Power converter

Abstract

In power converters connected to power grids, the stability of current control is improved while enhancing control responsiveness. [Solution] The power converter (40) is connected to the power system (15) via the system wiring (13) and converts AC power to DC power for output. The power converter includes a PFC coil (41), a common mode coil (51) having leakage inductance connected between the PFC coil and the system wiring, and a voltage command unit (80) that determines a voltage command value for controlling the output current to a current command value based on a predetermined feedforward term. The voltage command unit calculates a system approximate voltage that is closer to the system voltage of the power system than the input voltage of the PFC coil, based on the input voltage of the PFC coil and the voltage drop in the common mode coil, and calculates a predetermined feedforward term based on the calculated system approximate voltage.

Description

Technical Field

[0001] The present invention relates to a power conversion device that converts AC power into DC power and outputs it.

Background Art

[0002] For example, there is a grid-connected inverter connected to a pole-mounted transformer (power grid) via a wiring (see Patent Document 1). The grid-connected inverter described in Patent Document 1 estimates the grid impedance, which is the impedance of the wiring from the grid-connected inverter to the pole-mounted transformer, and sets control parameters so that the control system becomes stable based on the estimated grid impedance. The control parameter is a set value for controlling the operating characteristics of the grid-connected inverter.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] By the way, in order to charge an in-vehicle battery with the grid voltage of the power grid, there is a charger that converts AC power into DC power with a PFC converter, which is an AC / DC power conversion device, and controls the battery charging current with a DC / DC power conversion device. The PFC converter uses the input voltage of the PFC coil in the feed-forward (FF) term to increase the control responsiveness, and determines a voltage command value for controlling the output current to a current command value.

[0005] It has recently been discovered that when the inductance of the system wiring from the PFC converter to the pole-mounted transformer (power system) is large (system inductance), the phase margin of the PFC converter's current control system decreases, leading to unstable current control. The inventors of this invention focused on the fact that the large system inductance causes fluctuations in the input voltage of the PFC coil, and using this value in the FF term exacerbates current oscillations.

[0006] The present invention was made to solve the above problems, and its main objective is to improve the stability of current control while enhancing control responsiveness in a power conversion device connected to a power grid. [Means for solving the problem]

[0007] The first means to solve the above problem is, A power converter (40) connected to a power grid (15) via a grid wiring (13) that converts AC power to DC power and outputs it, PFC coil (41), A common mode coil (51) having leakage inductance is connected between the PFC coil and the system wiring, It includes a voltage command unit (81, 82, 86) that determines a voltage command value for controlling the output current to a current command value based on a predetermined feedforward term, The voltage command unit calculates a system approximate voltage that is closer to the power system voltage than the input voltage of the PFC coil, based on the input voltage of the PFC coil and the voltage drop in the common mode coil, and calculates the predetermined feedforward term based on the calculated system approximate voltage.

[0008] According to the above configuration, the power converter is connected to the power grid via grid wiring and converts AC power to DC power for output.

[0009] Here, the voltage command unit determines a voltage command value for controlling the output current to a current command value based on a predetermined feedforward term (hereinafter referred to as the "predetermined FF term"). If the voltage command unit calculates the predetermined FF term based only on the input voltage of the PFC coil, the control responsiveness can be improved, but the following problem arises. That is, if the inductance of the power grid wiring (hereinafter referred to as the "system inductance") is large, the input voltage of the PFC coil fluctuates, and the current control becomes unstable. On the other hand, the system voltage, which is the voltage of the power grid, is stable regardless of the magnitude of the system inductance. For this reason, the inventors of this application focused on the fact that current control can be stabilized if the system voltage is used in the predetermined FF term. However, the power converter cannot directly detect the system voltage.

[0010] In this regard, the common mode coil is connected between the PFC coil and the power grid wiring and has leakage inductance. Therefore, a voltage drop occurs due to the leakage inductance of the common mode coil, and this voltage drop in the common mode coil correlates with the voltage drop in the power grid wiring. Thus, by using the input voltage of the PFC coil and the voltage drop in the common mode coil, it is possible to calculate a system approximate voltage that is closer to the power grid voltage than the input voltage of the PFC coil. The voltage command unit then calculates the predetermined FF term based on the calculated system approximate voltage. Therefore, fluctuations in the predetermined FF term, and consequently fluctuations in the output current, can be suppressed compared to when the predetermined FF term is calculated based only on the input voltage of the PFC coil. Thus, in a power converter connected to a power grid, the stability of current control can be improved while enhancing control responsiveness.

