POWER CONVERTER

Abstract

An electrical power converter for converting electrical energy supplied to a rotating electrical machine (10) having a first winding group (11) and a second winding group (12), the power converter comprising: a first inverter (21) having a first switching element (211-216) associated with each of a plurality of phases in the first winding group (11); a second inverter (22) having a second switching element (221-226) associated with each of a plurality of phases in the second winding group (12); and a controller (40) including an instruction calculator (415, 425) which calculates a first voltage instruction value applied to the first winding group (11) and a second voltage instruction value applied to the second winding group (12), wherein an application voltage to the first winding group (11) is controlled (i) in a first period such that a first neutral point voltage (Vn1) shifts below a central output value (VM) of an outputtable range and a second neutral point voltage (Vn2) shifts above the central output value (VM) of the outputtable range, and (ii) in a second period such that the first neutral point voltage (Vn1) shifts above the central output value (VM) of the outputtable range and the second neutral point voltage (Vn2) shifts below the central output value (VM) of the outputtable range, the instruction calculator (415, 425) calculates the first instruction value and the second instruction value for enabling switching between the first period and the second period after a lapse of a switching period (Pc) which is variably calculated based on a preset physical quantity, characterized in that the preset physical quantity comprises a rotational speed of the rotating electrical machine (10), and the switching period (Pc) is calculated as a longer period when the rotational speed of the rotating electrical machine (10) is low and as a shorter period when the rotational speed is high.

Description

The present invention generally relates to a power converter for converting an electric energy.Generally, the power converter for converting an electric power supplied to a motor having two sets of winding groups is known in the art. For example, in a patent document JP 2011-188 674 A (corresponding to DE 10 211 001 171 A1), for the sake of ease of distributing heat generation to each of the winding groups, the power converters have two inverters associated with the two sets of winding groups, and the shift direction of the duty instruction signal regarding the driving of each of the inverters is switched at preset intervals.Further power converters are also known from U.S. Pat. No. 8 436 573 B2, U.S. Pat. No. 2014 / 0 253 006 A1, U.S. Pat. No. 2005 / 0 194 925 A1, DE 10 2007 021 368 B4 and U.S. Pat. No. 8 604 730 B2.As disclosed in JP 2011-188 674 A, when the shift direction of the duty instruction signal is switched, such switching may cause torque ripple as well as vibration and noise.In one aspect of the present invention, a power converter of the present invention that converts electric power supplied to a rotating electric machine having a first winding group and a second winding group includes a first inverter, a second inverter, and a controller.The first inverter has a first switching element associated with each of a plurality of phases in the first winding group.The second inverter has a second switching element associated with each of a plurality of phases in the second winding group.The controller includes an instruction computer. The instruction calculator calculates a first voltage instruction value applied to the first winding group and a second voltage instruction value applied to the second winding group.An application voltage to the first winding group is controlled in a first period and a second period in the following manner. That is, (i) in the first period, a first neutral point voltage applied to the first winding group shifts below a central output value of an outputable range of the inverter, and a second neutral point voltage applied to the second winding group shifts above the central output value of the outputable range of the inverter. Also, (ii), in the second period, the first neutral point voltage shifts above the central output value of the outputable range, and the second neutral point voltage shifts below the central output value of the outputable range of the inverter.Further, the instruction calculator calculates the first instruction value and the second instruction value for enabling switching between the first period and the second period after elapse (i.e., at an interval) of a switching period that is variably calculated based on a preset physical quantity.In the present invention, switching between the two states is performed (i.e., (i) one state in which the first neutral point voltage is shifted below the central output value and the second neutral point voltage is shifted above the central output value, and (ii) the other state in which the first neutral point voltage is shifted above the central output value and the second neutral point voltage is shifted below the central output value is performed). Thereby, the non-uniformity of heat generation in the switching elements is reduced.Objects, features and advantages of the present invention will become more apparent from the following detailed description made with reference to the accompanying drawings, in which: FIG. 1 is a schematic diagram of an electric power steering apparatus in a first embodiment of the present invention; FIG. 2 is a circuit diagram of an electrical configuration of a power converter in the first embodiment of the present invention; FIG. 3 is a block diagram of a controller in the first embodiment of the present invention; FIG. 4 is a flowchart of an instruction calculation process in the first embodiment of the present invention; FIG. 5 is a graph of a relationship of a rotational speed and a switching period in the first embodiment of the present invention; FIG. 6A / B is time charts showing a relationship of the rotational speed and the switching period in a second embodiment of the present invention; FIG. 7 is a graph showing a relationship of an electric current and the switching period in a third embodiment of the present invention; FIG. 8 is a graph showing a relationship of a voltage and the switching period in a fourth embodiment of the present invention; FIG. 9 is a graph of a relationship of a torque and the shift period in a fifth embodiment of the present invention; FIG. 10 is a graph showing a relationship of an inverter temperature and the switching period in a sixth embodiment of the present invention; FIG. 11A / B is diagrams of a relationship of a resonance frequency and a switching period in a seventh embodiment of the present invention; and FIG. 12 is a graph showing a relationship of the rotational speed and the switching period in an eighth embodiment of the present invention.Hereinafter, the power converter of the present invention will be described based on the drawings. In the following embodiments, like parts have like reference numerals, and the description of the same part will not be repeated.