Motor iron loss operation device and motor control device having the same

Abstract

A motor iron loss operation device (1) operates an iron loss value (PIL) of a motor on the basis of a 1st frequency (ω) and a q-axis current component (iq) in a dq-axis rotating coordinate in which a rotor flux direction of the motor is set as a d-axis.

Description

TECHNICAL FIELD

[0001] The present application relates to an electric motor iron loss calculation device and an electric motor control device having the same. BACKGROUND

[0002] An electric motor converts electric power into mechanical power, but a loss occurs along with the conversion. As a main loss, there are a loss of resistance of a winding of the electric motor, i.e., copper loss, and an iron loss of an iron core of the electric motor. In output control of the electric motor, there is a method of performing torque control by calculating a generated torque, and as a method of calculating the generated torque, in a conventional method described in Patent Document 1, an iron loss is calculated, and a generated torque is calculated based on the calculated iron loss value.

[0003] In the conventional iron loss calculation described in Patent Document 1, the iron loss W is separated into a hysteresis loss Wh and an eddy current loss We, and is calculated by the following equation.

[0004] W = Wh + We

[0005] Wh = σh - f1 - Bm2

[0006] We = σe - d2 - f1 2 - Bm2

[0007] Here, f1: 1st order frequency, Bm: maximum value of magnetic flux density of the iron core (maximum value of amplitude of the magnetic flux density), σh: constant determined by the material of the iron core, σe: constant determined by the electrical resistivity of the iron core, d: thickness of the iron core

[0008] Patent Document 1: Japanese Patent Application Laid-Open No. 2000-102298 SUMMARY

[0009] In the conventional iron loss calculation described in Patent Document 1, the calculation is performed based on a theoretical equation using information of a magnetic material, and when the operating condition of the electric motor changes, it is difficult to calculate a high-precision iron loss corresponding to the operating condition.

[0010] The present application discloses a technology for solving the above-described problem, and aims to provide an electric motor iron loss calculation device capable of calculating an iron loss of an electric motor with high precision corresponding to an operating condition of the electric motor.

[0011] In addition, the present application aims to provide an electric motor control device having the electric motor iron loss calculation device described above, capable of calculating an iron loss of an electric motor with high precision and reliably controlling the electric motor.

[0012] The electric motor iron loss calculation device disclosed in the present application calculates an iron loss value of an electric motor based on a 1st order frequency of the electric motor and a q-axis current component in a dq-axis rotating coordinate in which a rotor magnetic flux direction of the electric motor is set as a d-axis.

[0013] Further, the motor control device disclosed in the present application detects a current in a dq-axis rotating coordinate in which a rotor magnetic flux direction of a motor is set as a d-axis, generates a voltage command in a manner that a current follows a command value, and uses a phase of the dq-axis rotating coordinate that is calculated based on a 1st frequency to be applied for coordinate conversion between the dq-axis rotating coordinate and a three-phase stationary coordinate. Further, the motor iron loss calculation device calculates an iron loss value of the motor.

[0014] Effects of the Invention

[0015] According to the motor iron loss calculation device disclosed in the present application, an iron loss can be calculated with high precision corresponding to an operation state of a motor.

[0016] Further, according to the motor control device disclosed in the present application, an iron loss corresponding to an operation state of a motor can be calculated with high precision, and a motor can be controlled with high reliability. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 is a diagram showing a structure of the motor iron loss calculation device according to Embodiment 1.

[0018] Figure 2 is a diagram showing an example of a measurement result of an iron loss corresponding to a torque current of a motor.

[0019] Figure 3 is a reference diagram showing an example of a measurement result of an iron loss corresponding to a 1st magnetic flux of a motor.

[0020] Figure 4 is a diagram for explaining linear interpolation at the time of coefficient determination in the coefficient conversion section according to Embodiment 1.

[0021] Figure 5 is a diagram showing an example of a measurement result of an iron loss corresponding to a torque current of a motor.

[0022] Figure 6 is a diagram showing an example of a hardware structure of the motor iron loss calculation device according to Embodiment 1.

[0023] Figure 7 is a diagram showing a structure of the motor iron loss calculation device according to Embodiment 2.

[0024] Figure 8 is a diagram showing an example of an iron loss resistance corresponding to a torque current of a motor.

[0025] Figure 9is a graph showing an example of an iron loss resistance corresponding to a torque current of a motor.

[0026] Figure 10 is a graph showing a structure of a motor iron loss calculation device according to Embodiment 3.

[0027] Figure 11 is a graph showing a holding information of a function calculation section according to Embodiment 3.

[0028] Figure 12 is a graph showing a linear interpolation in a function calculation section according to Embodiment 3.

[0029] Figure 13 is a graph showing a structure of a motor iron loss calculation device according to another example of Embodiment 3.

[0030] Figure 14 is a graph showing a correction calculation corresponding to an excitation current according to Embodiment 4.

[0031] Figure 15 is a graph showing a correction calculation corresponding to an excitation current according to another example of Embodiment 4.

[0032] Figure 16 is a graph showing a correction calculation corresponding to an excitation current according to another example of Embodiment 4.

[0033] Figure 17 is a graph showing a correction calculation corresponding to an excitation current according to another example of Embodiment 4.