[0011] Furthermore, the leakage inductance of the common-mode coil is utilized to obtain a voltage drop that correlates with the voltage drop in the power grid wiring. In other words, by deliberately using a common-mode coil with leakage inductance, a voltage drop that correlates with the voltage drop in the power grid wiring is generated in the common-mode coil. Therefore, there is no need to add inductors or other components that generate a voltage drop correlated with the voltage drop in the power grid wiring, thus reducing the need for additional components. [Brief explanation of the drawing]

[0012] [Figure 1] A circuit diagram showing the onboard charger and its peripheral components. [Figure 2] Circuit diagram of the PFC converter according to the first embodiment. [Figure 3] A block diagram showing the procedure for determining the d-axis voltage command value of the PFC converter in Figure 2. [Figure 4] A flowchart showing the procedure for controlling the Id command value. [Figure 5] A graph showing the simulation results of the relationship between system inductance, phase margin, and proportionality constant. [Figure 6] This time chart shows the simulation results when the flowchart in Figure 4 is executed with a system inductance of 0. [Figure 7] This time chart shows the simulation results when the flowchart in Figure 4 is executed at the maximum value of the system inductance. [Figure 8] Circuit diagram of the PFC converter according to the second embodiment. [Figure 9] A block diagram showing the procedure for determining the d-axis voltage command value of the PFC converter in Figure 8. [Figure 10] Circuit diagram of the PFC converter according to the third embodiment. [Figure 11] A block diagram showing the procedure for determining the d-axis voltage command value of the PFC converter in Figure 10. [Figure 12] A flowchart illustrating an example of a change in the procedure for controlling the Id command value. [Figure 13]A flowchart showing another example of a procedure for controlling an Id command value. [Figure 14] A block diagram showing an example of a change in the procedure for determining the d-axis voltage command value.

Embodiments for Carrying Out the Invention

[0013] (First Embodiment) Hereinafter, the first embodiment implemented in an in-vehicle charger for charging an in-vehicle battery will be described with reference to the drawings.

[0014] As shown in FIG. 1, the vehicle 10 includes an in-vehicle battery 11 and an in-vehicle charger 20. The in-vehicle battery 11 is a secondary battery such as a lithium-ion battery, a nickel-metal hydride battery, or a lead-acid battery.

[0015] The in-vehicle charger 20 includes an isolated DC / DC converter 31, a PFC converter 40, DC terminals DCL1, DCL2, AC terminals ACL1, ACL2, ACL3, and a neutral point terminal ACLN, etc.

[0016] The DC terminal DCL1 is connected to the positive electrode of the in-vehicle battery 11. The DC terminal DCL2 is connected to the negative electrode of the in-vehicle battery 11.

[0017] When charging the in-vehicle battery 11, the AC terminals ACL1, ACL2, ACL3 are connected to the power grid 15 via the grid wiring 13, and the neutral point terminal ACLN is connected to the neutral point N of the power grid 15. The power grid 15 (corresponding to the power system) supplies three-phase AC power.

[0018] The PFC converter 40 (corresponding to the power conversion device) is connected to the isolated DC / DC converter 31, the AC terminals ACL1, ACL2, ACL3, and the neutral point terminal ACLN. That is, when charging the in-vehicle battery 11, the PFC converter 40 is connected to the power grid 15 via the grid wiring 13. The PFC converter 40 converts the AC power input from the AC terminals ACL1, ACL2, ACL3 into DC power and inputs it to the isolated DC / DC converter 31.

[0019] The isolated DC / DC converter 31 is an isolated type DC / DC converter. When charging the vehicle battery 11, the isolated DC / DC converter 31 controls the current flowing through the vehicle battery 11 and the voltage applied to the vehicle battery 11.

[0020] As shown in Figure 2, the PFC converter 40 includes an AC filter 50, a PFC coil 41, a current sensor 43, an inverter circuit 60, a smoothing capacitor 69, a voltage sensor 45, and a control device 80, etc.

[0021] The AC filter 50 is a circuit that removes AC power (AC current) within a predetermined frequency range from the AC power input from AC terminals ACL1, ACL2, and ACL3. A low-pass filter is used as the AC filter 50. The AC filter 50 has a series connection for each phase, consisting of two common coils 51 and 52 connected in series. The common coils 51 and 52 are connected between the PFC coil 41 and the system wiring 13. The AC filter 50 has X capacitors 53, 54, and 55, one end of which is connected to each phase. The other end of each X capacitor 53, 54, and 55 is connected to the neutral point terminal ACLN. In other words, the other ends of each X capacitor 53, 54, and 55 are connected to each other.

[0022] The PFC coil 41 is connected between the AC filter 50 and the inverter circuit 60. A PFC coil 41 is provided for each phase. The voltage sensor 42 detects the input voltages Vu, Vv, and Vw that are input to the PFC coil 41 of each phase. A current sensor 43 is provided for each phase and detects the currents Iu, Iv, and Iw that flow through the PFC coil 41 of each phase.

[0023] The inverter circuit 60 is composed of a full-bridge circuit, each having the same number of upper and lower arms as the number of phases in the three-phase system. The PFC coil 41 for each phase is connected to the connection point between the switching element of the upper arm and the switching element of the lower arm in each phase. A smoothing capacitor 69 is connected between the inverter circuit 60 and the isolated DC / DC converter 31. The drive state of the inverter circuit 60 is controlled by the control device 80.

[0024] The control device 80 controls the output current to the isolated DC / DC converter 31 while boosting the system voltage input to the PFC converter 40 through the action of the PFC coil 41, inverter circuit 60, and smoothing capacitor 69. In order to improve the power factor of the PFC converter 40, the control device 80 uses the input voltages Vu, Vv, Vw of each phase PFC coil 41 detected by the voltage sensor 42, and the currents Iu, Iv, Iw flowing through each phase PFC coil 41 detected by the current sensor 43 for control. Generally, the PFC converter 40 uses the input voltage of the PFC coil 41 in a feedforward (FF) term to determine the voltage command value for controlling the output current to the current command value in order to improve control responsiveness.