(First Embodiment)The power converter according to the first embodiment of the present invention will be described based on FIGS. 1-5.A power converter 1 of the present embodiment is disposed in a vehicle, not illustrated, together with an engine 10, and is applied to an electric power steering apparatus 5 for assisting a steering operation performed by the driver with the engine 10, which is also described as a rotating electric machine.FIG. 1 shows an overall configuration of a steering system 90 provided with the electric power steering apparatus 5.The steering system 90 includes a steering wheel (i.e., a steering wheel) 91, a steering shaft 92, a pinion gear 96, a rack shaft 97, wheels 98, the electric power steering apparatus 5, and the like.The steering wheel 91 is connected to the steering axle 92. On the steering shaft 92, a torque sensor 94 that detects the steering torque input from the steering wheel 91 to the steering shaft 92 by the driver's operation is disposed. The pinion gear 96 is disposed at the tip of the steering shaft 92, and the pinion gear 96 engages with the rack shaft 97. A pair of wheels 98 are connected to both ends of the rack shaft 97 via a tie rod, etc.Thereby, when the driver rotates the steering wheel 91, the steering shaft 92 connected to the steering wheel 91 rotates, the rotational motion of the steering shaft 92 is converted into the linear motion of the rack shaft 97 by the pinion gear 96, and the wheels 98 are steered by an angle depending on the amount of displacement of the rack shaft 97.The electric power steering device 5 includes a motor 10 that outputs an assist torque for assisting a steering operation of the steering wheel 91 by the driver, a power inverter 1 for driving control of the motor, a reduction gear 9 that reduces a rotation speed of the motor 10 for outputting the rotation to the steering shaft 92 or the rack shaft 97, and the like.The motor 10 is driven by receiving a supply of electric power from a battery 30 (see FIG. 2 ) which is a DC power supply, and rotates the reduction gear 9 back and forth (i.e., reciprocally).As shown in FIG. 2, the motor 10 is a three-phase brushless motor, and has a rotor and a stator (not illustrated). The rotor is a cylindrical component, and the permanent magnet is attached to the surface and has the magnetic pole. The stator has winding groups 11 and 12 wound on the stator. The first winding group 11 includes a U1 winding 111, a V1 winding 112, and a W1 winding 113. The second winding group 12 includes a U2 winding 121, a V2 winding 122, and a W2 winding 123.The power converter 1 includes a first inverter 21, a second inverter 22, a first current detector 26, a second current detector 27, a rotation angle sensor 29, a first power relay 31, a second power relay 32, a first capacitor 33, a second capacitor 34, a drive circuit (pre-drive) 35, a controller 40, and the like.The first inverter 21 has six first switching elements 211- 216 (the "switching element" may be referred to as the "SW element" hereinafter) and switches the power supply to the coils 111, 112, and 113 of the first winding group 11.The drain of the high potential side SW elements 211, 212 and 213 provided on the high potential side is connected to the positive electrode of the battery 30 via a first high side bus line 218, respectively.The source of the high potential side SW elements 211, 212, and 213 is connected to the drain of the low potential side SW elements 214, 215, and 216, respectively, which are provided on the low potential side.The source of the low potential side SW elements 214, 215 and 216 is connected to the negative electrode of the battery 30 via a first low side bus line 219, respectively.The node of the high potential side SW elements 211, 212, and 213 and the low potential side SW elements 214, 215, and 216 is connected to one end of the U1 winding 111, the V1 winding 112, and the W1 winding 113, respectively.The second inverter 22 has six second SW elements 221- 226 and switches the power supply of the coils 121, 122, and 123 of the second winding group 12.The drain of the high potential side SW elements 221, 222, and 223 is connected to the positive electrode of the battery 30, respectively, via a second high side bus line 228.The source of the high potential side SW elements 221, 222, and 223 is connected to the drain of the low potential side SW elements 224, 225, and 226, respectively.The source of the low potential side SW elements 224, 225, and 226 is connected to the negative electrode of the battery 30, respectively, via a second low side bus line 229.The node of the high potential side SW elements 221, 222, and 223 and the low potential side SW elements 224, 225, and 226 is connected to one end of the U2 winding 121, the V2 winding 122, and the W2 winding 123, respectively.Although the SW elements 211-216, 221-226 of the present embodiment are all metal oxide semiconductor field effect transistors (MOSFETs), an insulated gate bipolar transistor (IGBT), a thyristor, etc. may also be used as the SW elements.The first current detector 26 includes current detection elements 261, 262 and 263.The current detection element 261 is disposed on the low potential side of the SW element 214, and detects the electric current provided to the U 1 winding 111. The current detection element 262 is disposed on the low potential side of the SW element 215 and detects the electric current supplied to the V 1 winding 112. The current detection element 263 is disposed on the low potential side of the SW element 216, and detects the electric current supplied to the W 1 winding 113. The electric current detection values detected by the first