[0034] Figure 18 is a graph showing a structure of a motor iron loss calculation device according to Embodiment 5.

[0035] Figure 19 is a graph showing a structure of a motor control device according to Embodiment 6.

[0036] Figure 20 is a graph showing a structure of a motor control device according to Embodiment 7. DETAILED DESCRIPTION

[0037] Embodiment 1.

[0038] Figure 1 is a graph showing a structure of a motor iron loss calculation device according to Embodiment 1.

[0039] As shown in the drawing, the motor iron loss operation device 1 outputs an iron loss value PIL with the primary frequency ω of the motor, the torque current of the motor, i.e., the q-axis current component iq (hereinafter, referred to as the q-axis current iq or iq), and the excitation current, i.e., the d-axis current component id (hereinafter, referred to as the d-axis current id or id) as inputs. Further, the d-axis and q-axis currents id and iq are currents in the dq-axis rotating coordinate system in which the rotor magnetic flux direction of the motor is set as the d-axis.

[0040] Figure 2 is a graph showing an example of a measurement result of the iron loss corresponding to the torque current of the motor. In this case, the measurement result of the iron loss of a certain induction motor is used, the vertical axis is set as the iron loss value, and the horizontal axis is set as the q-axis current iq. In addition, the measurement results of several primary frequency conditions are collectively described, and in this case, waveforms 11, 12, and 13 of the iron loss corresponding to the high, medium, and low stages of the primary frequency ω are illustrated.

[0041] As shown in Figure 2 , the iron loss changes in accordance with the primary frequency ω, and as a result, the higher the primary frequency ω, the greater the iron loss.

[0042] As described above, the iron loss of the motor is roughly divided into two types of eddy current loss and hysteresis loss, and it is known that they strongly depend on the magnetic flux or the frequency (primary frequency ω) of the current at that time. In addition, the change in the iron loss corresponding to the primary frequency ω is significant, and the tendency is similarly observed in almost all motors.

[0043] Further, as shown in Figure 2 , the iron loss nonlinearly varies in accordance with the q-axis current iq.

[0044] The inventors of the present application found the phenomenon that, as shown in Figure 2 , the iron loss of the motor greatly changes in accordance with the increase and decrease of the q-axis current iq, the iron loss is uniquely determined by only the q-axis current iq under the condition that the primary frequency is fixed, and the iron loss is different under the power running operation condition and the regenerative operation condition.

[0045] The inventors also found the phenomenon that, as shown in Figure 3 , under the condition that the primary frequency is fixed, the iron loss cannot be uniquely expressed by only the norm of the stator magnetic flux (primary magnetic flux). Further, Figure 3 is a reference graph showing an example of a measurement result of the iron loss corresponding to the primary magnetic flux of the motor. Here, a waveform 14 of the iron loss corresponding to the norm of the primary magnetic flux is illustrated.

[0046] The motor iron loss operation device 1 according to the embodiment operates the iron loss value PIL based on the primary frequency ω of the motor and the q-axis current iq.

[0047] As shown in FIG. 1, the motor iron loss operation device 1 has a coefficient conversion section 2, a polynomial operation section 3, and a correction section 4. Figure 1

[0048] The polynomial operation section 3 is provided with a polynomial function F(ω) having the first frequency ω as a variable. Moreover, the polynomial function F(ω) is operated with the coefficients A2, A1, A0 of each term and the first frequency ω as inputs, and the operation result of the polynomial function F(ω) is output. In this case, since the polynomial function F(ω) uses a second order function, the second order, the first order, and the zero order (=1) are respectively operated in the square calculation sections 3a, 3b, 3c of each term. Moreover, the outputs of each square calculation section 3a, 3b, 3c are multiplied by the coefficients A2, A1, A0 and added, and the sum of the three terms is output.

[0049] In this case, F(ω) is expressed by the following equation (1).

[0050] F(ω) = A2- ω2+ A1- ω + A0- · · · (1)

[0051] The coefficient conversion section 2 holds the coefficient information 2a, 2b, 2c for each term of the polynomial function F(ω). Moreover, the q-axis current iq is input, and the coefficients A2(iq), A1(iq), A0(iq) of each term of the polynomial function F(ω) are output with reference to each coefficient information 2a, 2b, 2c. In addition, A2(iq), A1(iq), A0(iq) are the values of each coefficient A2, A1, A0 corresponding to the value of iq.

[0052] In the polynomial operation section 3, the coefficients A2(iq), A1(iq), A0(iq) output from the coefficient conversion section 2 and the first frequency ω are substituted into the set F(ω), and the operation result, that is, the iron loss value PILα is output. That is, PILα becomes the following equation (2).

[0053] PILα = A2(iq) - ω2+ A1(iq) - ω + A0(iq) - · · · (2)

[0054] Here, each coefficient information 2a, 2b, 2c held by the coefficient conversion section 2 is a function in which each one value of each coefficient A2, A1, A0 corresponds to the q-axis current iq as an argument, which is generated for one set value of the d-axis current id. The function of each coefficient information 2a, 2b, 2c is not limited to be expressed by an equation, but can be expressed by a data table or a data correspondence graph.