[0025] Here, the magnitude of the inductance Lg of the system wiring 13 (hereinafter referred to as "system inductance Lg") increases as the length of the system wiring 13 increases (it changes with length). Since the environments in which the on-board charger 20 is used are all different, neither the manufacturer nor the user can know in advance the magnitude of the system inductance Lg in the environment in which the on-board charger 20 is used. It has been newly discovered that when the system inductance Lg is large, the phase margin of the current control system of the PFC converter 40 decreases, and the current control becomes unstable. The voltages input to the above AC terminals ACL1, ACL2, ACL3, and consequently the input voltages Vu, Vv, Vw of the PFC coil 41, are more prone to fluctuation as the system inductance Lg increases. The inventors of this application have focused on the fact that when the system inductance Lg is large, using the input voltages Vu, Vv, Vw of the PFC coil 41 in the FF term exacerbates the oscillation of the currents Iu, Iv, Iw.

[0026] To solve the above problem, in this embodiment, the common coil 51 (corresponding to a common mode coil) has a leakage inductance La. The PFC converter 40 is equipped with a voltage sensor 45 for each phase. The voltage sensor 45 for each phase detects the voltages Vua, Vva, Vwa across the common coil 51 of each phase, that is, the voltage drop due to the leakage inductance La of the common coil 51 of each phase.

[0027] The system voltage of system 15 is stable regardless of the magnitude of the system inductance Lg. Therefore, the inventors of this invention focused on the fact that current control can be stabilized by using the system voltage in the FF term described above. However, the system voltage cannot be directly detected from the PFC converter 40. Here, the system voltages Vus, Vvs, Vws of each phase of system 15 are equal to the input voltages Vu, Vv, Vw of the PFC coil 41 of each phase plus the voltage drops ΔVus, ΔVvs, ΔVws due to the system inductance Lg of each phase. For example, Vus = Vu + ΔVus. Note that the voltage drops ΔVus, ΔVvs, ΔVws due to the system inductance Lg of the system wiring 13 of each phase are sufficiently larger than the voltage drop in the AC filter 50, so the voltage drop in the AC filter 50 is ignored.

[0028] The voltage drops Vua, Vva, and Vwa across the common coil 51 of each phase (i.e., the voltage drops occurring between the PFC coil 41 and the system wiring 13) are correlated with the voltage drops ΔVus, ΔVvs, and ΔVws due to the system inductance Lg of each phase (the voltage drops in the system wiring 13 of each phase). Therefore, by multiplying the voltage drops Vua, Vva, and Vwa across the common coil 51 of each phase by a proportionality constant K, the voltage drops ΔVus, ΔVvs, and ΔVws due to the system inductance Lg of each phase can be estimated. For example, ΔVus can be estimated (expressed) as ΔVus = Vua * K. The optimal value of the proportionality constant K is determined by the relationship between the inductance Lg of the system wiring 13 and the leakage inductance La of the common coil 51. The control device 80 changes the proportionality constant K to bring the product of the voltage drops Vua, Vva, Vwa and the proportionality constant K closer to the voltage drops ΔVus, ΔVvs, ΔVws due to the system inductance Lg of each phase.

[0029] Here, the current flowing through the common coil 51 of each phase towards the system wiring 13 includes high-frequency components (high-frequency noise) generated by the switching operation of the switching elements of the inverter circuit 60. The PFC converter 40 is equipped with an X capacitor 53, one end of which is connected to the connection point P between the system wiring 13 and the common coil 51, and the other end of which is connected to the neutral point N. Therefore, the current flowing through the system wiring 13 of each phase is the current after the high-frequency components have been removed by the X capacitor 53. Accordingly, in order to more accurately approximate the voltage drops ΔVus, ΔVvs, ΔVws in the system wiring 13 of each phase based on the voltage drops Vua, Vva, Vwa in the common coil 51 of each phase, the control device 80 estimates the voltage drops ΔVus, ΔVvs, ΔVws by multiplying the voltage drop obtained by removing (attenuating) the high-frequency components from the voltage drops Vua, Vva, Vwa in the common coil 51 of each phase by a proportionality constant K.

[0030] The control device 80 performs the following control using the voltages Vua, Vva, and Vwa across the common coil 51 for each phase, detected by the voltage sensors 45 for each phase. Figure 3 is a block diagram showing the procedure for determining the d-axis voltage command value Vd* of the PFC converter 40 in Figure 2. In this example, the d-axis direction is defined as the active power.

[0031] The control device 80 is mainly composed of a microcomputer equipped with a CPU, memory (ROM, RAM), input / output interface, and drive circuitry. The control device 80 implements the functions of a low-pass filter 89, a system voltage estimation unit 81, dq conversion units 82, 83, a subtractor 84, a PI controller 85, an adder 86, 87, and a vector control unit 88 by executing a stored program, for example.

[0032] The low-pass filter 89 is, for example, a digital low-pass filter. The low-pass filter 89 allows frequency components lower than a predetermined cutoff frequency to pass through and removes (attenuates) frequency components higher than the cutoff frequency. The cutoff frequency is a frequency at which high-frequency components generated by the switching operation of the switching elements of the inverter circuit 60 can be removed, for example, 1 kHz. The low-pass filter 89 receives the voltage drops Vua, Vva, Vwa at the common coil 51 of each phase detected by the voltage sensors 45 of each phase, and inputs the voltage drops Vua, Vva, Vwa after the high-frequency components have been removed to the system voltage estimation unit 81. The low-pass filter 89 may also be an analog low-pass filter circuit.

