_r的轨迹按时间变化映射到β轴，得到临界参考电压矢量V_limit在时域的波形。图左边的参考电压矢量V_r的β轴分量V_rsinθ_r为图右边的临界参考电压矢量V_limit在时域的形式。即临界参考电压矢量V_limit在[0，π/2]表达式如下：As shown in Fig. 18, if the locus of the reference voltage vector V _r in sector I of the space vector diagram is mapped to the β axis according to the time change, the waveform of the critical reference voltage vector V _limit in the time domain is obtained. The β-axis component V _r sin θ _r of the reference voltage vector V _r on the left side of the figure is the form of the critical reference voltage vector V _limit on the right side of the figure in the time domain. That is, the expression of the critical reference voltage vector V _limit in [0, π/2] is as follows:

${V V}_{limit limit} ((θ θ)) = = \{\begin{matrix} ρ ρ \frac{{U u}_{dc dc}}{\sqrt{33}} & 00 \leq \leq {θ θ}_{r r} < < \frac{π π}{66} \\ ((11 - - ρ ρ)) \frac{{U u}_{dc dc}}{\sqrt{33}} & \frac{π π}{66} \leq \leq {θ θ}_{r r} < < \frac{π π}{33} \\ \frac{{U u}_{dc dc}}{\sqrt{33}} & \frac{π π}{33} \leq \leq {θ θ}_{r r} < < \frac{π π}{22} \end{matrix} - - - - - - ((2828))$

根据电压有效值公式综合方程(28)可得，最大过调制方式下线电压基波幅值：According to the comprehensive equation (28) of the voltage effective value formula, the amplitude of the fundamental wave of the offline voltage in the maximum overmodulation mode is:

${U u}_{limit limit__fd fd} = = \frac{44}{π π} {&Integral; &Integral;}_{00}^{\frac{π π}{22}} {u u}_{limit limit} ((θ θ)) sin sin θdθ θdθ$

$= = \frac{44}{π π} {&Integral; &Integral;}_{00}^{\frac{π π}{66}} ρ ρ \frac{{U u}_{dc dc}}{\sqrt{33}} sin sin θdθ θdθ + + \frac{44}{π π} {&Integral; &Integral;}_{\frac{π π}{66}}^{\frac{π π}{33}} ((11 - - ρ ρ)) \frac{{U u}_{dc dc}}{\sqrt{33}} sin sin θdθ θdθ + + \frac{44}{π π} {&Integral; &Integral;}_{\frac{π π}{33}}^{\frac{π π}{22}} \frac{{U u}_{dc dc}}{\sqrt{33}} sin sin θdθ θdθ - - - - - - ((2929))$

$= = \frac{22}{π π} [[11 - - ((22 - - \sqrt{33})) ρ ρ]] \cdot \cdot {U u}_{dc dc}$

最大过调制情况下的电压利用率η_limit：The voltage utilization rate η _limit under the condition of maximum overmodulation:

${η η}_{limit limit} = = \frac{\sqrt{33} {U u}_{limit limit__fd fd}}{{U u}_{dc dc}} = = \frac{22 \sqrt{33}}{π π} [[11 - - ((22 - - \sqrt{33})) ρ ρ]] - - - - - - ((3030))$

因为SVPWM算法中，电压利用率最大值为最大线性调制下的电压利用率η_linear，而本发明的电流重构过调制方法中电压利用率的最大值可达到 $η_{\max} = η_{limit} = \frac{2 \sqrt{3}}{π} [1 - (2 - \sqrt{3}) ρ] .$ 因此本发明有效地提高了电压利用率。Because in the SVPWM algorithm, the maximum value of the voltage utilization rate is the voltage utilization rate η _linear under the maximum linear modulation, and the maximum value of the voltage utilization rate in the current reconstruction over-modulation method of the present invention can reach $η_{\max} = η_{limit} = \frac{2 \sqrt{3}}{π} [1 - (2 - \sqrt{3}) ρ] .$ Therefore, the present invention effectively improves the voltage utilization rate.