[0055] ​In this case, as a set value of the d-axis current id, the rated d-axis current idα of 100% is used, and under the condition of the d-axis current idα, the measured information of the iron loss of the target motor is collected in advance to generate the respective coefficient information 2a, 2b, 2c, which are stored in the coefficient conversion section 2.

[0056] The iron loss value PILα from the polynomial operation section 3 is input to the correction section 4, which performs a correction operation on the iron loss value PILα corresponding to the actual d-axis current id to output the iron loss value PIL.

[0057] Hereinafter, the coefficient conversion section 2 will be described in detail.

[0058] The coefficients A2, Al, A0 of the respective coefficient information 2a, 2b, 2c determine the data shape depending strongly on the characteristics of the motor. On the other hand, in the installation of the coefficient conversion section 2, the amount of operation is small, and in addition, the storage capacity used is reduced to achieve miniaturization and high speed. Therefore, the respective coefficient information 2a, 2b, 2c is appropriately selected from the data shape of the respective coefficients A2, Al, A0 in a manner in which the amount of operation is reduced or in a manner in which the data capacity is reduced.

[0059] For example, the coefficient information 2c in which the coefficient A0 is output from the coefficient conversion section 2 is expressed by the following polynomial (3) as one example, and the coefficient A0 can be operated by inputting the q-axis current iq. Here, D4, D3, D2, Dl, D0 are coefficients of the respective terms.

[0060] A0(iq) = D4 iq4+ D3 iq3+ D2 iq2+ Dl iq+ D0 (3)

[0061] In this case, it is set to a 4th order with respect to the q-axis current iq, but it is not limited thereto, and an expression in which a trigonometric function or the like is combined can also be used.

[0062] In addition, in a case where it is difficult to express by the expression as described above, as shown in Figure 4 , the coefficient can be operated using an interpolation operation. Furthermore, Figure 4 is a diagram illustrating linear interpolation when the coefficient in the coefficient conversion section is determined.

[0063] As shown in Figure 4 , the A0 data of a plurality of points P1, P2, P3, P4, P5 on the waveform 10 in which the coefficient A0 is held is output by performing a linear interpolation operation on the line segment 10A connecting the respective points, with respect to the value of the input q-axis current iq. In this case, an example in which the linear interpolation is used is shown, but other spline interpolation or the like can also be used.

[0064] As described above, the manner of each coefficient information 2a, 2b, 2c and the operation method of the coefficients A2, A1, A0 are selected in consideration of the operation amount, storage capacity, and precision in accordance with the iron loss characteristics of the motor and the like.

[0065] Further, Figure 1 The data shape of each coefficient A2, A1, A0 in each coefficient information 2a, 2b, 2c illustrated in FIG. 2 is one example for explanation and is not limited.

[0066] In addition, in this embodiment, the polynomial function F(ω) is used as a second order function using three coefficients A2, A1, A0, but the polynomial function F(ω) is not limited to the second order and can be a first order or a polynomial function of three or more. In this case, the number of the power calculation sections 3a, 3b, 3c and the number of the coefficient information 2a, 2b, 2c held by the coefficient conversion section 2 also change in correspondence with the order.

[0067] Next, the operation of the correction section 4 will be described.

[0068] Figure 5 is a graph showing an example of the measurement result of the iron loss corresponding to the torque current of the motor. In this case, it is the measurement result of the iron loss of the induction motor explained in Figure 2 , the vertical axis is set to the iron loss value and the horizontal axis is set to the q-axis current iq. In this case, it is the measurement result after changing the d-axis current id under the condition of the first order frequency (high) shown in Figure 2 . The waveforms 15, 16, 17 of the iron loss corresponding to three stages where the d-axis current id is 80%, 100%, and 120% are shown. Further, the waveform 16 corresponds to the waveform 11 in Figure 2 .

[0069] The inventors of the present application found the phenomenon that, as shown in Figure 5 , if the d-axis current id is changed, the data shape of the iron loss with respect to the q-axis current iq is substantially maintained and the iron loss changes in correspondence with the increase and decrease of the d-axis current id. That is, the change width of the iron loss change in correspondence with the increase and decrease of the d-axis current id is substantially not changed by the q-axis current iq, and the correction operation in the correction section 4 utilizes this phenomenon.

[0070] The correction section 4 operates the change width ΔPIL when the q-axis current iq is 0%, and adds the change width ΔPIL to the iron loss value PILα of the operation result of the polynomial operation section 3, thereby generating the iron loss value PIL.

[0071] That is, the iron loss value PIL generated by the motor iron loss operation device 1 is represented by the following equation (4).

[0072] PIL = PILa+ ΔPIL · · · (4)

[0073] As described above, each of the coefficient information 2a, 2b, 2c is generated under the condition of one set value of the d-axis current id, in this case, the rated d-axis current idα of 100%, and thus the operation result of the polynomial operation section 3, that is, the iron loss value PILa is also a value under the condition of the d-axis current idα.

[0074] The change width ΔPIL of the iron loss of the d-axis current id from idα (100%) to the actual id is found by the following equation (5).

[0075] ΔPIL = (PILa0 / idα) · (id - idα) · · · (5)

[0076] Here, PILa0 is the iron loss at the condition of idα (100%), iq (0%), that is, the point Y, that is, the iron loss value PILa at iq (0%) is expressed by the following equation (6) based on the above equation (2).