[0033] The system voltage estimation unit 81 receives the input voltages Vu, Vv, Vw of the PFC coils 41 of each phase detected by the voltage sensors 42 for each phase, and the voltage drops Vua, Vva, Vwa at the common coil 51 of each phase after passing through the low-pass filter 89. The system voltage estimation unit 81 estimates the system voltages Vus, Vvs, Vws of each phase by summing the detected input voltages Vu, Vv, Vw of the PFC coils 41 of each phase and the estimated voltage drops ΔVus, ΔVvs, ΔVws (corresponding to predetermined voltage drops) due to the system inductance Lg of each phase. For example, it is estimated using Vus = Vu + Vua * K. The estimated system voltages Vus, Vvs, Vws of each phase correspond to approximate system voltages that are closer to the actual system voltages Vus, Vvs, Vws of each phase than to the input voltages Vu, Vv, Vw of the PFC coils 41 of each phase.

[0034] The dq conversion unit 82 converts the estimated system voltages Vus, Vvs, and Vws of each phase into the d-axis voltage Vd and the q-axis voltage Vq. The dq conversion unit 82 inputs the d-axis voltage Vd as a feedforward term (corresponding to a predetermined feedforward term) to the adder 86.

[0035] The dq conversion unit 83 converts the currents Iu, Iv, and Iw detected by the current sensors 43 for each phase into the d-axis current Id and the q-axis current Iq. The dq conversion unit 83 inputs the d-axis current Id to the subtractor 84.

[0036] The subtractor 84 receives the Id command value (d-axis current command value Id*). The Id command value (corresponding to the current command value) is calculated, for example, based on the Id request value from the isolated DC / DC converter 31. The subtractor 84 inputs the difference ΔId between the Id command value and the d-axis current Id (corresponding to the output current) to the PI controller 85.

[0037] The PI controller 85 calculates a feedback term (corresponding to a predetermined feedback term) for controlling the d-axis current Id to the Id command value based on the input difference ΔId. The PI controller 85 inputs the calculated feedback term (FB term) to the adder 86.

[0038] Adder 86 inputs the sum of the d-axis voltage Vd (as a feedforward term) and the FB term to adder 87. Adder 87 adds a correction term ωLIq to the sum of the d-axis voltage Vd (as a feedforward term) and the FB term to calculate the d-axis voltage command value Vd*. That is, the d-axis voltage command value Vd* is determined based on the FF term and the FB term. Note that the calculation method of the correction term ωLIq is publicly known, so a detailed explanation is omitted. Adder 87 inputs the d-axis voltage command value Vd* to vector control unit 88. The voltage command unit is composed of the system voltage estimation unit 81, the dq conversion unit 82, and adder 86.

[0039] The vector control unit 88 calculates drive signals Duty_u to Duty_w for each switching element of the inverter circuit 60 based on the d-axis voltage command value Vd* using vector control. Various known methods can be applied to the calculation of the drive signals Duty_u to Duty_w, so a detailed explanation is omitted. The inverter circuit 60 is then driven by the drive signals Duty_u to Duty_w.

[0040] Figure 4 is a flowchart showing the procedure for controlling the Id command value. This series of processes is performed by the control device 80.

[0041] First, the proportionality constant K is set to 0, and the Id command value is set to a current command value equivalent to, for example, 1 / 5 of the rated output (corresponding to the first current) (S10). Alternatively, the Id command value may be set to a current command value equivalent to 1 / 10 to 1 / 5 of the rated output.

[0042] Next, it is determined whether the difference ΔId between the Id command value and the d-axis current Id is continuously oscillating, or whether the current output has been stopped due to overcurrent (S11). The d-axis current Id is calculated by converting the currents Iu, Iv, and Iw detected by the current sensors 43 for each phase into dq values. It can be determined that the difference ΔId is continuously oscillating, for example, if the peak-to-peak value of the difference ΔId (the difference between the maximum peak value and the minimum peak value) is greater than a threshold, or if the AC RMS value of the difference ΔId is greater than a threshold. It can be determined that the current output has been stopped due to overcurrent, for example, if the RMS value of the d-axis current Id is greater than a threshold. If this determination is negative (S11: NO), the process proceeds to S13. That is, if it is determined that the difference ΔId between the Id command value and the d-axis current Id is not continuously oscillating, and it is determined that the current output has not been stopped due to overcurrent, the control device 80 maintains the proportionality constant K=0 and calculates the d-axis voltage Vd as an FF term in the procedure shown in Figure 3. The control device 80 then drives the inverter circuit 60 based on the d-axis voltage command value Vd* calculated using the d-axis voltage Vd. Note that if the difference ΔId between the Id command value and the d-axis current Id continues to oscillate, it corresponds to the degree of oscillation of the current flowing through the PFC coil 41 being greater than a predetermined degree.