如图1所示，基于以上电流重构过调制方法的描述，本发明的电流重构过调制装置操作步骤为：直流电流采样模块13将直流侧的电流传感器4两次采样的直流电流采样值i_dc输入到交流电流重构模块14中。交流电流重构模块14结合SVPWM过调制模块11输入的扇区号信号，重构出三相交流电流值i_a、i_b和i_c。矢量控制器2根据三相交流电流值i_a、i_b和i_c分别计算得出参考电压矢量V_r的实部μ_α和虚部μ_β。SVPWM过调制模块11根据V_r＝μ_α+jμ_β计算得出参考电压矢量V_r，从而根据方程(1)得到调制比M。若产生如图6所示的调制比M过低，或者如图7所示的参考电压矢量V_r接近六个有效矢量中的某一个时，则SVPWM过调制模块11选择修改开关状态法对各邻近有效矢量的作用时间进行调整，从而使得PWM变频器3中的作用时间满足最小采样时间要求。若V_r落在无效区域，则SVPWM过调制模块11选择电流重构过调制方法对参考电压矢量V_r进行修改，进而通过SVPWM算法计算出T₁，T₂，T₀的大小，使得T₁，T₂，T₀满足最小采样时间要求。As shown in FIG. 1 , based on the description of the current reconstruction overmodulation method above, the operation steps of the current reconstruction overmodulation device of the present invention are as follows: the DC current sampling module 13 samples the DC current sampling value twice sampled by the current sensor 4 on the DC side i _dc is input into the alternating current reconstruction module 14 . The AC current reconstruction module 14 combines the sector number signal input by the SVPWM overmodulation module 11 to reconstruct the three-phase AC current values _ia , _ib and _ic . The vector controller 2 respectively calculates the real part μ _α and the imaginary part μ _β of the reference voltage vector V _r according to the three-phase AC current values _ia , _ib and _ic . The SVPWM overmodulation module 11 calculates the reference voltage vector V _r according to V _r =μ _α +jμ _β , so as to obtain the modulation ratio M according to equation (1). If the modulation ratio M shown in Figure 6 is too low, or the reference voltage vector V _r shown in Figure 7 is close to one of the six effective vectors, then the SVPWM over-modulation module 11 selects to modify the switch state method for each The action time of adjacent active vectors is adjusted so that the action time in the PWM inverter 3 meets the minimum sampling time requirement. If V _r falls in the invalid region, the SVPWM overmodulation module 11 selects the current reconstruction overmodulation method to modify the reference voltage vector V _r , and then calculates the size of T ₁ , T ₂ , and T ₀ through the SVPWM algorithm, so that T ₁ , T ₂ , T ₀ meet the minimum sampling time requirement.

其中，电流重构过调制方法的具体步骤为：SVPWM过调制模块11根据调制比M与η_linear、η_edge和η_limit比较的结果选择上述三种过调制方法，并进行叠加，然后利用SVPWM算法计算有效矢量作用时间T₁、T₂和T₀。叠加的原则是保持实际线电压基波幅值U_{1_fd}与预期相等；设应用本发明的电流重构过调制方法后得到的调制后参考电压矢量为V′_r，其中，对最大线性调制、一般过调制方式和最大过调制方式调制情况的具体叠加法则如下：Wherein, the specific steps of the current reconstruction overmodulation method are: the SVPWM overmodulation module 11 selects the above three overmodulation methods according to the comparison results of the modulation ratio M with η _linear , η _edge and η _limit , and superimposes them, and then uses the SVPWM algorithm Calculate effective vector action times T ₁ , T ₂ and T ₀ . The principle of superposition is to keep the actual line voltage fundamental wave amplitude U _{1_fd} equal to the expectation; assume that the modulated reference voltage vector obtained after applying the current reconstruction over-modulation method of the present invention is V′ _r , wherein, for the maximum linear modulation, the general The specific superposition rules of the overmodulation mode and the maximum overmodulation mode modulation are as follows:

A)若M≤η_linear，则此时参考电压矢量V_r处于线性调制区，此时不需要对参考电压矢量V_r做调整，也即调制后参考电压矢量V′_r＝V_r。A) If M≤η _linear , then the reference voltage vector V _r is in the linear modulation region at this time, and there is no need to adjust the reference voltage vector V _r at this time, that is, the reference voltage vector V′ _r =V _r after modulation.