[0077] PILa0 = A2(0) · ω2+ A1(0) · ω + A0(0) · · · (6)

[0078] As described above, the correction section 4 that performs the correction operation based on the d-axis current id is provided, and thus each of the coefficient information 2a, 2b, 2c can be generated and held only by one set value of the d-axis current id, and the number of conditions can be reduced to suppress the data capacity required for the iron loss operation.

[0079] The correction operation using the above equations (4) to (6) becomes no correction at id = idα (100%), and the iron loss value PILa input to the correction section 4 is still output as the iron loss value PIL. In addition, at id = iq = 0, the iron loss value PIL output from the correction section 4 is 0.

[0080] The motor iron loss operation device 1 according to the embodiment 1 operates the iron loss value PIL of the motor based on the one frequency ω and the q-axis current iq of the motor. As described above, the q-axis current iq that changes successively by the operation state (load state) of the motor is input to the motor iron loss operation device 1 to perform successive operation, and thus the iron loss that changes in accordance with the operation state (load state) of the motor can be operated with high accuracy.

[0081] Further, the motor iron loss operation device 1 has a polynomial operation section 3 in which a polynomial function F(ω) having a first frequency ω as a variable is set in advance, and a coefficient conversion section 2 that holds coefficient information 2a, 2b, 2c for each term of the polynomial function F(ω), and outputs coefficients A2, A1, A0 of each term of the polynomial function F(ω) with the q-axis current iq as an input. Further, the polynomial operation section 3 outputs the iron loss value PILα by substituting the coefficients A2, A1, A0 output from the coefficient conversion section 2 and the first frequency ω into the polynomial function F(ω).

[0082] As described above, the motor iron loss operation device 1 approximates the iron loss of the motor using the polynomial function F(ω) and the coefficients A2, A1, A0. Further, based on this approximation, the operation of estimating the iron loss value PILα is performed and output.

[0083] Therefore, by inputting only the first frequency ω and the q-axis current iq, the iron loss value PILα of high precision can be easily and reliably operated.

[0084] Further, the coefficient information 2a, 2b, 2c held by the coefficient conversion section 2 for each term of the polynomial function F(ω) is a function that generates one set value idα (100%) for the d-axis current id and makes one coefficient correspond to each value of the q-axis current iq. Further, the operation result of the polynomial function F(ω), that is, the iron loss value PILα is corrected in correspondence with the actual d-axis current id.

[0085] Therefore, both the operation amount and the data capacity of the motor iron loss operation device 1 can be reduced, miniaturization and high speed can be achieved, and a low-cost device structure can be provided.

[0086] Further, in the hardware that configures the motor iron loss operation device 1, for example, the processor 8 and the storage device 9 shown in FIG. 8 can be used in combination. Figure 6

[0087] The processor 8 executes a control program input from the storage device 9. The storage device 9 has a secondary storage device and a volatile storage device. The control program is input to the processor 8 from the secondary storage device via the volatile storage device. The processor 8 outputs data such as operation results to the volatile storage device of the storage device 9, and saves these data in the secondary storage device via the volatile storage device as necessary.

[0088] Embodiment 2.

[0089] Figure 7 is a diagram showing the structure of the motor iron loss operation device according to Embodiment 2.

[0090] ​As shown in the figure, the motor iron loss operation device 1A outputs the iron loss value PIL with the fundamental frequency ω of the motor, the torque current of the motor, i.e., the q-axis current iq, and the excitation current, i.e., the d-axis current id as inputs.

[0091] The motor iron loss operation device 1A according to this embodiment 2 operates the iron loss resistance RIL based on the fundamental frequency ω of the motor and the q-axis current iq. In this case, the iron loss is expressed by the power loss through the iron loss resistance, and the iron loss resistance obtained by dividing the iron loss by the square of the current vector norm (id2+iq2) is operated.

[0092] Further, the iron loss value PILα is generated based on the iron loss resistance RIL, and then, similarly to the above-described embodiment 1, the correction section 4 performs a correction operation on the iron loss value PILα based on the d-axis current id, and outputs the iron loss value PIL.

[0093] Figure 8 is a graph showing an example of the iron loss resistance corresponding to the torque current of the motor. In this case, the iron loss shown in the graph of the above-described embodiment 1 is divided by the square of the current vector norm (id2+iq2) to be converted into the iron loss resistance. Waveforms 21, 22, and 23 of the iron loss resistance corresponding to the three stages of high, medium, and low of the fundamental frequency ω are shown. The waveforms 21, 22, and 23 of the iron loss resistance correspond to the respective waveforms 11, 12, and 13 of the iron loss in Figure 2 Figure 2

[0094] As shown in the graph of Figure 8 , the iron loss resistance rises in correspondence with the increase of the fundamental frequency ω, and the shape changes in correspondence with the torque current iq, and the iron loss resistance is uniquely determined by only the q-axis current iq in the condition that the fundamental frequency is fixed. In addition, the value is different between the motoring condition and the regenerative condition, the peak value of the iron loss resistance is slightly shifted to the motoring side, and the like, and the shape becomes asymmetric with respect to the torque current iq.