[0043] On the other hand, in the determination in S11, if it is determined that the difference ΔId between the Id command value and the d-axis current Id is continuously oscillating, or if it is determined that the current output has stopped due to overcurrent (S11:YES), the value obtained by adding the coefficient change amount ΔK to the proportionality constant K is set as the new proportionality constant K (S12). The coefficient change amount ΔK can be set in advance to an appropriate value by conducting tests or simulations using the magnitude of the system inductance Lg expected in the operating environment of the onboard charger 20. Then, using the new proportionality constant K, the d-axis voltage Vd as the FF term is calculated in the procedure shown in Figure 3. Then, the control device 80 drives the inverter circuit 60 based on the d-axis voltage command value Vd* calculated using this d-axis voltage Vd.

[0044] Next, it is determined whether the difference ΔId between the Id command value and the d-axis current Id is stable (S13). Stability of the difference ΔId between the Id command value and the d-axis current Id can be determined, for example, by whether the peak-to-peak value of the difference ΔId is greater than the threshold, or whether the AC RMS value of the difference ΔId is greater than the threshold, similar to the process in S11. The threshold in the process of S13 may be the same as the threshold in the process of S11, or it may be a different value. If it is determined in this determination that the difference ΔId between the Id command value and the d-axis current Id is not stable (S13:NO), the process is repeated starting from S11.

[0045] On the other hand, in the determination in S13, if it is determined that the difference ΔId between the Id command value and the d-axis current Id is stable (S13: YES), the Id command value is set to, for example, a current command value equivalent to the rated output (corresponding to the second current) (S14). Then, the control device 80 calculates the d-axis voltage command value Vd* using the Id command value equivalent to the rated output and the current proportionality coefficient K in the procedure shown in Figure 3, and drives the inverter circuit 60 based on this d-axis voltage command value Vd*. After that, this series of processes is terminated (END). That is, if the oscillation degree of the current flowing through the PFC coil 41 is greater than a predetermined degree while the Id command value is set to the first current, the proportionality coefficient K is increased from the current proportionality coefficient K to reduce the oscillation degree of the current flowing through the PFC coil 41 to less than a predetermined degree, and then the Id command value is set to a second current greater than the first current.

[0046] Figure 5 is a graph showing the simulation results of the relationship between system inductance Lg, phase margin, and proportionality constant K. Lg_Max is the maximum value of system inductance Lg expected in the operating environment of the onboard charger 20. When the phase margin becomes 0° (phase lag of 180°), the current may oscillate and resonate.

[0047] When the proportionality constant K=0, the phase margin is greater than 0° for system inductances Lg from 0 to Lg2, and current control is stable. However, when the system inductance Lg becomes greater than Lg2, the phase margin becomes 0°, and there is a risk of current oscillation and resonance.

[0048] When the proportionality constant K=K2 (for example, K2=K+ΔK), the system inductance Lg has a phase margin greater than 0° from Lg1 to Lg_Max, and the current control is stable.

[0049] Therefore, for example, by setting the proportionality constant K=0 when the system inductance Lg is from 0 to Lg2, and the proportionality constant K=K2 when the system inductance Lg is from Lg2 to Lg_Max, current control can be stabilized when the system inductance Lg is from 0 to Lg_Max.

[0050] Figure 6 is a time chart showing the simulation results when the flowchart in Figure 4 is executed with a system inductance Lg of 0.

[0051] At time t11, the control device 80 is set to a proportionality constant K=0 and an Id command value equivalent to 1 / 5 of the rated output. As shown in Figure 5, with system inductance Lg=0 and proportionality constant K=0, the phase margin becomes greater than 0°, and current control becomes stable. For this reason, the difference ΔId between the Id command value and the detected Id value (d-axis current Id calculated by dq conversion) is approximately 0, and the difference ΔId is stable.

[0052] At time t12, the control device 80 determines that the difference ΔId is stable and maintains the proportionality constant K=0, setting the Id command value to the value equivalent to the rated output. Even after the Id command value becomes equivalent to the rated output, the difference ΔId between the Id command value and the detected Id value is approximately 0, indicating that the difference ΔId is stable.

[0053] Figure 7 is a time chart showing the simulation results when the flowchart in Figure 4 is executed at Lg_Max for the system inductance Lg.

[0054] At time t21, the control device 80 is set to a proportionality constant K=0 and an Id command value equivalent to 1 / 5 of the rated output. As shown in Figure 5, with system inductance Lg=Lg_Max and proportionality constant K=0, the phase margin becomes 0 and the current oscillates. As a result, the difference ΔId between the Id command value and the detected Id value becomes larger than the threshold.

[0055] At time t22, the control device 80 determines that the difference ΔId is oscillating continuously and updates the proportionality constant K to the proportionality constant K2 (for example, K2 = K + ΔK). As shown in Figure 5, with system inductance Lg = Lg_Max and proportionality constant K = K2, the phase margin becomes greater than 0°, and the current control becomes stable. Therefore, the difference ΔId between the Id command value and the Id detection value is approximately 0, and the difference ΔId is stable. In other words, the control device 80 changes the proportionality constant K so that the oscillation degree of the currents Iu, Iv, and Iw flowing through the PFC coil 41 is smaller than the current oscillation degree.

[0056] At time t23, the control device 80 determines that the difference ΔId is stable and maintains the proportionality constant K=K2, setting the Id command value to the rated output equivalent. Even after the Id command value reaches the rated output equivalent, the difference ΔId between the Id command value and the Id detected value is approximately 0, indicating that the difference ΔId is stable. Furthermore, even if the Id detected value is about to fluctuate slightly, the Id detected value converges to the Id command value due to the action of the FB term calculated by the PI controller 85 in Figure 3.