B)若η_linear＜M≤η_edge，此时可用一般过调制方式与最大线性调制方式的线性叠加。令系数B) If η _linear <M≤η _edge , then the linear superposition of the general overmodulation mode and the maximum linear modulation mode can be used. order coefficient

${k k}_{11} = = \frac{M m - - {η η}_{linear linear}}{{η η}_{edge edge} - - {η η}_{linear linear}} - - - - - - ((3131))$

调制后参考电压矢量V′_r＝k₁V_edge+(1-k₁)V_linear。其中V_linear、V_edge分别通过方程(13)和方程(23)得到。The modulated reference voltage vector V' _r =k ₁ V _edge +(1-k ₁ )V _linear . Among them, V _linear and V _{edge are respectively} obtained by equation (13) and equation (23).

C)若η_edge＜M≤η_limit，此时可用一般过调制方式与最大过调制方式的线性叠加。令系数C) If η _edge <M≤η _limit , then the linear superposition of the general overmodulation mode and the maximum overmodulation mode can be used. order coefficient

${k k}_{22} = = \frac{M m - - {η η}_{edge edge}}{{η η}_{limit limit} - - {η η}_{edge edge}} - - - - - - ((3232))$

调制后参考电压矢量V′_r＝k₂V_limit+(1-k₂)V_edge。其中V_edge，V_limit分别通过方程(23)和方程(28)得到。记调制后参考电压矢量V′_r的极坐标形式为 $V_{r}^{'} = | V_{r}^{'} | e^{j θ_{r}^{'}} .$ The modulated reference voltage vector V' _r =k ₂ V _limit +(1-k ₂ )V _edge . Among them, V _edge and V _limit are respectively obtained through Equation (23) and Equation (28). Note that the polar coordinate form of the reference voltage vector V′ _r after modulation is $V_{r}^{'} = | V_{r}^{'} | e^{j θ_{r}^{'}} .$

由于V_linear、V_edge和V_limit均在有效区域内部，因此调制后参考电压矢量V′_r始终位于有效区域内。此时SVPWM过调制模块11利用SVPWM算法计算邻近矢量和零矢量的作用时间。Since V _linear , V _edge and V _limit are all within the effective region, the modulated reference voltage vector V′ _r is always within the effective region. At this time, the SVPWM overmodulation module 11 uses the SVPWM algorithm to calculate the action time of the adjacent vector and the zero vector.

$\{\begin{matrix} {T T}_{11} = = \sqrt{33} {T T}_{s the s} | | {V V}_{r r}^{' '} | | sin sin {θ θ}_{r r}^{' '} / / {U u}_{dc dc} \\ {T T}_{22} = = \sqrt{33} {T T}_{s the s} | | {V V}_{r r}^{' '} | | sin sin ((π π / / 33 - - {θ θ}_{r r}^{' '})) / / {U u}_{dc dc} \\ {T T}_{00} = = {T T}_{s the s} - - {T T}_{11} - - {T T}_{22} \end{matrix} - - - - - - ((3333))$