current detector 26 regarding the electric current provided to the respective phases of the first winding group 11 will be referred to as first current detection values lu 1, Iv 1, and Iw 1 hereinafter.The second current detector 27 includes current detection elements 271, 272, and 273.The current detection element 271 is disposed on the low potential side of the SW element 224 and detects the electric current supplied to the U2 winding 121. The current detection element 272 is disposed on the low potential side of the SW element 225, and detects the electric current provided to the V 2 winding 122. The current detection element 273 is disposed on the low potential side of the SW element 226 and detects the electric current supplied to the W 2 winding 123. The electric current detection values detected by the second current detector 27 regarding the electric current provided to the respective phases of the second winding group 12 will be referred to as second current detection values Iu 2, Iv 2, and Iw 2 hereinafter.The current detection elements 261- 263, 271- 273 of the present embodiment are shunt resistors.The rotation angle sensor 29 detects the rotation angle of the motor 10.An electric angle θ of the motor 10 detected by the rotation angle sensor 29 is output to the controller 40.The first power relay 31 can cut off the electric power supply from the battery 30 to the first inverter 21. Also, the second power relay 32 may cut off the electric power supply from the battery 30 to the second inverter 22.Although the power relays 31 and 32 are the same MOSFET as the SW element 211 and the like, the power relays 31, 32 may be IGBTs, mechanical relays, or the like.Further, when the power relays 31 and 32 are realized as MOSFETs, etc., it is preferable to use a reverse connection protection relay (not shown) connected in series with the power relays 31 and 32 for protecting the power relays 31, 32 from accidental reverse connection of the battery 30 or the like.The first capacitor 33 is connected in parallel to the battery 30 and the first inverter 21. The second capacitor 34 is connected in parallel to the battery 30 and the second inverter 22.The capacitors 33 and 34 assist in supplying the electric power of the inverters 21 and 22 and control the noise component (e.g., the peak current) by storing electric charge.In the present embodiment, the first winding group 11, and the first inverter 21, the first current detector 26, the first power relay 31, and the first capacitor 33 for controlling the power supply of the first winding group 11 are collectively referred to as a "first system 101"; and the second winding group 12 as well as the second inverter 22, the second current detector 27, the second power relay 32, and the second capacitor 34 for controlling the power supply of the second winding group 12 are collectively referred to as a "second system 102". Hereinafter, the numeral "1" may be used as an index of the component involving control of the first system 101, and the numeral "2" may be used as an index of the component involving control of the second system 102.The controller 40 performs control of the power converter 1 as a whole, and is constituted by a microcomputer, etc., which performs various operations and calculations. Each process in the controller 40 may be a software process by execution of a pre-stored program by a central processing unit (CPU), or may be a hardware process realized by an electronic circuit dedicated to such a process.The controller 40 generates a control signal that controls ON and OFF switching of the SW elements 211- 216, 221- 226 based on the steering torque obtained from the torque sensor 94, the electric angle θ obtained from the rotation angle sensor 29, etc. The generated control signal is output to the gate of the SW elements 211- 216, 221- 226 via the drive circuit (pre-drive) 35.As shown in FIG. 3, the controller 40 has a first instruction calculator 41 related to controlling the first system 101 and a second instruction calculator 42 related to controlling the second system 102.Since the function block of the first instruction calculator 41 to which a numeral 41 xis assigned and the function block of the second instruction calculator 42 to which a numeral 42 xis assigned are substantially the same (e.g., 411=1211), only the function blocks 41 xwill be described below.The first instruction calculator 41 includes a 3-to-2-phase converter 410, subtractors 411 and 412, and a control part 413, a 2-to-3-phase converter 414, and a duty calculator 415.The 3-to-2-phase converter 410 performs dq conversion of the electric current detection values lu 1, Iv 1, and Iw 1 detected by the first current detector 26 and corrected as to an offset error, a gain error, etc., and calculates a d-axis electric current detection value Id 1 and a q-axis electric current detection value Iq 1.The d-axis subtractor 411 calculates a d-axis electric current deviation ΔId 1, which is a deviation between a d-axis electric current instruction value Id* 1 and the d-axis electric current detection value Id 1.The q-axis subtractor 412 calculates a q-axis electric current deviation ΔIq 1 which is a deviation between a q-axis electric current instruction value Iq* 1 and the q-axis electric current detection value Iq 1.The control part 413 calculates a d-axis voltage instruction value Vd*1 and a q-axis voltage instruction value Vq*1 by PI calculation etc. so that the electric current deviations ΔId1 and ΔIq1 converge toward zero, respectively.The 2-to-3-phase converter 414 performs inverted dq conversion of the d-axis voltage instruction value Vd*1 and the q-axis voltage instruction value Vq*1 based on the electrical angle θ, and calculates voltage instruction values Vu*1, Vv*1, and Vw*1.The duty calculator 415 calculates duty instruction values Du 1, Dv 1, and Dw 1 based on the voltage instruction values Vu* 1, Vv* 1, and Vw* 1.A duty calculator 425 of the second instruction calculator 42 calculates duty instruction values Du 2, Dv 2, and Dw 2 based on the voltage instruction values Vu* 2, Vv* 2, and Vw* 2.Hereinafter, the calculation of the first duty instruction values Du 1, Dv 1, Dw 1, which are instructions / commands regarding control of the first inverter 21, and the calculation of the second duty instruction values Du 2, Dv 2, Dw 2, which are instructions / commands regarding control of the second inverter 22, are