[0057] The embodiment described in detail above has the following advantages.

[0058] The control device 80 calculates the FF term based on a system approximation voltage that is closer to the system voltages Vus, Vvs, Vws than to the input voltages Vu, Vv, Vw of the PFC coil 41. Therefore, fluctuations in the FF term and, consequently, oscillations in the d-axis current Id (Id detection value) can be suppressed compared to when the FF term is calculated based only on the input voltages Vu, Vv, Vw of the PFC coil 41. Thus, in the PFC converter 40 connected to the system 15, the stability of current control can be improved while enhancing control responsiveness.

[0059] The voltage drops Vua, Vva, and Vwa that occur between the PFC coil 41 and the system wiring 13 are correlated with the voltage drops ΔVus, ΔVvs, and ΔVws in the system wiring 13. Therefore, the approximate system voltage can be calculated using the input voltages Vu, Vv, and Vw of the PFC coil 41 and the voltage drops Vua, Vva, and Vwa that occur between the PFC coil 41 and the system wiring 13.

[0060] The voltage drops Vua, Vva, and Vwa in the common coil 51 are correlated with the voltage drops ΔVus, ΔVvs, and ΔVws in the system wiring 13. Therefore, the voltage drops ΔVus, ΔVvs, and ΔVws in the system wiring 13 can be approximated by predetermined voltage drops (Vua*K, Vva×K, Vwa*K) calculated by multiplying the voltage drops Vua, Vva, and Vwa in the common coil 51 by a proportionality constant K. The system voltages Vus, Vvs, and Vws are approximately equal to the sum of the input voltages Vu, Vv, and Vw of the PFC coil 41 and the voltage drops ΔVus, ΔVvs, and ΔVws in the system wiring 13. Therefore, by using the sum of the input voltages Vu, Vv, and Vw of the PFC coil 41 and the predetermined voltage drops as the approximate system voltage, fluctuations in the FF term and, consequently, oscillations in the d-axis current Id can be suppressed.

[0061] To obtain voltage drops correlated with the voltage drops ΔVus, ΔVvs, and ΔVws in the system wiring 13, the leakage inductance La of the common coil 51 is utilized. In other words, by deliberately using a common coil 51 with leakage inductance La, voltage drops Vua, Vva, and Vwa, which correlate with the voltage drops ΔVus, ΔVvs, and ΔVws in the system wiring 13, are generated in the common coil 51. Therefore, there is no need to add inductors or other components that generate voltage drops correlated with the voltage drops ΔVus, ΔVvs, and ΔVws in the system wiring 13, thus reducing the need for additional components.

[0062] The PFC converter 40 is equipped with a low-pass filter 89 that attenuates frequency components higher than the cutoff frequency. The control device 80 takes the sum of the input voltages Vu, Vv, Vw of the PFC coils 41 of each phase and a predetermined voltage drop (Vua*K, Vva×K, Vwa*K) calculated by multiplying the voltage drops Vua, Vva, Vwa after passing the voltage drops Vua, Vva, Vwa at the common coil 51 of each phase through the low-pass filter 89 by a proportionality constant K, and uses this sum as the system approximate voltage. Therefore, the system approximate voltage can be brought even closer to the system voltages Vus, Vvs, Vws, and the fluctuation of the FF term and, consequently, the oscillation of the output current can be suppressed.

[0063] By changing the proportionality constant K, the predetermined voltage drops (Vua*K, Vva×K, Vwa*K) can be changed, and consequently, the system approximation voltage used in calculating the FF term can be changed. Therefore, even if the system voltages Vus, Vvs, and Vws cannot be actually measured, the oscillation degree of the currents Iu, Iv, and Iw flowing through the PFC coils 41 of each phase can be reduced by changing the proportionality constant K, thereby stabilizing the current control.

[0064] If the oscillation degree of the current flowing through the PFC coil 41 is greater than a predetermined degree, the control device 80 can set the Id command value to the equivalent of 1 / 5 of the rated output. Then, after reducing the oscillation degree of the current flowing through the PFC coil 41 to below the predetermined degree, the control device 80 sets the Id command value to the equivalent of the rated output. Therefore, the PFC converter 40 can suppress the flow of a current equivalent to the rated output through the PFC coil 41 when the oscillation degree of the current flowing through the PFC coil 41 is greater than a predetermined degree. As a result, damage to the PFC coil 41 can be suppressed, and the PFC coil 41 can be miniaturized.

[0065] The control device 80 determines the Id command value not only based on the FF term but also on the FB term. This makes it easier to accurately control the d-axis current Id to the Id command value.

[0066] (Second Embodiment) The second embodiment will now be described, focusing on the differences from the first embodiment. In this embodiment, the method for determining the voltage drops Vua, Vva, and Vwa in the common coil 51 of each phase differs from that of the first embodiment. Parts identical to those in the first embodiment will be denoted by the same reference numerals for further explanation.

[0067] Instead of the voltage sensor 45 in Figure 2, the PFC converter 40 of this embodiment is equipped with a current sensor 46 as shown in Figure 8. Each phase current sensor 46 is connected between the connection point P between the X capacitor 53 and the common coil 51 and the connection point Q between the X capacitor 54 and the common coil 51 in each phase. Each phase current sensor 46 detects the currents Iua, Iva, and Iwa flowing through the common coil 51 of each phase.

