JP7483651B2 - Three-phase load driver and current sensor error detection method - Google Patents

Description

The present invention relates to a three-phase load driver and a method for detecting errors in a current sensor.

A three-phase load driver that controls a three-phase load such as a three-phase AC motor is known. The three-phase load driver sets a target current value for the drive torque, and detects the phase current value of each phase flowing through the three-phase load in order to control it with the target current value. The three-phase load is then controlled by feedback control or the like based on the detected phase current value and the target current value. In this type of control, the current sensor that detects the phase current value of each phase is important. If control is continued while there is an error in the detection value of the current sensor, there may be an excess or deficiency relative to the target current value, and the drive torque may not match. It is known that current sensors generally have an offset error in which the output of the current sensor is offset above or below the true value by a predetermined value, and a gain error in which the output of the current sensor is doubled by a predetermined value.

Patent document 1 discloses a technology that eliminates the effects of changes in offset error and gain error of a current sensor caused by environmental changes such as temperature while the motor is being driven.

JP 2010-110067 A

In addition to offset error and gain error, current sensors have linearity error with a certain error for each current value flowing through the current sensor. However, in Patent Document 1, linearity error is not taken into consideration, and even if offset error and gain error are detected, the remaining linearity error cannot be detected, making it impossible to control the three-phase load machine with high precision.

The three-phase load driving device according to the present invention comprises a current control unit which supplies current to the three-phase load in six current patterns in which current is supplied to one phase of the three-phase load and a current in the opposite direction is supplied to the other phase; a current detection unit which detects current inter-phase current values of two phases supplied with current when current is supplied in the six current patterns based on the current values from a current sensor which detects the current values of each phase flowing through the three-phase load; and an error calculation unit which calculates a current inter-phase error, which is the difference between the absolute values of the current inter-phase current values of the two phases detected by the current detection unit, for each of the six current patterns, and calculates an estimate of the linearity error corresponding to the current value of each phase of the current sensor using a pseudo-inverse matrix of a calculation matrix which represents the correspondence between the linearity error corresponding to the current value of each phase of the current sensor and the current inter-phase error.
A current sensor error detection method according to the present invention is a current sensor error detection method in a three-phase load driving device equipped with a current sensor that detects the current values of each phase flowing through a three-phase load, wherein current is passed through the three-phase load in six current patterns in which current is passed through one phase of the three-phase load and a current in the opposite direction is passed through the other phase, and based on the current values from the current sensor, current-to-phase current values of two phases passed when current is passed in the six current patterns are detected, and a current-to-phase error, which is the difference between the absolute values of the current-to-phase current values of the detected two phases, is calculated for each of the six current patterns, and an estimate of the linearity error corresponding to the current value of each phase of the current sensor is calculated using a pseudo-inverse matrix of a calculation matrix that represents the correspondence between the linearity error corresponding to the current value of each phase of the current sensor and the current-to-phase error.

According to the present invention, it is possible to control a three-phase load machine with high precision by detecting the linearity error of the current sensor.

FIG. 1 is a system configuration diagram of a three-phase load driver. FIG. 2 is a configuration diagram of an inverter and a three-phase load machine. FIG. 4 is a timing chart showing two-phase energization. 11 is a table showing the correspondence between alternating processes and current patterns of power semiconductor elements. 4 is a flowchart showing a process of the three-phase load driver according to the first embodiment. 10 is a flowchart showing a process of a current flow detection unit. FIG. 13 is a diagram illustrating a pseudo-inverse matrix. 5 is a flowchart showing a process of an error correction unit in the first embodiment. 10 is a flowchart showing a process of a three-phase load driver according to a second embodiment. FIG. 13 is a diagram showing a table in which a set DUTY and an output current value IOUT are linked by an INDEX. 10 is a flowchart showing a process of an error correction unit in the second embodiment. FIG. 13 is a diagram illustrating linear interpolation.

The following describes an embodiment of the present invention with reference to the drawings. The following description and drawings are examples for explaining the present invention, and some parts have been omitted or simplified as appropriate for clarity of explanation. The present invention can also be implemented in various other forms. Unless otherwise specified, each component may be singular or plural.

In addition, the following description may describe processing performed by executing a program, but the program is executed by a processor (e.g., a CPU or GPU) to perform a defined process using storage resources (e.g., a memory) and/or an interface device (e.g., a communication port) as appropriate, so the subject of the processing may be the processor. Similarly, the subject of the processing performed by executing a program may be a controller, device, system, computer, or node having a processor. The subject of the processing performed by executing a program may be any computing unit, and may include a dedicated circuit (e.g., an FPGA or ASIC) that performs a specific process.