[0095] In addition, the iron loss value PIL is generated based on the iron loss resistance RIL, and then, similarly to the above-described embodiment 1, the correction section 4 performs a correction operation on the iron loss value PIL based on the d-axis current id, and outputs the iron loss value PIL. Figure 9 is a graph showing an example of the iron loss resistance corresponding to the torque current of the motor, in which the iron loss shown in the graph of the above-described embodiment 1 is divided by the square of the current vector norm (id2+iq2) to be converted into the iron loss resistance. Waveforms 26, 25, and 24 of the iron loss resistance corresponding to the three stages of 80%, 100%, and 120% of the d-axis current id are shown. In addition, the waveform 25 corresponds to the waveform 21 in Figure 5 Figure 8

[0096] ​​​​The polynomial operation section 3, like the above-described embodiment 1, sets in advance a polynomial function G(ω) that takes the 1st frequency ω as a variable. Further, the polynomial function G(ω) is output with the coefficients B2, B1, B0 of each term and the 1st frequency ω as inputs. In this case, since the polynomial function G(ω) uses a 2nd function, the 2nd power, the 1st power, and the 0th power (=1) are respectively calculated in the power calculation sections 3a, 3b, 3c. Further, the outputs of the respective power calculation sections 3a, 3b, 3c are multiplied by the coefficients B2, B1, B0 and added, and the sum of the 3 terms is output.

[0097] In this case, G(ω) is expressed by the following equation (7).

[0098] G(ω) = B2- ω2+ B1- ω + B0- · · · (7)

[0099] The coefficient conversion section 5 holds the coefficient information 5a, 5b, 5c for each term of the polynomial function G(ω). Further, the q-axis current iq is input, and the coefficients B2(iq), B1(iq), B0(iq) of each term of the polynomial function G(ω) are output with reference to the respective coefficient information 5a, 5b, 5c. Note that B2(iq), B1(iq), B0(iq) are the values of the respective coefficients B2, B1, B0 corresponding to the value of iq.

[0100] In the polynomial operation section 3, the coefficients B2(iq), B1(iq), B0(iq) output from the coefficient conversion section 5 and the 1st frequency ω are substituted into the set G(ω), and the operation result, i.e., the iron loss resistance RIL, is output. That is, RIL becomes the following equation (8).

[0101] RIL = B2(iq) - ω2+ B1(iq) - ω + B0(iq) - · · · (8)

[0102] The respective coefficient information 5a, 5b, 5c held by the coefficient conversion section 5, like the above-described embodiment 1, is a function in which each of the respective coefficients B2, B1, B0 corresponds to one value for a set value of the d-axis current id, with the q-axis current iq as an argument. The function of the respective coefficient information 5a, 5b, 5c is not limited to being expressed by an equation, but can be expressed by a data table or a data correspondence chart.

[0103] In this case, as one set value of the d-axis current id, the rated d-axis current idα of 100% is used, and the respective coefficient information 5a, 5b, 5c is generated by collecting in advance information of the iron loss resistance value based on the measured value of the iron loss of the target motor under the condition of the d-axis current idα and stored in the coefficient conversion section 5.

[0104] The motor iron loss operation device 1A multiplies the operation result of the polynomial operation section 3, that is, the iron loss resistance RIL by the square of the current vector norm at the time of the rated d-axis current idα (idα2+ iq2) to generate the iron loss value PILα. That is, the iron loss value PILα is represented by the following equation (9).

[0105] PILα= RIL · (idα2+ iq2) · · · (9)

[0106] The correction section 4, like the above-described embodiment 1, operates the variation amplitude ΔPIL at the time of the q-axis current iq of 0%, and adds the variation amplitude ΔPIL to the iron loss value PILα to thereby generate the iron loss value PIL.

[0107] That is, the iron loss value PIL generated by the motor iron loss operation device 1 is represented by the following equation (10).

[0108] PIL= PILα+ ΔPIL

[0109] = PILα+ (PILα0 / idα) · (id- idα) · · · (10)

[0110] Here, PILα0 is the iron loss at the point Y (refer to Figure 5 ) of the condition of idα (100%), iq (0%), that is, the iron loss value PILα at the time of iq (0%), and is represented by the following equation (11) based on the above-described equations (8), (9).

[0111] PILα0= (B2(0) · ω2+ B1(0) · ω + B0(0)) · (idα2) · · · (11)

[0112] As described above, the motor iron loss operation device 1A related to this embodiment 2 operates the iron loss resistance RIL based on the first-order frequency ω and the q-axis current iq which changes successively by the operation state (load state) of the motor, and operates the iron loss value PIL based on the iron loss resistance RIL. Therefore, the iron loss which changes in correspondence with the operation state of the motor can be operated with high precision.

[0113] In addition, the motor iron loss operation device 1A has the polynomial operation section 3 which is provided with the polynomial function G(ω) which sets the first-order frequency ω as a variable in advance, and the coefficient conversion section 5 which outputs the coefficients B2, B1, B0 of each term of the polynomial function G(ω) as input of the q-axis current iq to operate the iron loss resistance RIL.

[0114] As described above, the motor iron loss operation device 1A approximates the iron loss resistance of the motor using the polynomial function G(ω) and the coefficients B2, B1, B0. Moreover, based on this approximation, an operation to estimate the iron loss resistance RIL is performed and output.