[0068] The voltage drops Vua, Vva, and Vwa across the common coil 51 of each phase can be calculated based on the currents Iua, Iva, and Iwa flowing through the common coil 51 of each phase using the following formula. For example, Vua = La × dIua / dt. Similar to the first embodiment, the voltage drops ΔVus, ΔVvs, and ΔVws due to the system inductance Lg of each phase can be estimated by multiplying the voltage drops Vua, Vva, and Vwa across the common coil 51 of each phase by a proportionality constant K. For example, ΔVus can be estimated (expressed) as ΔVus = La × (dIua / dt) * K. Similar to the first embodiment, the control device 80 uses the voltage drops Vua, Vva, and Vwa after passing through the low-pass filter 89 as the voltage drops Vua, Vva, and Vwa across the common coil 51 of each phase.

[0069] As shown in Figure 9, the system voltage estimation unit 81 receives the input voltages Vu, Vv, Vw of the PFC coils 41 of each phase detected by the voltage sensors 42 of each phase, and the voltage drops Vua, Vva, Vwa (i.e., La × dIua / dt, La × dIva / dt, La × dIwa / dt) at the common coil 51 of each phase after passing through the low-pass filter 89. The system voltage estimation unit 81 estimates the system voltages Vus, Vvs, Vws of each phase by summing the detected input voltages Vu, Vv, Vw of the PFC coils 41 of each phase and the estimated voltage drops ΔVus, ΔVvs, ΔVws (corresponding to predetermined voltage drops) due to the system inductance Lg of each phase. For example, Vus is estimated using the formula Vus = Vu + La × (dIua / dt) * K. The estimated system voltages Vus, Vvs, and Vws for each phase correspond to approximate system voltages that are closer to the system voltages Vus, Vvs, and Vws for each phase than to the input voltages Vu, Vv, and Vw of the PFC coils 41 for each phase.

[0070] The parts other than the system voltage estimation unit 81 are the same as in the first embodiment. The PFC converter 40 of this embodiment can also produce the same effects as in the first embodiment.

[0071] (Third embodiment) The third embodiment will now be described, focusing on the differences from the first embodiment. In this embodiment, the method for estimating the voltage drops ΔVus, ΔVvs, and ΔVws due to the system inductance Lg of each phase differs from that of the first embodiment. Parts identical to those in the first embodiment will be denoted by the same reference numerals for further explanation.

[0072] Instead of the voltage sensor 45 in Figure 2, the PFC converter 40 of this embodiment is equipped with a voltage sensor 47 as shown in Figure 10. The u-phase and v-phase voltage sensors 47 detect the input voltages Vub and Vvb of the u-phase and v-phase common coils 51. The input voltage Vwb of the w-phase common coil 51 can be calculated using the formula Vwb = -(Vub + Vvb). Furthermore, the PFC converter 40 does not have a w-phase voltage sensor 42. The input voltage Vw of the w-phase PFC coil 41 can be calculated using the formula Vw = -(Vu + Vv).

[0073] Here, the difference Vub-Vu, Vvb-Vv, Vwb between the input voltages Vub, Vvb, Vwb of the common coil 51 of each phase and the input voltages Vu, Vv, Vw of the PFC coil 41 of each phase, is equal to the voltage drop across the common coils 51 and 52 connected between the system wiring 13 and the PFC coil 41. Note that the common coils 51 and 52 constitute a predetermined electrical circuit.

[0074] The voltage drops Vub-Vu, Vvb-Vv, Vwb-Vw (corresponding to circuit voltage drops) in a given electrical circuit are correlated with the voltage drops ΔVus, ΔVvs, ΔVws in the system wiring 13. Therefore, the voltage drops ΔVus, ΔVvs, ΔVws in the system wiring 13 can be estimated (approximated) by the predetermined voltage drop calculated by multiplying the voltage drops Vub-Vu, Vvb-Vv, Vwb-Vw in the given electrical circuit by a proportionality constant K. For example, ΔVus = (Vub-Vu) * K. The optimal value of the proportionality constant K is determined by the relationship between the inductance Lg of the system wiring 13 and the inductance of the given electrical circuit. The control device 80 uses the input voltages Vub, Vvb, Vwb after passing through the low-pass filter 89 as the input voltages Vub, Vvb, Vwb of the common coil 51 of each phase.

[0075] As shown in Figure 11, the system voltage estimation unit 81 receives the input voltages Vu, Vv, Vw of the PFC coils 41 of each phase detected by the voltage sensors 42 of each phase, and the input voltages Vub, Vvb, Vwb of the common coils 51 of each phase after passing through the low-pass filter 89. The system voltage estimation unit 81 estimates the system voltages Vus, Vvs, Vws of each phase by summing the detected input voltages Vu, Vv, Vw of the PFC coils 41 of each phase and the estimated voltage drops ΔVus, ΔVvs, ΔVws (corresponding to predetermined voltage drops) due to the system inductance Lg of each phase. For example, it is estimated using Vus = Vu + (Vub - Vu) * K. The estimated system voltages Vus, Vvs, Vws of each phase correspond to approximate system voltages that are closer to the actual system voltages Vus, Vvs, Vws of each phase than to the input voltages Vu, Vv, Vw of the PFC coils 41 of each phase.