A program may be installed in a device such as a computer from a program source. The program source may be, for example, a program distribution server or a computer-readable storage medium. When the program source is a program distribution server, the program distribution server may include a processor and a storage resource that stores the program to be distributed, and the processor of the program distribution server may distribute the program to be distributed to other computers. Also, in the following description, two or more programs may be realized as one program, and one program may be realized as two or more programs.

[First embodiment]
FIG. 1 is a system configuration diagram of a three-phase load driver 100 that controls the drive of a three-phase load machine 200. As shown in FIG.

The three-phase load driving device 100 converts DC power supplied from the battery 300 into AC power and applies AC current to the three-phase load machine 200 to drive the three-phase load machine 200. The three-phase load driving device 100 includes a current sensor 2, an inverter 3, a gate signal generating unit 4, a current control unit 5, a current detection unit 6, an error calculation unit 7, an error memory unit 8, and an error correction unit 9.

The current control unit 5 outputs current phase commands 51 and DUTY commands 52 to the gate signal generating unit 4 to energize the three-phase load 200 in six current patterns in which current is applied to one phase of the three-phase load 200 and a current in the opposite direction is applied to the other phase, as will be described in detail later.

The gate signal generating unit 4 generates a PWM signal for driving the inverter 3 based on the current phase command 51 and the duty command 52 from the current control unit 5.

When the current sensor 2 detects a linearity error, the inverter 3 energizes the three-phase load 200 in six energization patterns, energizing one phase of the three-phase load 200 and energizing the other phase in the reverse direction, based on a PWM signal from the energization control unit 5 based on an energization phase command 51 and a DUTY command 52. The inverter 3 also normally switches multiple power semiconductor elements in the inverter 3 based on a PWM signal from the drive control unit 10 (described later) to convert the DC power supplied from the battery 300 into three-phase AC power.

The current sensor 2 detects the currents of the U, V, and W phases flowing from the inverter 3 to the three-phase load 200.

The energizing current detection unit 6 detects the energizing phase-to-phase current values of the two phases that are energized when energized in the six energization patterns based on the current values detected by the current sensor 2.

The error calculation unit 7 calculates an inter-phase current error, which is the difference between the absolute values of the inter-phase current values of the two phases detected by the current detection unit 6, for each of the six current patterns. Furthermore, it calculates an estimated value of the linearity error corresponding to the current value of each phase of the current sensor 2 using a pseudo-inverse matrix of a calculation matrix that represents the correspondence between the linearity error corresponding to the current value of each phase of the current sensor 2 and the inter-phase current error. The details of the current control unit 5, the current detection unit 6, and the error calculation unit 7 will be described later.

The error storage unit 8 stores the estimated value of the linearity error calculated by the error calculation unit 7 .
The error correction unit 9 corrects the current value detected by the current sensor 2 based on the estimated value of the linearity error stored in the error storage unit 8 .
The drive control unit 10 controls the three-phase load machine 200 by feedback control or the like, with reference to the current values of each phase of the current sensor 2 corrected by the error correction unit 9 and the target current value for a target drive command input from a higher-level control unit (not shown). The drive control unit 10 outputs a drive command required for this control to the gate signal generation unit 4. Normally, the three-phase load machine 200 is driven and controlled by the control of the drive control unit 10. During a period in which the drive control unit 10 is not operating, when it becomes necessary to calculate an estimated value of the linearity error according to the current values of each phase of the current sensor 2, the current control unit 5 is operated to calculate an estimated value of the linearity error.

FIG. 2 is a configuration diagram of the inverter 3 and the three-phase load machine 200.
The inverter 3 includes power semiconductor elements T1, T2, and T3 that configure upper arm circuits of the U-phase, V-phase, and W-phase, and power semiconductor elements T4, T5, and T6 that configure lower arm circuits of the U-phase, V-phase, and W-phase. In this figure, the power semiconductor elements T1, T2, T3, T4, T5, and T6 are illustrated with switch symbols for convenience of explanation, but the power semiconductor elements T1, T2, T3, T4, T5, and T6 are, for example, IGBTs. Also, the same reference numerals are used for the same parts as in Figure 1, and explanations thereof will be omitted.

With reference to Figure 2, an example of two-phase current flow for determining linearity error will be described. Figure 2 illustrates the operation of an alternating process in which a positive current flows through the U phase and a negative current flows through the V phase. This alternating process will be referred to as the first alternating process.