respectively described in full detail.In the present embodiment, based on a first neutral point voltage Vn 1 which is a neutral point voltage of the voltages Vu 1, Vv 1, Vw 1 applied to the first winding group 11 and a second neutral point voltage Vn 2 which is a neutral point voltage of the voltages Vu 2, Vv 2, Vw 2 applied to the second winding group 21, the duty instruction values Du 1, Dv 1, Dw 1, Du 2, Dv 2, Dw 2 are calculated so that one of Vn 1 and Vn 2 is shifted above a center output value VM which is a center value of an outputable voltage range and the other of Vn 1 and Vn 2 is shifted below the center output value VM. In the present embodiment, a first period P 1 is defined as a period in which the first neutral point voltage Vn 1 is shifted below the center output value VM and the second neutral point voltage Vn 2 is shifted above the center output value VM; and a second period P 2 is defined as a period in which the first neutral point voltage Vn 1 is shifted above the center output value VM and the second neutral point voltage Vn 2 is shifted below the center output value VM.The duty instruction values Du 1, Dv 1, Dw 1, Du 2, Dv 2, and Dw 2 are calculated to be within a preset value range in consideration of a required electric current detection time of the current detectors 26 and 27. According to the present embodiment, a lower limit value DL of the duty instruction values Du 1, Dv 1, Dw 1, Du 2, Dv 2, and Dw 2 is set to 4 [%], and an upper limit value DH is set to 93 [%]. Further, the duty conversion value of the center output value VM is set as a center output duty DM. The center initial setting period DM is 50 [%].By shifting one of the first neutral point voltage Vn 1 and the second neutral point voltage Vn 2 above the central output value VM and shifting the other below the value VM, the vector generation timing of the first inverter 21 (that is, when a zero voltage vector and an effective voltage vector are generated in the first inverter 21) and the vector generation timing of the second inverter 22 (that is, when a zero voltage vector and an effective voltage vector are generated in the second inverter 22) become different timings, respectively. In this way, the ripple current of the capacitors 33 and 34 is reduced.When the first neutral point voltage Vn 1 is shifted below the central output value VM, a longer ON period occurs for the low potential side SW elements 214- 216 than for the high potential side SW elements 211- 213 in the first inverter 21.Similarly, when the second neutral point voltage Vn 2 is shifted below the central output value VM, a longer ON period occurs for the low potential side SW elements 224- 226 than for the high potential side SW elements 221- 223 in the second inverter 22.When the first neutral point voltage Vn 1 is shifted above the central output value VM, a longer ON period occurs for the high potential side SW elements 211- 213 than for the low potential side SW elements 214- 216, in the first inverter 21.Similarly, when the second neutral point voltage Vn 2 is shifted above the central output value VM, a longer ON period occurs for the high potential side SW elements 221- 223 than for the low potential side SW elements 224- 226 in the second inverter 22.Therefore, in the present embodiment, in order to avoid overheating of part of the SW elements 211- 216, 221- 226, two states are switched (i.e., a first state in which the first neutral point voltage Vn 1 is shifted downward and the second neutral point voltage Vn 2 is shifted upward, and a second state in which the first neutral point voltage Vn 1 is shifted upward and the second neutral point voltage Vn 2 is shifted downward). Thereby, generation of heat from the SW elements 211- 216, 221- 226 is distributed to two systems 101 and 102.In such a case, when the rotation speed of the motor 10 is low, the SW elements 211- 216, 221- 226 may easily have mutually nonuniform ON times and OFF times, thereby having a large temperature difference therebetween. On the other hand, when the rotation speed of the motor 10 is high, it may be difficult to occur for such non-uniformity of ON / OFF times. Therefore, according to the present embodiment, the switching period Pc is variably defined based on the rotational speed of the motor 10. In this case, the rotation speed of the motor 10 may be the number of revolutions in a unit of "rpm" or the like, or may be a rotation angle speed or the like.An instruction calculation process of the present embodiment will be described based on the flowchart shown in FIG. 4. The process of FIG. 4 is performed at predetermined intervals by the duty calculator 415. The calculation of the first duty instruction values Du 1, Dv 1, Dw 1 and the calculation of the second duty instruction values Du 2, Dv 2, Dw 2 are the same process, and therefore a description of the calculation of Du 2-Dw 2 is omitted.In the first step (S) S 101, among the pre-shift duty instruction values Du 1_ b, Dv 1_ b, Dw 1_ b, which are the duty conversion values of the voltage instruction values Vu* 1, Vv* 1, Vw* 1, the largest value is referred to as "D 1_max", and the smallest value is referred to as "D 1_min".In S 102, the switching period Pc is calculated based on the rotational speed of the motor 10. The switching period Pc is calculated as a longer period when the rotation speed of the motor 10 is low as shown in FIG. 5 and as a shorter period when the rotation speed is high.In S 103, it is determined whether a count value Cnt of a counter that counts a elapse time from switching of the shift directions is larger than a count determination value Cth that is a predetermined count value depending on the switching period Pc.When the count value Cnt is determined to be equal to or less than the count determination value Cth (that is, when the switching period Pc has not elapsed after the switching of the shifting directions) (S 103: NO), the process proceeds to S 106. The shift flag is maintained as the previous value at such time.When it is determined that the count value Cnt is larger than the count determination value Cth (that is, when the switching period Pc has elapsed after the switching of the shift directions) (S 103: YES), the process proceeds to S 104.The shift flag is changed in S104. Specifically, when the previous value of the shift flag is "an up shift", the shift flag is changed to "a down shift". When the previous