T₁、T₂和T₀的大小满足最小采样时间T_min，将T₀、T₁和T₂发送到PWM信号调节模块12中，PWM信号调节模块12根据T₀、T₁和T₂的大小，采用七段式对PWM变频器3开关进行控制。PWM变频器3按T₀、T₁和T₂的大小调整开关状态分布以满足电流重构的要求，同时PWM信号调节模块12输出采样点控制信号到直流电流采样模块13。直流电流采样模块13将两次采样得到的直流电流采样值i_dc发送到交流电流重构模块14中，交流电流重构模块14结合扇区号信号重构出三相交流电流。The size of T ₁ , T ₂ and T ₀ satisfies the minimum sampling time T _min , and T ₀ , T ₁ and T ₂ are sent to the PWM signal adjustment module 12, and the PWM signal adjustment module 12 according to T ₀ , T ₁ and T ₂ The size of the PWM inverter 3 switches is controlled by a seven-stage method. The PWM frequency converter 3 adjusts the switch state distribution according to the magnitude of T ₀ , T ₁ and T ₂ to meet the requirements of current reconstruction, and the PWM signal adjustment module 12 outputs the sampling point control signal to the DC current sampling module 13 . The DC current sampling module 13 sends the DC current sampling value i _dc obtained by two samplings to the AC current reconstruction module 14, and the AC current reconstruction module 14 reconstructs a three-phase AC current in combination with the sector number signal.

下面通过一实施例对本发明的装置和方法作进一步说明。The device and method of the present invention will be further described below through an embodiment.

本实施例的电流重构过调制装置中，PWM变频器3为传统的交-直-交型变换器。整流侧采用三相二极管不控整流桥，逆变侧为三相IGBT逆变桥。直流侧电流传感器4采用LEM公司的LA25-NP电流传感器。PWM变频器3的驱动电机5为一永磁同步电机。In the current reconstruction overmodulation device of this embodiment, the PWM frequency converter 3 is a traditional AC-DC-AC converter. The rectifier side uses a three-phase diode uncontrolled rectifier bridge, and the inverter side uses a three-phase IGBT inverter bridge. The DC side current sensor 4 adopts the LA25-NP current sensor of LEM Company. The driving motor 5 of the PWM frequency converter 3 is a permanent magnet synchronous motor.

首先设定永磁同步电机参数：d轴电感Ld＝7.418mH，q轴电感Lq＝12.285mH，转子电阻R＝0.618Ω，极对数p＝2，转动惯量J＝5.59×10-4kgm2，反电势常数KE＝0.1128，参考转速2400n_ref＝2400rpm(转/分)。在此条件下分别利用SVPWM算法和本发明的电流重构过调制方法进行仿真。First set the parameters of the permanent magnet synchronous motor: d-axis inductance Ld=7.418mH, q-axis inductance Lq=12.285mH, rotor resistance R=0.618Ω, number of pole pairs p=2, moment of inertia J=5.59×10-4kgm2, inversely Potential constant KE=0.1128, reference speed 2400n_ref=2400rpm (rotation/minute). Under this condition, the SVPWM algorithm and the current reconstruction overmodulation method of the present invention are used for simulation respectively.

利用SVPWM算法进行仿真时，设定参数：直流母线电压U_dc＝150V，负载转矩T_L＝2nm，采样周期T_s＝100μs，控制周期T_c＝T_s，死区时间T_dead＝0，最小采样持续时间T_min＝10μs。When using the SVPWM algorithm for simulation, set parameters: DC bus voltage U _dc =150V, load torque T _L =2nm, sampling period T _s =100μs, control period T _c =T _s , dead time T _dead =0, Minimum sampling duration T _min =10 μs.

当采用本发明的电流重构过调制方法时，设定进行仿真的永磁同步电机参数：直流母线电压U_dc＝135v，负载转矩T_L＝2nm，采样周期T_s＝100μs，控制周期T_c＝T_s，死区时间T_dead＝0，最小采样持续时间T_min＝10μs。When using the current reconstruction overmodulation method of the present invention, set the parameters of the permanent magnet synchronous motor for simulation: DC bus voltage _Udc = 135v, load torque T _L = 2nm, sampling period T _s = 100μs, control period T _c =T _s , dead time T _dead =0, minimum sampling duration T _min =10 μs.