[0115] Therefore, by inputting the frequency ω and the q-axis current iq only once, the iron loss resistance RIL can be easily and reliably operated with high precision.

[0116] In addition, the coefficient conversion section 5 generates the coefficient information 5a, 5b, 5c held for each term of the polynomial function G(ω) for one set value idα (100%) of the d-axis current id, and functions in which one coefficient corresponds to each value of the q-axis current iq. Moreover, the iron loss PILα of the iron loss resistance RIL based on the operation result of the polynomial function G(ω) is corrected in correspondence with the actual d-axis current id.

[0117] Therefore, it is possible to reduce both the operation amount and the data capacity of the motor iron loss operation device 1A, and to provide a device structure that achieves miniaturization, high speed, and low cost.

[0118] In addition, in the above-described embodiment 2, the iron loss is expressed by the power loss of the iron loss resistance. In this case, under the condition of the set d-axis current id, the information of the iron loss resistance of the target motor is collected in advance to generate each coefficient information 5a, 5b, 5c of the coefficient conversion section 5. Depending on the motor model, the iron loss resistance is used, and the number of terms of the polynomial function G(ω) or the function of each coefficient information 5a, 5b, 5c is suppressed to be low, and a high-precision operation result (RIL) can be obtained. Therefore, it is possible to further reduce the operation amount and the data capacity of the motor iron loss operation device 1A, and to further achieve miniaturization and high speed.

[0119] Moreover, in a case where the calculated iron loss PIL is used for torque control of the motor, depending on the motor model or the operating condition, the precision of the operation is improved by using an operation using the value of the iron loss resistance RIL.

[0120] Furthermore, the iron loss is expressed by the power loss of the iron loss resistance, and therefore the iron loss value becomes zero under the condition of current zero, and cannot be applied to the iron loss operation of a permanent magnet motor in which the magnet flux is generated even if the current is zero.

[0121] Embodiment 3.

[0122] Figure 10 is a diagram showing the structure of the motor iron loss operation device according to Embodiment 3.

[0123] As shown in the figure, the motor iron loss operation device IB outputs an iron loss value PIL with the fundamental frequency ω of the motor, the torque current of the motor, i.e., the q-axis current iq, and the field current, i.e., the d-axis current id, as inputs.

[0124] In the above-described embodiments 1 and 2, the iron loss PIL is operated based on the approximate expression using the polynomial functions F(ω), G(ω) and the coefficients of the fundamental frequency ω, but in this embodiment 3, the iron loss PIL is operated using a function with the fundamental frequency ω and the q-axis current iq as two variables.

[0125] As shown in the figure, the motor iron loss operation device IB has a function operation section 6 and a correction section 4. Figure 10

[0126] The function operation section 6 is provided with a function with the fundamental frequency ω and the q-axis current iq as two variables in advance and holds it. Further, the fundamental frequency ω and the q-axis current iq are inputted, and the operation result of the function, i.e., the iron loss value PILα, is outputted.

[0127] Here, the function held by the function operation section 6 is generated for one set value of the d-axis current id, and one iron loss value PILα corresponds to the function for each combination of the fundamental frequency ω and the q-axis current iq. The function is information expressed by a 2-dimensional data correspondence chart, a 2-dimensional data table, or the like.

[0128] Further, as shown in the figure, the function operation section 6 is provided with a function with the fundamental frequency ω and the q-axis current iq as two variables in advance and holds it. Further, the fundamental frequency ω and the q-axis current iq are inputted, and the operation result of the function, i.e., the iron loss value PILα, is outputted. Figure 10

[0129] Figure 11 is a figure for explaining the held information of the function operation section 6, and shows a 2-dimensional data table. One iron loss value (α11 to α55) corresponds to each combination of the fundamental frequency ω and the q-axis current iq.

[0130] In this case, as one set value of the d-axis current id, the rated d-axis current idα of 100% is used, and under the condition of the d-axis current idα, the measured information of the iron loss of the target motor is collected in advance to generate the function, which is stored in the function operation section 6.

[0131] In addition, the interpolation operation can be used as in the above-described embodiment 1, and one example of linear interpolation will be described below. Figure 12 is a figure for explaining the linear interpolation in the function operation section 6.

[0132] ​​As shown in the figure, the iron loss value involved in each combination of the 1st frequency ω and the q-axis current iq is the data of each lattice point (intersection point). In the case where the iron loss value of point 6P is calculated, the data of 4 points, i.e., points A, B, C, and D, around point 6P are extracted, and the iron loss value of point E, which is the data of the same iq as point 6P, is calculated, for example, in the following manner. First, linear interpolation is performed on the line connecting points A and B, and the iron loss value of point E, which is the data of the same iq as point 6P, is calculated. Next, linear interpolation is performed on the line connecting points C and D, and the iron loss value of point F, which is the data of the same iq as point 6P, is calculated. Furthermore, linear interpolation is performed on the line connecting points E and F, and the iron loss value of point 6P, which is the data of the same ω as point 6P, is calculated.

[0133] Here, linear interpolation is used as an example, but other interpolation methods can be used.

[0134] As described above, the manner in which the function operation section 6 holds the information as a function and the operation method are selected in consideration of the calculation amount, the storage capacity, and the precision, based on the iron loss characteristics of the motor and the like.