Based on the energization phase command 51 and DUTY command 52 from the energization control unit 5, the gate signal generation unit 4 simultaneously turns on the upper arm power semiconductor element (T1 in this example) of one of the U, V, or W phases and the lower arm power semiconductor element (T5 in this example) of one of the remaining two phases for a predetermined time (t0), while keeping the other power semiconductor elements (T2, T3, T4, T6) constantly off to allow phase current to flow. At this time, a positive current IU is generated in the U phase, which had the upper arm turned on, and a negative current IV is generated in the V phase, which had the lower arm turned on.

Next, the upper arm power semiconductor element (T1 in this example) of one phase that has been turned on the upper arm side and the lower arm power semiconductor element (T5 in this example) of one phase that has been turned on the lower arm side of one of the remaining two phases are turned off at the same timing, and instead the lower arm power semiconductor element (T4 in this example) of one phase that has been turned on the upper arm side until now and the upper arm power semiconductor element (T2 in this example) of one of the remaining two phases that has been turned on the lower arm side until now are turned on at the same timing for the same period of time as the predetermined period of time, thereby alternating. This performs the current supply process in the first alternating step of the inverter 3 and the three-phase load machine 200.

Fig. 3 is a timing chart of two-phase energization, showing an example of the first alternating step shown in Fig. 2, i.e., a positive current is passed through the U phase and a negative current is passed through the V phase.
The upper arm power semiconductor element T1 and the lower arm power semiconductor element T5 are simultaneously turned on for a predetermined time t0, and at the next predetermined time t0, the lower arm power semiconductor element T4 and the upper arm power semiconductor element T2 are simultaneously turned on. That is, the power semiconductor elements T1, T5 and the power semiconductor elements T4, T2 are energized at a constant DUTY.

At a predetermined time t0+t0, a positive current IU is generated in the U-phase with the upper arm turned on, and a negative current IV is generated in the V-phase with the lower arm turned on. The current IU has a current peak value IUP, and the current IV has a current peak value IVM. In the first alternating stroke, the current IW of the W-phase is zero amperes. These current values are detected by the current sensor 2.
In this way, by alternately turning on and off the power semiconductor elements, it is possible to generate triangular output currents with opposite polarities in the current-carrying phases.

Figure 4 is a table showing the correspondence between the alternating steps and the current patterns that turn on/off the power semiconductor elements T1 to T6.

In the first alternating step 1, in the current direction UP (positive side), the power semiconductor elements T1 and T5 are turned on, and in the current direction DOWN (negative side), the power semiconductor elements T2 and T4 are turned on. In the second alternating step 2, in the current direction UP, the power semiconductor elements T2 and T4 are turned on, and in the current direction DOWN, the power semiconductor elements T1 and T5 are turned on. In the third alternating step 3, in the current direction UP, the power semiconductor elements T2 and T6 are turned on, and in the current direction DOWN, the power semiconductor elements T3 and T5 are turned on. In the fourth alternating step 4, in the current direction UP, the power semiconductor elements T3 and T5 are turned on, and in the current direction DOWN, the power semiconductor elements T2 and T6 are turned on. In the fifth alternating step 5, in the current direction UP, the power semiconductor elements T3 and T4 are turned on, and in the current direction DOWN, the power semiconductor elements T1 and T6 are turned on. In the sixth alternating step 6, when the current direction is UP, the power semiconductor elements T1 and T6 are turned on, and when the current direction is DOWN, the power semiconductor elements T3 and T4 are turned on.

As shown in FIG. 4, the current control unit 5 turns on and off the power semiconductor elements T1 to T6, and the current detection unit 6 detects the current values when alternating in each arm using the current sensor 2.

FIG. 5 is a flowchart showing the process of the three-phase load driver 100.
In step S501, the error correction unit 9 determines whether a linearity error has been estimated. Specifically, it determines whether an estimated value of the linearity error is stored in the error storage unit 8. If an estimated value is not stored, the linearity error has not been estimated, and the linearity error estimation process shown in steps S502 to S507 is performed. If an estimated value is stored, the correction process of the current sensor 2 shown in step S508 is performed based on the estimated value of the linearity error stored in the error storage unit 8.

In step S502, the current control unit 5 sets the current pattern from the first alternating step 1 to the sixth alternating step 6 as described with reference to Figures 2, 3, and 4. The DUTY for currenting each of the power semiconductor elements T1, T2, T3, T4, T5, and T6 is constant.

In step S503, the energization control unit 5 inputs the energization phase command 51 and the DUTY command 52 to the gate signal generation unit 4 based on the energization pattern. This causes the energization process of the inverter 3 and the three-phase load machine 200 to be performed.