value of the shift flag is "the down shift", the shift flag is changed to "the up shift".The count value Cnt is initialized in S 105.The count value Cnt is incremented in S 106.In S107, it is determined whether the shift flag indicates the down shift. When it is determined that the shift flag does not indicate the down shift (that is, when the shift flag is the up shift) (S 107: NO), the process proceeds to S 111. When it is determined that the shift flag indicates the down shift (S 107: YES), the process proceeds to S 108.In S 108, the shift amount Sft is calculated. The shift amount Sft is a difference between the duty conversion value Dn 1 and the center output duty DM of the neutral point voltage Vn 1 in the present amplitude, and is represented by an equation (1). Hereinafter, an "equation" may indicate either an equation or an inequality.The shift amount Sft is limited in S 109.The upper limit value of the displacement amount Sft is set to 0. The lower limit value of the shift amount Sft is represented by an equation (2).In S 110, the first duty instruction values Du 1, Dv 1, Dw 1 are calculated. The first duty instruction values Du 1, Dv 1, Dw 1 are represented by equations (3-1) to (3-3).In S 111 performed when the shift flag is the upward shift (S 107: NO), the shift amount Sft is calculated by an equation (1) just like S 108.The displacement amount Sft is limited in S 112. The lower limit value of the displacement amount Sft is set to 0. The upper limit value of the shift amount Sft is represented by an equation (4).In S 113, the first duty instruction values Du 1, Dv 1, Dw 1 are calculated. The first duty instruction values Du 1, Dv 1, Dw 1 are given by equations (5- 1) to (5- 3).According to the present embodiment, a switching cycle or period is shortened when the rotation speed of the motor 10 is low. This reduces the non-uniformity of heat generation between the elements.For example, when an integrated value of the electric current is used as a parameter regarding the switching between the upward shift and the downward shift, the increase of the integrated value of the electric current is slow when the rotation speed of the motor 10 is low. Here, the switching takes more time to reach the threshold, making it possible to have non-uniform heat generation among many SW elements.In consideration of the above, the shift cycle is configured to be short when the rotation speed of the motor 10 is low in the above embodiment, as mentioned above. Thereby, the non-uniformity of heat generation in the SW elements is appropriately reduced.Further, when the rotation speed of the motor 10 is high, the switching cycle is extended to be a longer period because it is difficult to cause the nonuniformity of heat generation in the SW elements under such a condition. This reduces the torque ripple accompanying the switching of the upward shift and the downward shift.According to the present embodiment, the shift cycle is variably set based on the rotational speed of the motor 10. The rotational speed of the motor 10 can be regarded as an "instantaneous value of a physical quantity concerning the driving of the motor", which is easier to calculate as compared with the integrated value of the electric current that requires more memory and other calculation resources for storing and calculating an integrated value of the electric current and the like. That is, control of the switching period based on the engine rotation speed enables reduction of system resources such as a storage area and the like.As described above in full detail, the power converter 1 of the present embodiment converts the electric power for the motor 10 having the first winding group 11 and the second winding group 12, and the inverter 1 is provided with the first inverter 21, the second inverter 22, and the controller 40.The first inverter 21 includes the first SW elements 211- 216 provided for each of the plurality of phases of the first winding group 11 in a corresponding manner.The second inverter 22 includes the second SW elements 221- 226 respectively provided for each of the plurality of phases of the second winding group 12.The controller 40 has the duty calculators 415 and 425 that calculate (i) the first duty instruction values Du 1, Dv 1, Dw 1 concerning the voltage applied to the first winding group 11 and (ii) the second duty instruction values Du 2, Dv 2, and Dw 2.Further, in the first period P 1, the first neutral point voltage Vn 1 applied to the first winding group 11 shifts below the center output value VM of the outputable range of the inverter ( 21), and the second neutral point voltage Vn 2 applied to the second winding group 12 shifts above the center output value VM of the outputable range of the inverter ( 22).Further, in the second period P 2, the first neutral point voltage Vn 1 shifts above the center output value VM, and the second neutral point voltage Vn 2 shifts below the center output value VM.When the switching period Pc, which is variably calculated based on the preset physical quantity, has elapsed, the duty calculators 415 and 425 calculate the first duty instruction values Du 1, Dv 1, Dw 1 and the second duty instruction values Du 2, Dv 2, Dw 2 so that the first period P 1 and the second period P 2 are switched.According to the present embodiment, two states are switched (i.e., alternated) in an interval of the switching period Pc (that is, one state in which the first neutral point voltage Vn 1 is shifted above the center output value VM and the second neutral point voltage Vn 2 is shifted below the center output value VM; and the other state in which the first neutral point voltage Vn 1 is shifted below the center output value VM and the second neutral point voltage Vn 2 is shifted above the center output value VM). In this way, non-uniform heat generation among the SW elements 211- 216 and 221- 226 is reduced.Further, based on the preset physical quantity, the switching period Pc is variably set. Thereby, the vibration and the noise accompanying the switching between the first period P 1 and the second period P 2 are reduced.In the present embodiment, the preset physical quantity is the rotation speed of the motor 10.Therefore, the nonuniformity of heat generation in the SW elements is reduced when the rotation speed of the motor 10 is low. Further, the torque ripple accompanying the switching at a time when the rotational speed of the motor 10 is high is reduced.In the present embodiment, the duty calculators 415 and 425 correspond to an "instruction calculator". The first duty instruction values Du 1, Dv 1, Dw 1 correspond to a "first voltage instruction value", and the second duty instruction values Du 2, Dv 2, Dw 2 correspond to a "second voltage instruction value", The rotational speed of the motor 10 corresponds to a "preset physical quantity".