如图19、图20所示，在相同的负载转矩下，永磁同步电机控制系统参考转速输入2400rpm时，装置在本发明提出的电流重构过调制方法和SVPWM方式下均能稳定运行在2400rpm。由于采用电流重构过调制方法时的直流母线电压U_dc为135v，而当采用SVPWM算法时的直流母线电压U_dc为150v。因此，本发明提出的电流重构过调制方法在直流母线电压低的情况下，系统的电压利用率高。As shown in Figure 19 and Figure 20, under the same load torque, when the reference speed input of the permanent magnet synchronous motor control system is 2400rpm, the device can run stably under the current reconstruction overmodulation method and the SVPWM mode proposed by the present invention. 2400rpm. Since the DC bus voltage U _dc is 135v when the current reconstruction overmodulation method is used, and the DC bus voltage U _dc is 150v when the SVPWM algorithm is used. Therefore, the current reconstruction overmodulation method proposed by the present invention has a high voltage utilization rate of the system when the DC bus voltage is low.

本发明方法和装置的实施例仅用于说明本发明，其中各部件的结构、设置位置、连接方式，及方法步骤的设置和顺序都是可以有所变化的，凡是在本发明技术方案的基础上进行的改进和等同变换，均不应排除在本发明的保护范围之外。The embodiments of the method and device of the present invention are only used to illustrate the present invention, wherein the structure of each component, the setting position, the connection mode, and the setting and order of the method steps can be changed to some extent. Improvements and equivalent transformations made above shall not be excluded from the protection scope of the present invention.

Claims

1. the electric current reconstructing and over-modulating device of an air conditioning frequency converter, it is characterized in that: it comprises electric current reconstructing device, vector controller, pulse width modulated frequency convertor, DC side current sensor, motor and encoder; Wherein, described electric current reconstructing device comprises sinusoidal voltage pulse-width modulation ovennodulation module, pulse-width signal adjustment module, direct current sampling module and alternating current reconstructed module; Described electric current reconstructing device and vector controller to the step that pulse width modulated frequency convertor carries out closed-loop control are jointly: at first, described direct current sampling module is under the effect of the sampled point control signal of described pulse-width signal adjustment module input, by described DC side current sensor the DC side of described pulse width modulated frequency convertor is carried out twice direct current sampling, described twice direct current sampled value is input in the described alternating current reconstructed module, described alternating current reconstructed module goes out three-phase alternating current in conjunction with the sector number signal reconstruction of described sinusoidal voltage pulse-width modulation ovennodulation module input, described three-phase alternating current flow valuve is imported described vector controller, in conjunction with the rotor position angle of described encoder output, calculate the value of reference voltage vector Vr;

Described sinusoidal voltage pulse-width modulation ovennodulation module is to described reference voltage vector V _rCarry out resolution of vectors,, then adjust the described action time that closes on effective vector if the process resolution of vectors obtains closing on the requirement of not satisfying minimal sampling time action time of effective vector; Described pulse-width signal adjustment module will be sent to the adjusted described action time that closes on effective vector, described pulse-width signal adjustment module is according to described each adjusted duration of controlling the state of each switch in the described pulse width modulated frequency convertor action time that closes on vector, and to described direct current sampling module output sampled point control signal, described direct current sampling module carries out twice direct current sampling by described DC side current sensor to the DC side of described pulse width modulated frequency convertor under the effect of sampled point control signal, described twice direct current sampled value is input in the described alternating current reconstructed module, and described alternating current reconstructed module goes out three-phase alternating current in conjunction with the sector number signal reconstruction of described sinusoidal voltage pulse-width modulation ovennodulation module input.