[0076] The parts other than the system voltage estimation unit 81 are the same as in the first embodiment. The PFC converter 40 of this embodiment can also produce the same effects as in the first embodiment. Furthermore, with the PFC converter 40 of this embodiment, the number of voltage sensors 42, 47 can be reduced compared to the number of voltage sensors 42, 45 in the first embodiment.

[0077] Furthermore, the first to third embodiments can also be implemented with the following modifications. Parts identical to those in the first to third embodiments are denoted by the same reference numerals, and their descriptions are used accordingly.

[0078] The procedure for controlling the Id command value shown in Figure 4 can also be changed as shown in Figure 12. That is, the control device 80 is set to proportionality coefficient K=K3 and the Id command value is set to a current command value equivalent to, for example, 1 / 5 of the rated output (S10A). The process in S11 is the same as the process in S11 in Figure 4. Subsequently, the value obtained by subtracting the coefficient change amount ΔK from the proportionality coefficient K is set as the new proportionality coefficient K (S12A). The processes in S13 and S14 are the same as the processes in S13 and S14 in Figure 4.

[0079] The procedure for controlling the Id command value shown in Figure 4 can also be changed as shown in Figure 13. That is, the control device 80 is set to proportionality constant K=0 and the Id command value is set to a current command value equivalent to the rated output, for example (S10B). The processing in S11 to S13 is the same as the processing in S11 to S13 in Figure 4. The processing in S14 in Figure 4 is omitted. Even with this configuration, fluctuations in the FF term, and consequently oscillations in the d-axis current Id, can be suppressed more effectively than when the FF term is calculated based only on the input voltages Vu, Vv, Vw of the PFC coil 41. Therefore, in the PFC converter 40 connected to the system 15, the stability of current control can be improved while enhancing control responsiveness.

[0080] The low-pass filter 89 can also be omitted. Even in this case, fluctuations in the FF term, and consequently oscillations in the d-axis current Id (Id detection value), can be suppressed compared to when the FF term is calculated based only on the input voltages Vu, Vv, Vw of the PFC coil 41. Therefore, in the PFC converter 40 connected to the system 15, the stability of current control can be improved while enhancing control responsiveness.

[0081] As shown in Figure 14, the FB term can also be omitted when determining the d-axis voltage command value Vd*. This configuration also suppresses fluctuations in the FF term and, consequently, oscillations in the d-axis current Id, compared to when the FF term is calculated based only on the input voltages Vu, Vv, and Vw of the PFC coil 41. [Explanation of Symbols]

[0082] 13...System wiring, 15...System, 40...PFC converter, 41...PFC coil, 51...Common coil, 81...System voltage estimation unit, 82...dq conversion unit, 86...Adder, 87...Adder.

Claims

1. A power converter (40) connected to a power grid (15) via a grid wiring (13) that converts AC power to DC power and outputs it, PFC coil (41) and A common mode coil (51) having leakage inductance is connected between the PFC coil and the system wiring, The system includes a voltage command unit (81, 82, 86) that determines a voltage command value for controlling the output current to a current command value based on a predetermined feedforward term, The voltage command unit calculates a system approximate voltage that is closer to the system voltage of the power system than the input voltage of the PFC coil, based on the input voltage of the PFC coil and the voltage drop in the common mode coil, and calculates the predetermined feedforward term based on the calculated system approximate voltage, in this power conversion device.

2. An X capacitor (53) has one end connected to the connection point between the aforementioned system wiring and the common mode coil, and the other end connected to the neutral point. It includes a low-pass filter (89) that attenuates frequency components higher than the cutoff frequency, The power conversion device according to claim 1, wherein the voltage command unit uses the sum of the input voltage of the PFC coil and a predetermined voltage drop calculated by multiplying the voltage drop in the common mode coil by a proportionality constant after passing it through the low-pass filter as the approximate system voltage.

3. An X capacitor (53) has one end connected to the connection point between the aforementioned system wiring and the common mode coil, and the other end connected to the neutral point. A low-pass filter (89) that attenuates frequency components higher than the cutoff frequency, A predetermined electrical circuit (51, 52) is connected between the PFC coil and the system wiring and includes the common mode coil, The power conversion device according to claim 1, wherein the voltage command unit uses the sum of the input voltage of the PFC coil and a predetermined voltage drop calculated by multiplying the circuit voltage drop obtained by subtracting the input voltage of the PFC coil from the voltage drop after the input voltage of the common mode coil has passed through the low-pass filter by a proportionality constant, as the approximate system voltage.

4. The power conversion device according to claim 2 or 3, wherein the voltage command unit changes the proportionality coefficient so that the degree of oscillation of the current flowing through the PFC coil is smaller than the current degree of oscillation.

5. The power conversion device according to claim 4, wherein, when the current command value is set to a first current, the voltage command unit increases the proportionality constant from the current proportionality constant to reduce the current oscillation degree of the current flowing through the PFC coil to below the predetermined degree, and then sets the current command value to a second current that is greater than the first current.

6. The power conversion device according to any one of claims 1 to 3, wherein the voltage command unit determines the voltage command value based on the predetermined feedforward term and the predetermined feedback term.

Images