(Second Embodiment)The second embodiment of the present invention will be described based on FIG. 6A / B.According to the present embodiment, the duty calculator 415 of the first instruction calculator 41 performs a modulation process of the voltage instruction values Vu*1, Vv*1, Vw*1, and calculates the first duty instruction values Du 1, Dv 1, Dw 1, and the duty calculator 425 of the second instruction calculator 42 performs the modulation process of the voltage instruction values Vu*2, Vv*2, Vw*2, and calculates the second duty instruction values Du 2, Dv 2, and Dw 2.Here, the modulation process will be described.According to the present embodiment, a "low-end setting modulation process" is defined as a modulation process in which one of the smallest among the first duty instruction values Du 1, Dv 1, and Dw 1 is set to a preset value. Similarly, the "low-end setting modulation process" is defined as a modulation process in which one of the smallest among the second duty instruction values Du 2, Dv 2, and Dw 2 is set to a preset value. By performing the low-end modulation process, the neutral point voltages Vn 1 and Vn 2 are shifted below the center output value VM.Further, a "high-end setting modulation process" is defined as a modulation process in which one of the highest ones among the first duty instruction values Du 1, Dv 1, and Dw 1 is set to a preset value. Similarly, the "high end setting modulation process" is defined as a modulation process in which one of the highest ones among the second duty instruction values Du 2, Dv 2, and Dw 2 is set to a preset value. By performing the high-end modulation process, the neutral point voltages Vn 1 and Vn 2 are shifted over the center output value. That is, according to the present embodiment, by performing the high-end adjusting modulation process, the neutral point voltages Vn 1 and Vn 2 are shifted over the center output value VM.By performing the modulation process, the voltage utilization efficiency is improved.According to the present embodiment, one of the first duty instruction values Du 1, Dv 1, Dw 1 and the second duty instruction values Du 2, Dv 2, Dw 2 performs the low-end setting modulation process, and the other performs the high-end setting modulation process. Thereby, one of the first neutral point voltage Vn 1 and the second neutral point voltage Vn 2 is shifted below the center output value VM, and the other is shifted above the center output value VM.According to the present embodiment, the switching period Pc is variably set based on the rotational speed of the motor 10, just as in the above-mentioned embodiment. The calculation method of the switching period Pc is the same as that of the above-mentioned embodiment.In the present embodiment, as shown in FIG. 6A / B, the switching period Pc 1 at a time when the rotation speed of the motor 10 is low is shorter than the switching period Pc 2 when the rotation speed is high. That is, Pc1<Pc2. In this way, the same effects as in the above-mentioned embodiment are obtained.(Third Embodiment)The third embodiment of the present invention will be described based on FIG. 7.The calculation method of the switching period Pc is different from the above embodiments in the third to sixth embodiments. The other points which are the same as in the above embodiments (i.e., the modulation process) may be performed just as in the first embodiment or may not be performed just as in the second embodiment.According to the present embodiment, the switching period Pc is calculated based on the current detection values lu1, Iv1, Iw1, Iu2, Iv2, and Iw2 detected by the current detectors 26 and 27.The electric current supplied to the switching elements 211- 216, 221- 226 is large when the current detection values Iu 1, Iv 1, Iw 1, Iu 2, Iv 2, and Iw 2 are large. Therefore, the switching period Pc is shortened when, for example, the amplitude of the current detection values lu1, Iv1, Iw1, Iu2, Iv2, and Iw2 becomes large, as shown in FIG. 7.Further, as a value concerning the electric current flowing to the winding groups 11 and 12, instead of using the current detection values lu 1, Iv 1, Iw 1, Iu 2, Iv 2, and Iw 2, the switching period Pc may be calculated using the electric current instruction value or the electric current estimation value.The electric current detection values and / or the electric current instruction values used for such calculation may be derived from each of the three phases or may be derived from each of the d / q axes.Further, the switching period Pc may be calculated based on a value concerning the electric current flowing in only one of the first winding group 11 and the second winding group 12.In the present embodiment, the preset physical quantity is at least one of the electric current flowing in only one of the first winding group 11 and the second winding group 12. Thereby, the non-uniformity of heat generation among the SW elements 211- 216, 221- 226 is suitably reduced,(Fourth Embodiment)The fourth embodiment of the present invention will be described based on FIG. 8.According to the present embodiment, the voltages applied to the first winding group 11 and the second winding group 12 are detected, and the switching period Pc is calculated based on the voltage detection value. The voltage detection value may be, for example, a terminal voltage detection value.The greater the voltages applied to the first winding group 11 and the second winding group 12, the higher the probability of high heat generation by the switching elements 211- 216, 221- 226 becomes. Therefore, for example, the switching period Pc is made shorter the larger the amplitude of the voltage detection value is, as shown in FIG. 8.In such a case, the switching period Pc may be calculated based on the voltage instruction value instead of using the voltage detection value.Further, the voltage instruction value may be a duty-ratio converted value.Further, the electric current detection values and / or the electric current instruction values used for such calculation may be derived from each of the three phases, or may be derived from each of the d / q axes.Further, the switching period Pc may be calculated based on the value concerning either only the voltage applied to the first winding group 11 or the voltage applied to the second winding group 12.In the present embodiment, the preset physical quantity is at least one of the voltage applied to the first winding group 11 and the voltage applied to the second winding group 12. Thereby, the non-uniformity of heat generation among the SW elements 211- 216, 221- 226 is appropriately reduced.