2. the electric current reconstructing ovennodulation method of the air conditioning frequency converter that installs according to claim 1 of a realization, it is characterized in that: the device of air conditioning frequency converter electric current reconstructing ovennodulation at first is set, and described air conditioning frequency converter electric current reconstructing device comprises electric current reconstructing device, vector controller, pulse width modulated frequency convertor, DC side current sensor; Wherein, described electric current reconstructing device comprises sinusoidal voltage pulse-width modulation ovennodulation module, pulse-width signal adjustment module, direct current sampling module and alternating current reconstructed module; Described direct current sampling module is under the effect of the sampled point control signal of described pulse-width signal adjustment module input, by the DC side current sensor DC side of described pulse width modulated frequency convertor is carried out twice direct current sampling, and with twice direct current sampled value i _DcBe input in the described alternating current reconstructed module, described alternating current reconstructed module goes out three-phase alternating current i in conjunction with the sector number signal reconstruction of described sinusoidal voltage pulse-width modulation ovennodulation module input _a, i _bAnd i _c, with described three-phase alternating current flow valuve i _a, i _bAnd i _cImport described vector controller, the rotor position angle θ in conjunction with encoder output calculates reference voltage vector V _rReal part μ _αWith imaginary part μ _β, described sinusoidal voltage pulse-width modulation ovennodulation module is according to V _r=μ _α+ j μ _βCalculate described reference voltage vector V _rValue, then with described reference voltage vector V _rBe input to described sinusoidal voltage pulse-width modulation ovennodulation module, described sinusoidal voltage pulse-width modulation ovennodulation module is calculated modulation ratio M according to equation (1),

M = \frac{| V_{r} |}{U_{dc} / \sqrt{3}} - - - (1)

U wherein _DcBe DC bus-bar voltage; Calculate the voltage utilization η under the maximum linear modulation then _Linear, the voltage utilization η under the general ovennodulation mode _EdgeWith the voltage utilization η under the maximum ovennodulation mode _Limit

If modulation ratio M is low excessively, perhaps described reference voltage vector V _rDuring near in six effective vectors any, then described sinusoidal voltage pulse-width modulation ovennodulation module selects to revise the on off state method to adjusting action time of each contiguous effective vector; If described reference voltage vector V _rDrop on inactive area, then select electric current reconstructing toning method for making adjusting the action time that respectively closes on vector; Make and satisfy action time in the described pulse width modulated frequency convertor described minimal sampling time requirement;

Described electric current reconstructing ovennodulation method step comprises: described sinusoidal voltage pulse-width modulation ovennodulation module is with the voltage utilization η under described modulation ratio M and the maximum linear modulation _Linear, the voltage utilization η under the general ovennodulation mode _EdgeWith the voltage utilization η under the maximum ovennodulation mode _LimitRelatively:

Reference voltage vector is V after the modulation that obtains after the described electric current reconstructing ovennodulation method if use _r',

A) if the voltage utilization η under modulation ratio M≤maximum linear modulation _Linear, do not need described reference voltage vector V this moment _rAdjust, also reference voltage vector V after the i.e. modulation _r'=V _r

B) if the voltage utilization η under the maximum linear modulation _LinearVoltage utilization η under '＜modulation ratio M≤general ovennodulation mode _Edge, this seasonal coefficient

k_{1} = \frac{M - η_{linear}}{η_{edge} - η_{linear}}

Modulation back reference voltage vector V ' _r=k ₁V _Edge+ (1-k ₁) V _Linear

C) if the voltage utilization η under the general ovennodulation mode _EdgeVoltage utilization η under＜modulation ratio M≤maximum ovennodulation mode _Limit, this seasonal coefficient

k_{2} = \frac{M - η_{edge}}{η_{limit} - η_{edge}}

Modulation back reference voltage vector V ' _r=k ₂V _Limit+ (1-k ₂) V _Edge

Described sinusoidal voltage pulse-width modulation ovennodulation module utilizes sinusoidal voltage pulse width modulation algorithm calculating two to close on effective vector T action time ₁, T ₂T action time with zero vector ₀:

[\begin{matrix} T_{1} = \sqrt{3} T_{s} | V_{r}^{'} | \sin θ_{r}^{'} / U_{dc} \\ T_{2} = \sqrt{3} T_{s} | V_{r}^{'} | \sin (π / 3 - θ_{r}^{'}) \\ T_{0} = T_{s} - T_{1} - T_{2} \end{matrix}, / U_{dc}]

Wherein, θ ' _rFor modulating the back reference voltage vector is V ' _rPhase angle, described sinusoidal voltage pulse-width modulation ovennodulation module is closed on effective vector T action time with described two ₁, T ₂T action time with zero vector ₀Send in the described pulse-width signal adjustment module, described pulse-width signal adjustment module is closed on effective vector T action time according to described two ₁, T ₂T action time with zero vector ₀Size, adopt seven segmentations that described pulse width modulated frequency convertor switch is controlled; Simultaneously described pulse-width signal adjustment module output sampled point control signal is to described direct current sampling module; The direct current sampled value that described direct current sampling module obtains double sampling sends in the described alternating current reconstructed module, and described alternating current reconstructed module goes out three-phase alternating current in conjunction with the sector number signal reconstruction that described sinusoidal voltage pulse width modulation module sends.

3. the electric current reconstructing ovennodulation method of a kind of air conditioning frequency converter as claimed in claim 2 is characterized in that:

Voltage utilization η under the described maximum linear modulation _LinearFor

η_{linear} = \min (1, \frac{2 (1 - ρ)}{\sqrt{3}})

Reference voltage vector V under the described maximum linear modulation _LinearFor

V_{linear} = \frac{u_{dc}}{\sqrt{3}} η_{linear} \cdot e^{j θ_{r}}

Wherein, θ _rBe the reference voltage vector V under the described maximum linear modulation _LinearPhase angle, ρ is minimal sampling time T _MinWith control cycle T _sRatio.

4. the electric current reconstructing ovennodulation method of a kind of air conditioning frequency converter as claimed in claim 2 is characterized in that:

Voltage utilization η under the described general ovennodulation mode _EdgeFor

Reference voltage vector V under the described general ovennodulation mode _EdgeFor

θ wherein _rBe the reference voltage vector V under the described general ovennodulation mode _EdgePhase angle, ρ is minimal sampling time T _MinWith control cycle T _sRatio;

Be the hexagonal centre of resolution of vectors figure vector and vector V to a summit of inactive area ₁Between angle:

5. the electric current reconstructing ovennodulation method of a kind of air conditioning frequency converter as claimed in claim 2 is characterized in that:

Voltage utilization η under the described maximum ovennodulation mode _LimitFor

η_{limit} = \frac{2 \sqrt{3}}{π} [1 - (2 - \sqrt{3}) ρ]

Reference voltage vector V under the described maximum ovennodulation mode _LimitFor

V_{limit} (θ) = \{\begin{matrix} ρ \frac{U_{dc}}{\sqrt{3}} & 0 \leq θ_{r} < \frac{π}{3} \\ (1 - ρ) \frac{U_{dc}}{\sqrt{3}} & \frac{π}{6} \leq θ_{r} < \frac{π}{3} \\ \frac{U_{dc}}{\sqrt{3}} & \frac{π}{3} \leq θ_{r} < \frac{π}{2} \end{matrix}

Wherein, θ _rBe the reference voltage vector V under the described maximum ovennodulation mode _LimitPhase angle, ρ is minimal sampling time T _MinWith control cycle T _sRatio.

6. as claim 3 or 4 or 5 described a kind of electric current reconstructing ovennodulation methods, it is characterized in that: described inactive area is the diamond-shaped area at place, six summits among the hexagon resolution of vectors figure, promptly drops on the reference voltage vector V in the described inactive area _rThrough after decomposing, obtain any action time in two described effective vectors that close on less than described minimal sampling time T _Min