[0135] The correction section 4, like the above-described embodiment 1, calculates the variation amplitude ΔPIL when the q-axis current iq is 0%, and adds the variation amplitude ΔPIL to the iron loss value PILα, thereby generating the iron loss value PIL.

[0136] That is, the iron loss value PIL generated by the motor iron loss operation device 1 is represented by the following equation (12).

[0137] PIL = PILα + ΔPIL

[0138] = PILα + (PILα0 / idα) · (id - idα) · · · (12)

[0139] Here, PILα0 is the iron loss at point Y (refer to Figure 5 ) which is the condition of idα (100%) and iq (0%), i.e., the iron loss value PILα at iq (0%), and is calculated by the function operation section 6.

[0140] As described above, the motor iron loss operation device IB according to the above-described embodiment 3 calculates the iron loss value PIL based on the 1st frequency ω and the q-axis current iq, which changes in accordance with the operation state (load state) of the motor, and thus, like the above-described embodiment 1, can calculate the iron loss, which changes in accordance with the operation state of the motor, with high precision.

[0141] In addition, the function operation section 6, which is provided, holds a function in which the 1st frequency ω and the q-axis current iq are set as two variables, and outputs the iron loss value PILα.

[0142] As described above, the motor iron loss operation device IB performs an operation of estimating the iron loss value PILα by approximating the iron loss of the motor using the information held by the function operation section 6 as a function, and outputs the iron loss value PILα.

[0143] Therefore, by inputting the frequency ω and the q-axis current iq only once, the iron loss value PILα can be easily and reliably operated with high precision.

[0144] Further, the information held by the function operation section 6 as a function is generated for one set value idα (100%) of the d-axis current id, and is a function in which one coefficient corresponds to each value of the q-axis current iq. Moreover, the operation result of the function operation section 6, that is, the iron loss value PILα is corrected in correspondence with the actual d-axis current id.

[0145] Therefore, it is possible to reduce both the operation amount and the data capacity of the motor iron loss operation device IB, and to provide a device structure that achieves miniaturization, high speed, and low cost.

[0146] Further, depending on the model of the motor or the current condition, the iron loss or the iron loss resistance of the motor varies complicatedly, and in particular, is greatly affected by magnetic flux saturation under a large current condition.

[0147] In this case, as in Embodiments 1 and 2 described above, in the approximate expression using the polynomial functions F(ω), G(ω) of the first order frequency ω and the coefficients, in order to ensure the operation precision, the number of the polynomial functions and the increase in the coefficient information are sometimes required, and the operation amount and the data capacity increase. Further, in order to ensure the precision of the calculation of the power of the first order frequency ω, the following problems occur. If the number of the power increases, the processing value becomes very large, and on the contrary, the coefficient involved in the power term becomes very small, and in particular, it is necessary to consider the case where a fixed-point microcomputer or the like is used.

[0148] In this embodiment 3, the above problems do not occur, and the iron loss is operated using a function expressed by a 2-dimensional data corresponding graph or a 2-dimensional data table or the like. Therefore, the number of data held is reduced, and the interpolation operation such as linear interpolation is effectively used, and thus it is possible to balance the suppression of the data capacity and the precision of the iron loss operation.

[0149] Further, in this embodiment 3, as in Embodiment 2 described above, the iron loss resistance RIL is operated, and the iron loss value PILα can be generated from the iron loss resistance RIL, and the iron loss value PILα is generated based on Figure 13 As shown below.

[0150] As Figure 13As shown, the motor iron loss operation device 1C has the function operation section 7 and the correction section 4. The function operation section 7 holds a function that sets the 1st frequency ω and the q-axis current iq as two variables in advance. Moreover, the function operation section 7 outputs the operation result of the function, that is, the iron loss resistance RIL with the 1st frequency ω and the q-axis current iq as inputs.

[0151] Here, the function held by the function operation section 7 is generated for one set value (idα(100%)) of the d-axis current id, and is a function that makes the value of one iron loss resistance RIL correspond to each combination of the 1st frequency ω and the q-axis current iq. The function is information indicated by a 2-dimensional data correspondence chart, a 2-dimensional data table, or the like.

[0152] Moreover, the motor iron loss operation device 1C, like the above-described embodiment 2, generates the iron loss value PILα by multiplying the operation result of the function operation section 7, that is, the iron loss resistance RIL by the square (idα^2+iq^2) of the current vector norm in the rated d-axis current idα. Then, the correction section 4 operates the variation amplitude ΔPIL when the q-axis current iq is 0%, and adds the variation amplitude ΔPIL to the iron loss value PILα, thereby generating the iron loss value PIL.

[0153] Thus, the motor iron loss operation device 1C obtains the same effect as the above-described embodiment 3, and has the advantage that the operation amount and the data capacity can be reduced according to the model of the motor by using the iron loss resistance RIL, and obtains the same effect as the above-described embodiment 2.

[0154] Embodiment 4.

[0155] In this embodiment 4, the correction operation of the correction section 4 shown in the above-described embodiments 1 to 3 is described.

[0156] Figure 14 is a graph that describes the correction operation corresponding to the field current involved in the embodiment 4. In addition, Figure 15 to 17 are graphs that describe the correction operation corresponding to the field current involved in another example of the embodiment 4, respectively.