In step S504, the energizing current detection unit 6 detects the energizing phase-to-phase current values of the two phases that are energized when energized in the six energization patterns based on the current values detected by the current sensor 2. Details of the processing by the energizing current detection unit 6 will be described later with reference to the flowchart in FIG. 6.

In step S505, the error calculation unit 7 calculates, for each of the six current patterns, an energization phase-to-phase error, which is the difference between the absolute values of the energization phase-to-phase current values of the two phases detected by the energization current detection unit 6. The details of this energization phase-to-phase error calculation process will be described later.

Then, in step S506, the error calculation unit 7 calculates an estimate of the linearity error corresponding to the current value of each phase of the current sensor 2 using a pseudo-inverse matrix of a calculation matrix that represents the correspondence between the linearity error corresponding to the current value of each phase of the current sensor 2 and the current-carrying phase error. Details of this linearity error estimation calculation process will be described later.

In step S507, the error calculation unit 7 stores the calculated estimated value of the linearity error in the error storage unit 8.

If the estimated value is stored in the error storage unit 8 in step S501, the process proceeds to step S508, where the error correction unit 9 corrects the current value detected by the current sensor 2 based on the estimated value of the linearity error stored in the error storage unit 8. Details of this correction process will be described later with reference to the flowchart in FIG. 8.

Fig. 6 is a flowchart showing the process of the energizing current detection unit 6. It is a detailed flowchart of step S504 shown in Fig. 5, which is a process for detecting the current value between energized phases.
In step S601, the current alternating process is confirmed. For example, if it is the first alternating process, the current U-phase current (IU) and V-phase current (IV) are captured and stored in step S602.

In step S603, if the currently acquired U-phase current (IU) is greater than the previously stored U-phase maximum current (IUPz), the U-phase maximum current (IUP) is updated to the currently acquired U-phase current (IU) in step S604.

If the condition in step S603 is NO, the previous U-phase maximum current (IUPz) is retained as the U-phase maximum current (IUP) in step S605, and the previous U-phase maximum current value (IUPz) is updated in step S606.

Similarly, in step S607, if the currently acquired V-phase current (IV) is smaller than the previously stored V-phase minimum current (IVMz), the V-phase minimum current (IVM) is updated to the currently acquired V-phase current (IV) in step S608.

If the condition in step S607 is NO, the previous V-phase minimum current (IVMz) is held as the V-phase minimum current (IVM) in step S609, and the previous V-phase maximum current value (IVMz) is updated in step S610.

If the current alternating process is not the first alternating process in step S601, the process proceeds to step S611. In step S611, the U-phase maximum value (IUPz) at the time when the first alternating process ends is stored as IUP1 in a non-volatile memory or the like, and the V-phase minimum value (IVMz) is temporarily stored as IVM1 in a non-volatile memory or the like.

The flowchart in Figure 6 explains the first alternating process as an example, but the maximum and minimum values of each phase are obtained for the other alternating processes by similar processing, and are temporarily stored in a non-volatile memory, etc.

Next, we will explain the current-carrying phase-to-phase error calculation process of the error calculation unit 7. This is a detailed explanation of step S505 shown in Figure 5.

The error calculation unit 7 reads out the maximum and minimum values of each phase temporarily stored in a non-volatile memory or the like, and calculates the interphase error. The interphase error is an element for calculating an error corresponding to the flowing current value from the interphase current value.

In the first alternating process, the U phase is a positive current and the V phase is a negative current, so if the inter-phase error based on the U phase maximum value IUP1 and the V phase minimum value IVM1 obtained in the first alternating process is Eupvm, the error can be defined by the following equations (1) and (2). Note that the U phase positive error (Eum1) and V phase negative error (Evm1), which are errors corresponding to the current value and are error elements of the inter-phase error Eupvm, can be expressed as in equation (2), but since they cannot be observed directly, they can be found by a matrix calculation equation described later.
Eupvm=|IUP1|-|IVM1| ... (1)
Eupvm=Eup1+Evm1 ... (2)

Because this is a current detection value obtained by applying current to two phases, if there were no error in the current sensor 2, the absolute values would be equal, and the inter-phase error Eupvm value would be 0 A. However, since the current sensor 2 actually contains an error, the error amount excluding the true value is extracted.

Furthermore, the current peak value, inter-phase error, and error element obtained in the alternating process between the upper and lower arm phases can be defined by the following equations (3) to (12).
When U-phase current is set to the negative side and V-phase current is set to the positive side,
Eumvp = |IUM1| - |IVP1| ... (3)
Eumvp=Eum1+Evp1 ... (4)

When V-phase current is set to the positive side and W-phase current is set to the negative side,
Evpwm=|IVP2|-|IWM1| ... (5)
Evpwm=Evp1+Ewm1 (6)
Incidentally, since the IVP value obtained here is an IVP value that exists when the UV phases are energized, the suffix 2 is added to distinguish it from the IVP value obtained when the UV phases are energized, and thus it is referred to as IVP2.