(Fifth Embodiment)The fifth embodiment of the present invention will be described based on FIG. 9.According to the present embodiment, the shift period Pc is calculated based on the torque of the motor 10. Based on, for example, the current detection values or the current instruction values, the torque of the motor 10 is calculable. The greater the torque of the motor 10, the higher the probability of high heat generation by the switching elements 211- 216, 221- 226 becomes. Therefore, the switching period Pc is shortened as the torque of the motor 10 increases, for example, as shown in FIG. 9.Further, the switching period Pc may be calculated based on the steering torque detected by the torque sensor 94, instead of using the torque of the motor 10.In the present embodiment, the preset physical quantity is the torque of the motor 10.(Sixth Embodiment)The sixth embodiment of the present invention will be described based on FIG. 10.According to the present embodiment, the switching period Pc is calculated based on the inverter temperature, which is the temperature of the inverters 21 and 22. That is, a temperature sensor, which is not illustrated, may be disposed on the inverters 21 and 22, and the switching period Pc may be calculated based on the detection value of the temperature sensor in question.The switching period Pc may also be calculated based on the estimated temperature based on the electric current detection value or the electric current instruction value. That is, as shown in FIG. 10, the switching period Pc can be shortened as the temperature of the inverters 21 and 22 increases.The switching period Pc may also be calculated based on the temperature of either the first inverter 21 or the second inverter 22.In the present embodiment, the preset physical quantity is the temperature of at least one of the first inverter 21 and the second inverter 22, and thereby, based on the temperature of the first inverter 21 and the second inverter 22, the nonuniformity of heat generation among the SW elements 211- 216, 221- 226 is more appropriately reduced.(Seventh Embodiment)The seventh embodiment of the present invention will be described based on FIGS. 11A / B.According to the present embodiment, the power converter 1 is disposed in a vehicle. That is, in the present embodiment, a vehicle corresponds to a "system in which the rotating electric machine is installed". As shown in FIG. 11B, in a vehicle, the magnitude of the gain changes depending on the frequency. In the present embodiment, when the gain peaks at certain frequencies, such frequencies are defined as resonant frequencies X, Y and Z.Further, in the power converter 1, when the switching between the upward shift and the downward shift is performed in the same cycle as the resonance frequencies X, Y, and Z, the vibration and / or the noise may become larger than that in the switching in the other cycles.Therefore, according to the present embodiment, the switching period Pc is calculated so that the switching period Pc does not fall within a resonance range Rx, Ry, or Rz, each defined as a specific frequency range including a time Tx, a time Ty, or a time Tz, each derived as a specific time conversion value of the resonance frequency X, Y, or Z, as shown in FIG. 11A.Although FIGS. 11A / B illustrate an example of calculation of the switching period Pc based on the rotation speed of the motor 10, the same scheme is applicable when the switching period Pc is calculated based on the electric current, the voltage, the torque, or the temperature of the inverters 21 and 22. The same applies to the eighth embodiment mentioned later.Although FIGS. 11A / B each illustrate an example having three resonance regions Rx, Ry, Rz, the number of resonance regions may be arbitrarily determined depending on the resonance frequency of the vehicle. Further, the width of the resonant region may vary from region to region. The resonant regions Rx, Ry, Rzmay each have different region widths.According to the present embodiment, the switching period Pc is calculated by avoiding the resonance ranges Rx, Ry, Rz, which are respectively calculated as ranges including the time conversion values Tx, Ty, Tz of the resonance frequencies X, Y, Z. Thereby, an increase in vibration and noise due to resonance is avoided.(Eighth Embodiment)The eighth embodiment of the present invention will be described based on FIG. 12.According to the present embodiment, generation of a human audible sound is prevented by setting the switching period Pc to a frequency range outside a human audible range. Specifically, when the lower limit frequency of the human audible range is Hmin (e.g., 20 Hz), the switching period Pc may be set to be equal to or higher than a lower limit value Tmin (e.g., 50 ms), which is a time conversion value of Hmin.That is, when the rotational speed is less than a threshold Sth, the switching period Pc is set to the lower threshold Tmin, and when the rotational speed increases (that is, exceeds the threshold Sth), the switching period Pc may also be increased. In such a manner, the switching period Pc is controlled to be equal to or greater than the lower limit value Tmin, generation of the noise accompanying the switching between the upward shift and the downward shift is avoided.When, instead of using the rotational speed, the electric current, the voltage, the torque, or the inverter temperature is used, the switching period Pc may be set to the lower limit value Tmin when the above-described parameter is equal to or greater than a limit value. Thereby, the calculated switching period Pc becomes equal to or greater than the lower limit value Tmin.According to the present embodiment, the switching period Pc is calculated so as not to fall within a human audible frequency range. In such a manner, the noise accompanying the switching between the first period P 1 and the second period P 2 is avoided.