When V-phase current is set to the negative side and W-phase current is set to the positive side,
Evmwp = |IVM2| - |IWP1| ... (7)
Evmwp=Evm1+Ewp1 (8)
Incidentally, since the IVM value obtained here is an IVM value that exists when the UV phases are energized, the subscript 2 is added to differentiate it from the IVM value obtained here, that is, IVM2.

When W-phase current is set to the positive side and U-phase current is set to the negative side,
Ewpum = |IWP2| - |IUM2| ... (9)
Ewpum=Ewp1+Eum1 (10)

When W-phase current is set to the negative side and U-phase current is set to the positive side,
Ewmup = |IWM2| - |IUP2| ... (11)
Ewmup=Eum1+Ewm1 ... (12)
Incidentally, the IUP, IUM, IWP, and IWM values obtained here are given the subscript 2 for differentiation because the energized phases are different from those when the UV phase or VW phase is energized.

Next, the error calculation unit 7 performs an estimation calculation process for the linearity error. This is a detailed explanation of step S506 shown in FIG. 5.

Here, the relationship between the current conduction pattern of the power semiconductor elements T1, T2, T3, T4, T5, and T6 and the interphase error and correlation error elements obtained at that time can be replaced with the determinant of the following equation (13).

The columns of the determinant correspond, from the left, to the U-phase upper arm (T1), V-phase upper arm (T2), W-phase upper arm (T3), U-phase lower arm (T4), V-phase lower arm (T5), and W-phase lower arm (T6). A "1" in the matrix indicates that the arm is in the on state, and a "0" indicates that the arm is in the off state.

By substituting the first matrix term on the left side of equation (13) as A, the second matrix term on the left side as X, and the matrix on the right side as b, we obtain equation (14).
A × X = b ... (14)

Since X in the second term on the left side of equation (14) is the error in the desired specified current value for each phase, that is, the so-called linearity error, by transforming equation (14) into an equation for calculating the error X, we obtain the following equation (15). Here, since the matrix in the first term on the left side of equation (13) has a rank of 5, it does not have a normal inverse matrix. Therefore, a pseudo-inverse matrix is applied to A ⁺ shown in equation (15). By using the pseudo-inverse matrix, X' indicates the optimal solution for equation (14) that minimizes the error norm |AX-b|. Here, X' is an approximation of X.
X'=A ⁺ ×b (15)

Fig. 7 is a diagram showing the pseudo-inverse matrix A ⁺ of A. The pseudo-inverse matrix A ⁺ is a matrix as shown in Fig. 7, but the calculation approach and numerical precision of each element value can be set according to the usage environment.

The error calculation unit 7 stores the calculated linearity error in the error storage unit 8 in step S507 of FIG. 5. The linearity error may be calculated when the three-phase load driving device 100 is manufactured or adjusted, and may be stored in advance in the error storage unit 8. The processes shown in the flowcharts of FIG. 5 and FIG. 6 are executed by a processor or the like in the three-phase load driving device 100, but these processes may be executed by a device external to the three-phase load driving device 100 and may be stored in advance in the error storage unit 8. The determination of whether the linearity error of the current sensor 2 can be corrected may be determined based on whether the linearity error is stored in the error storage unit 8.

Figure 8 is a flowchart showing the processing of the error correction unit 9. It is a detailed flowchart of step S508 shown in Figure 5, and is a correction process that corrects the current value detected by the current sensor 2 based on the estimated value of the linearity error stored in the error storage unit 8.

8, a U-phase current sensor value IU of current sensor 2 is obtained. In step S802, it is determined whether the current value obtained in step S801 is a positive value. If it is a positive value, in step S803, the U-phase current sensor value is corrected by the following equation (16) using the U-phase positive side error Eup stored in error storage unit 8 in the linearity error estimation calculation, to obtain a corrected U-phase current IU'.
IU′=IU−Eup (16)

If the condition is not satisfied in step S802 and the current value is negative in step S804, in step S805, the U-phase negative value side error Eum stored in the error memory unit 8 in the linearity error estimation calculation is used to correct the U-phase current sensor value using the following equation (17), thereby obtaining the corrected U-phase current IU'.
IU'=IU-Eum (17)

If the conditions are not satisfied in steps S801 and S804, the current value is 0 A, and therefore, in step S806, the corrected U-phase current IU' is obtained from the following equation (18).
IU′=IU . . . (18)
Although the above description has been given with respect to the U-phase current sensor, the same correction is also performed for the V-phase and W-phase current sensors.

[Second embodiment]
In the first embodiment, the linearity error for each polarity of each phase is corrected, but in reality, the linearity error of the current sensor 2 may differ depending on the current value, so it is necessary to perform error correction according to the current value. Therefore, in the second embodiment, an example in which the linearity error is corrected according to the current value in each phase will be described.

Figure 9 is a flowchart showing the processing of the three-phase load driving device 100 in the second embodiment. The system configuration diagram of the three-phase load driving device 100 is the same as that of the first embodiment shown in Figure 1. The alternating process is also basically the same as that of the first embodiment described with reference to Figures 2 to 4, but differs from the first embodiment in that the pulse ON time for switching the power semiconductor elements T1, T2, T3, T4, T5, and T6 is variable.

In the first embodiment, the pulse ON time when switching the power semiconductor elements T1, T2, T3, T4, T5, and T6 was set to a predetermined time t0, so a certain true output current value was targeted, but it is possible to adjust the output current by varying the pulse ON time.

In step S901 of FIG. 9, the error correction unit 9 determines whether a linearity error has been estimated. If it determines that a linearity error has not been estimated, it performs the linearity error estimation process shown in steps S902, S903, and the above-described steps S503 to S507. If a linearity error has been estimated, it performs the correction process of the current sensor 2 shown in step S908 based on the estimated value of the linearity error stored in the error storage unit 8.

In step S902, the energization control unit 5 sequentially outputs DUTY commands 52 with different numerical values DUTY(1) to DUTY(n), and first outputs one of the DUTY commands 52, DUTY(1), to the gate signal generation unit 4. The gate signal generation unit 4 outputs an energization pattern to the inverter 3 according to one of the input DUTY commands 52, DUTY(1).

In step S903, the energization control unit 5 determines whether the DUTY command 52 is within the range up to DUTY(n). If it is within the range, it performs the same linearity error estimation process as steps S503 to S507 shown in FIG. 5.

After processing in step S507, the process returns to step S902, the DUTY command 52 is updated to the next DUTY (2), and the processes in steps S903 and S503 to S507 are performed. The same process is then repeated up to DUTY (n).

If the energization control unit 5 determines in step S903 that it is not within the range of the DUTY command 52 up to DUTY(n), this indicates that estimation of the linearity error for all predetermined DUTY(n)s in which the pulse ON time is variable has been completed. In this case, the process returns to step S901. In step S901, the error correction unit 9 determines that estimation of the linearity error has been performed, and performs the correction process of the current sensor 2 shown in step S908 based on the estimated value of the linearity error stored in the error memory unit 8.

The output current values obtained by the current conduction pattern in which the current conduction times of DUTY(1) to DUTY(n) are changed by repeating the processes of steps S903 and S503 to S507 described above are designated as output current value IOUT(1) to output current value IOUT(n). Then, a group of linearity errors corresponding to the obtained output current value IOUT(1) to output current value IOUT(n) are designated as Euvw(1) to Euvw(n) as shown in the following equations (19) to (21). Note that each linearity error is calculated for each of DUTY(1) to DUTY(n) according to the above-mentioned equations (13) to (15).

Figure 10 is a diagram showing a table in which the set DUTY and the output current value IOUT are linked by INDEX. Each linearity error Euvw calculated by equations (19) to (21) can be regarded as the linearity error of the current sensor 2 corresponding to the output current value, and can be linked by the table shown in Figure 10. The table shown in Figure 10 is stored in a non-volatile memory or the like.

Figure 11 is a flowchart showing the processing of the error correction unit 9 in the second embodiment. It is a flowchart showing the details of step S908 in Figure 9, and shows the correction processing of the current sensor 2.

11 is a flowchart showing an example of a procedure for correcting the linearity error in the U-phase current sensor 2. In this example, it is assumed that the absolute value IUABS of the acquired U-phase current value is within the range shown in the following equation (22).
IOUT(3)<IUABS≦IOUT(4) (22)

In step S001, the U-phase current sensor value IU is acquired. The current value acquired in step S001 is subjected to absolute value processing in step S002. If the condition for the current value IUABS obtained in step S002 is satisfied in step S003 and the condition is not satisfied in step S004, the process transitions to step S005. In step S005, it is determined whether the current value acquired in step S001 is a positive value. If it is a positive value, the process transitions to step S006.