(Other Embodiments)(a) Instruction ComputerAccording to the above-mentioned embodiments, the first duty instruction values calculated by the duty calculator correspond to a "first voltage instruction value", and the second duty instruction values calculated by the duty calculator correspond to a "second voltage instruction value".According to other embodiments, pre-duty conversion voltage instruction values may undergo a shift process, and the post-shift voltage instruction values after the shift process may be converted into the dutyThe same may apply to the modulation process. That is, the voltage instruction values before duty conversion may be regarded as, for example, the first voltage instruction value and the second voltage instruction value, and the function blocks such as a modulation processor and a shift processor may be disposed at a position between the controllers 413 and 423 of the first embodiment and the duty calculators 415 and 425. In such a case, the modulation processor and the shift processor correspond to an "instruction calculator".In the above-mentioned embodiment, the time constant of the low-pass filter performing filtering of the target shift amount is the preset time constant.According to other embodiments, considering the point that the responsiveness of the motor changes depending on the magnitude of the electric current, the time constant may be a value variable depending on the magnitude of the electric current. Here, "the magnitude of the electric current" may indicate a value based on the electric current detection value or based on the electric current instruction value.According to the above-mentioned embodiment, the target shift amount is filtered.According to other embodiments, by using post-filtering other parameters than the target shift amount, the calculation of the first voltage instruction value and the second voltage instruction value may be configured to control the time constant of the change of the first neutral point voltage and the second neutral point voltage to be larger than the preset time constant.(b) Electric current detectorAccording to the above-mentioned embodiments, the electric current detector is a shunt resistor and is disposed at a position on the low potential side of the low potential side SW element.According to other embodiments, the electric current detector need not necessarily be the shunt resistor, but may also be a Hall IC, etc., for example.Further, according to other embodiments, the electric current detector may be disposed at various positions other than the low potential side of the low potential side SW element, such as a position on the high potential side of the high potential side SW element, a position between the winding group and the inverter, or the like.(c) Rotating Electric MachineIn the above-mentioned embodiments, the rotating electric machine is a brushless motor having three phases.In other embodiments, the polyphase rotating electrical machine may also be usable except for the three phases (e.g., having four phases or more).Further, other than the brushless motor, any motor may be used.Further, the rotating electric machine may be not only a motor but also a generator, and may be a motor generator having the functions of the electric motor and the generator.According to the above-mentioned embodiments, the rotating electric machine is applied to an electric power steering apparatus.According to other embodiments, the rotating electric machine may be applied to devices other than the electric power steering device.Although the present invention has been described in connection with preferred embodiments thereof with reference to the accompanying drawings, it is to be noted that various changes and modifications will become apparent to those skilled in the art, and such changes, modifications and summarized schemes are intended to be within the scope of the present invention as defined by the appended claims.

Claims

An electric power converter for converting an electric power provided to a rotating electric machine (10) having a first winding group (11) and a second winding group (12), the power converter comprising: a first inverter (21) having a first switching element (211-216) associated with each of a plurality of phases in the first winding group (11); a second inverter (22) having a second switching element (221-226) associated with each of a plurality of phases in the second winding group (12); and a controller (40) including an instruction calculator (415, 425) that calculates a first voltage instruction value applied to the first winding group (11) and a second voltage instruction value applied to the second winding group (12), wherein an application voltage to the first winding group (11) is controlled (i) in a first period such that a first neutral point voltage (Vn1) shifts below a central output value (VM) of an outputable range and a second neutral point voltage (Vn2) shifts above the central output value (VM) of the outputable range, and (ii) in a second period such that the first neutral point voltage (Vn1) shifts above the central output value (VM) of the outputable range and the second neutral point voltage (Vn2) shifts below the central output value (VM) of the outputable range, the instruction calculator (415, 425) calculates the first instruction value and the second instruction value for enabling switching between the first period and the second period after elapse of a switching period (Pc) which is variably calculated based on a preset physical quantity, characterized in that the preset physical quantity includes a rotational speed of the rotating electric machine (10), and the switching period (Pc) is calculated as a longer period when the rotational speed of the rotating electric machine (10) is low and as a shorter period when the rotational speed is high.The electric power converter according to claim 1, wherein the switching period (Pc) is calculated by avoiding a resonance range including a time conversion value of a system resonance frequency regarding a system in which the rotating electric machine (10) is installed.The electric power converter according to claim 1, wherein the switching period (Pc) is calculated by avoiding a human audible range corresponding to a human audible frequency range.

Images