In step S006, the linearity errors Eup(4) and Eup(3) corresponding to the current range INDEX are extracted by referring to the table shown in FIG.
In step S007, the linearity error Eup is calculated by linear interpolation using Eup(3) and Eup(4) extracted in step S006.

尚、Ｅｕｐ（４）およびＥｕｐ（３）は前述の直線性誤差演算にて不揮発性メモリ等に保存されていることが望ましい。 12 is a diagram showing linear interpolation, where IOUT(X) and IOUT(X-1) correspond to Eab(X) and Eab(X-1), where a=u, v, w, and b=p, m.
In this example, since IUABS has the relationship shown in equation (22), Eup can be calculated by the following equation (23).

Incidentally, it is desirable that Eup(4) and Eup(3) are stored in a non-volatile memory or the like in the above-mentioned linearity error calculation.

In step S008, the obtained Eup is used to correct the value of U-phase current sensor 2. In step S008, the value of U-phase current sensor 2 is corrected by the following equation (24) to obtain corrected U-phase current IU'.
IU'=IU-Eup (24)

On the other hand, if the condition is not satisfied in step S005 and the current value is a negative value in step S009, the process proceeds to step S010.
In step S010, the linearity errors Eum(4) and Eum(3) corresponding to the current range INDEX are extracted by referring to the table shown in FIG.

尚、Ｅｕｍ（４）およびＥｕｍ（３）は前述の直線性誤差演算にて不揮発性メモリ等に保存されていることが望ましい。 In step S011, the linearity error Eum is calculated using the linear interpolation shown in FIG.
In this example, since IUABS has the relationship shown in equation (22), Eum can be calculated by the following equation (25).

Incidentally, it is desirable that Eum(4) and Eum(3) are stored in a non-volatile memory or the like after the linearity error calculation described above.

In step S012, the obtained Eum is used to correct the value of U-phase current sensor 2. In step S012, the value of U-phase current sensor 2 is corrected by the following equation (26) to obtain a corrected U-phase current IU'.
IU'=IU-Eum... (26)

When the condition is not met in step S009, the corrected U-phase current IU' is obtained using equation (18) to indicate that the current value is 0 A.

When the condition in step S003 is not satisfied, if IUABS is equal to or less than IOUT(3), the same processing as in steps S003 to S013 is performed in step S014. In this case, in the processing corresponding to step S003, IOUT(3) is replaced with, for example, IOUT(2), and in the processing corresponding to step S004, IOUT(4) is replaced with, for example, IOUT(3). In addition, in the processing corresponding to steps S006 and S007, Eup(4) and Eup(3) are replaced with, for example, Eup(3) and Eup(2), respectively, and in the processing corresponding to steps S010 and S011, Eum(4) and Eum(3) are replaced with, for example, Eum(3) and Eum(2), respectively.

When the condition is met in step S004, if IUABS is greater than IOUT(4), the same processing as steps S003 to S013 is performed in step S015. In this case, in the processing corresponding to step S003, IOUT(3) is replaced with, for example, IOUT(4), and in the processing corresponding to step S004, IOUT(4) is replaced with, for example, IOUT(5). In addition, in the processing corresponding to steps S006 and S007, Eup(4) and Eup(3) are replaced with, for example, Eup(5) and Eup(4), respectively, and in the processing corresponding to steps S010 and S011, Eum(4) and Eum(3) are replaced with, for example, Eum(5) and Eum(4), respectively.

Furthermore, when IUABS is in another range, linear interpolation can be performed using the linearity error value corresponding to that range by the same process as above, and the linearity errors Eup and Eum can be calculated.

The above explanation describes one current condition for the U-phase current sensor 2, but similar corrections can be made for other current values and the V-phase and W-phase current sensors.

In the second embodiment, the linearity error may be calculated during manufacture or adjustment of the three-phase load driver 100 and stored in advance in the error storage unit 8. The processes shown in the flowchart of FIG. 9 and the like have been described as being executed by a processor or the like within the three-phase load driver 100, but these processes may be executed by a device external to the three-phase load driver 100 and stored in advance in the error storage unit 8. The determination of whether the linearity error of the current sensor 2 can be corrected may be determined based on whether the linearity error is stored in the error storage unit 8.

According to the embodiment described above, the following advantageous effects can be obtained.
(1) The three-phase load driving device 100 includes an energization control unit 5 that energizes the three-phase load 200 in six energization patterns in which one phase of the three-phase load 200 is energized and the other phase is energized in the opposite direction, an energization current detection unit 6 that detects the energized interphase current values of two phases that are energized when energized in the six energization patterns based on the current values from a current sensor 2 that detects the current values of each phase flowing through the three-phase load 200, and an error calculation unit 7 that calculates an energization interphase error that is the difference between the absolute values of the energization interphase current values of the two phases detected by the energization current detection unit 6 for each of the six energization patterns, and calculates an estimate of the linearity error corresponding to the current value of each phase of the current sensor 2 using a pseudo inverse matrix of a calculation matrix that indicates the correspondence between the linearity error corresponding to the current value of each phase of the current sensor 2 and the energization interphase error. This makes it possible to detect the linearity error of the current sensor and control the three-phase load with high accuracy.

(2) The error detection method of the current sensor 2 is a method of detecting an error of the current sensor 2 in a three-phase load driving device 100 equipped with a current sensor 2 that detects the current value of each phase flowing through the three-phase load 200, and energizes the three-phase load 200 in six energization patterns in which one phase of the three-phase load 200 is energized and the other phase is energized in the opposite direction. Based on the current value from the current sensor 2, the energized phase-to-phase current values of the two phases that are energized during energization in the six energization patterns are detected. For each of the six energization patterns, an energization phase-to-phase error, which is the difference between the absolute values of the detected energization phase-to-phase current values, is calculated, and an estimated value of the linearity error corresponding to the current value of each phase of the current sensor 2 is calculated using a pseudo-inverse matrix of a calculation matrix that represents the correspondence between the linearity error corresponding to the current value of each phase of the current sensor 2 and the energization phase-to-phase error. This allows the linearity error of the current sensor to be detected and the three-phase load to be controlled with high accuracy.

The present invention is not limited to the above-described embodiments, and other forms that are conceivable within the scope of the technical concept of the present invention are also included within the scope of the present invention, so long as they do not impair the characteristics of the present invention. In addition, configurations that combine the above-described embodiments may also be used.

2...Current sensor, 3...Inverter, 4...Gate signal generation unit, 5...Electrical supply control unit, 6...Electrical supply current detection unit, 7...Error calculation unit, 8...Error memory unit, 9...Error correction unit, 10...Drive control unit, 51...Electrical supply phase command, 52...DUTY command, 100...Three-phase load drive device, 200...Three-phase load machine, 300...Battery, T1...U-phase upper arm power element semiconductor, T2...V-phase upper arm power element semiconductor, T3...W-phase upper arm power element semiconductor, T4...U-phase lower arm power element semiconductor, T5...V-phase lower arm power element semiconductor, T6...W-phase lower arm power element semiconductor.

Claims (5)

an energization control unit that energizes the three-phase load in six energization patterns in which a current is applied to one phase of the three-phase load and a current is applied in a reverse direction to the other phase;
a current detection unit that detects current values between two phases that are energized when energized in the six current patterns based on current values from a current sensor that detects a current value of each phase flowing through the three-phase load machine;
a current detection unit for detecting an absolute value of the current-to-phase error of the current sensor, the current detection unit detecting an absolute value of the current-to-phase error of the current sensor, and a current detection unit for detecting an absolute value of the current-to-phase error of the current sensor. 2. The three-phase load driver according to claim 1,
The three-phase load driving device further comprises an error correction unit that corrects the current value from the current sensor based on the estimated value of the linearity error obtained by the error calculation unit. 3. The three-phase load driver according to claim 1,
The current control unit changes a current application time in the six current application patterns,
the energizing current detection unit detects, based on the current values from a current sensor detecting the current values of the phases flowing through the three-phase load machine, energizing phase-to-phase current values of two phases that are energized during energization in the six energization patterns for each of the changed energization times;
The error calculation unit calculates an estimated value of the linearity error based on the current-carrying phase-to-phase current value detected during each current-carrying time. A method for detecting an error in a current sensor in a three-phase load driving device including a current sensor for detecting a current value of each phase flowing through a three-phase load machine, comprising:
energizing one phase of the three-phase load machine and energizing the other phase in a reverse direction in six current patterns;
Detecting inter-phase current values of two phases that are energized when energized in the six energization patterns based on the current values from the current sensors;
A current sensor error detection method, which calculates an energization phase-to-phase error, which is the difference in absolute values of the energization phase-to-phase current values of the detected two phases, for each of the six energization patterns, and calculates an estimated value of the linearity error corresponding to the current value of each phase of the current sensor using a pseudo-inverse matrix of a calculation matrix that represents the correspondence between the linearity error corresponding to the current value of each phase of the current sensor and the energization phase-to-phase error. 5. The current sensor error detection method according to claim 4,
storing the calculated estimated value of the linearity error;
A current sensor error detection method, comprising: correcting the current value from the current sensor based on the stored estimated value of the linearity error.

Images