[0157] In Figure 14 to 17 , in any case, a solid line waveform of the iron loss when the d-axis current id is set to the reference idα(100%) and a broken line waveform of the iron loss obtained by the correction operation when the d-axis current id is 80% and 120% are shown.

[0158] In the above-described embodiment 1, the correction operation of the correction section 4 is shown by the above-described equations (4) to (6), and the change width ΔPIL of the iron loss from idα(100%) to the actual id of the axis current id is added to the iron loss value PILα to correct it by the above-described equation (5), and the iron loss value PIL is output.

[0159] In this case, the iron loss value PIL output from the correction section 4 is 0 at id = iq = 0, and the iron loss changes in accordance with the waveform of a linear function fl shown in Figure 14 In this case, the condition of idα(100%), iq(0%) that is the point Y (refer to Figure 5 ) corresponds to the point Yl.

[0160] In addition, the iron loss under the condition of iq = 0 can also change in accordance with the waveform of a linear function f2 shown in Figure 15 Here, the condition of idα(100%), iq(0%) that is the point Y (refer to Figure 5 ) corresponds to the point Y2.

[0161] In this case, the change width ΔPIL of the iron loss from idα(100%) to the actual id is expressed with respect to the change amount of id using an arbitrary value α, an arbitrary offset β, by the following equation (13).

[0162] ΔPIL = α · (id - idα) + β · · · (13)

[0163] In addition, the iron loss under the condition of iq = 0 can also change in accordance with the waveform of a linear function f2 shown in Figure 16 Here, the condition of idα(100%), iq(0%) that is the point Y (refer to Figure 5 ) corresponds to the point Y3.

[0164] In this case, the iron loss changes in accordance with the waveform of a quadratic function f3 shown in

[0165] In this case, the change width ΔPIL of the iron loss from idα(100%) to the actual id is expressed with respect to the change amount of id using an arbitrary value α, an arbitrary offset β, by the following equation (13).

[0166] In addition, the iron loss under the condition of iq = 0 can also change in accordance with the waveform of a linear function f2 shown in

[0167] In addition, the iron loss under the condition of iq = 0 can also change in accordance with the waveform of a linear function f2 shown in Figure 17The waveform of the 2nd order function f4 shown changes. Here, the condition of idα (100%), iq (0%) is the point Y (refer to Figure 5 ) which corresponds to the point Y4.

[0168] In this case, the change range ΔPIL of the iron loss from idα (100%) to the actual id is expressed by the following equation (15) using arbitrary values γ, δ, ε.

[0169] ΔPIL = γ (id2- idα2) + δ (id2) + ε · · · (15)

[0170] As for the operation of the change range ΔPIL of the iron loss added to the iron loss value PILα, the above equation (5) explained in Embodiment 1 is the most simple and does not use parameters of arbitrary values. On the other hand, in the other equations (13), (14), (15), the parameters increase, but according to the iron loss characteristics of the motor, the iron loss operation accuracy can be improved. At the time of obtaining the iron loss data, the d-axis current id is appropriately changed and measured, and an appropriate equation is selected.

[0171] In addition, in the induction motor, the value obtained by multiplying the d-axis current id by the mutual inductance becomes a value corresponding to the 2nd order flux Φdr. In each of the above embodiments, the correction operation in the correction unit 4 based on the d-axis current id is shown, but instead of the d-axis current id, the 2nd order flux Φdr can be used to obtain the same effect. In this case, in the above equations (5), (13), (14), (15), if the d-axis current id is replaced by the 2nd order flux Φdr and the rated d-axis current idα (100%) is replaced by the rated 2nd order flux Φdrα (100%), the equation configuration can be directly followed.

[0172] According to the model of the motor, the saturation of the mutual inductance is significant, and the correction operation based on the 2nd order flux Φdr can sometimes ensure the operation accuracy of the iron loss. In addition, in the case where the iron loss PIL output from the motor iron loss operation device is used for the control of the induction motor, when the 2nd order flux Φdr is controlled in the induction motor, by using the correction operation of the 2nd order flux Φdr, the control design and installation become easy.

[0173] In addition, in the case of the synchronous motor, in the above equations (5), (13), (14), (15), the d-axis current id is replaced by the d-axis flux (Φe + Ld · id) and the rated d-axis current idα (100%) is replaced by Φe, respectively, and the correction operation can be similarly performed. Furthermore, Φe is the rotor induced voltage, and Ld is the d-axis inductance.

[0174] Embodiment 5.

[0175] Figure 18 is a view showing the structure of the motor iron loss operation device according to Embodiment 5. In Embodiments 1 and 2 described above, the iron loss PIL is operated based on the approximate expression of the polynomial functions F(ω), G(ω) and the coefficients using the first frequency ω. In this Embodiment 5, the iron loss PIL is approximately expressed by the polynomial function H(ω, id) using the first frequency ω and the d-axis current id or the second magnetic flux Φdr as variables.

[0176] As shown in FIG. 1D, the motor iron loss operation device 1D has a polynomial operation section 50 and a coefficient conversion section 20. In this embodiment, unlike each of the above-described Embodiments 1 to 4, the correction section 4 is not provided. Figure 18

[0177] The polynomial operation section 50 is provided with the polynomial function H(ω, id) using the first frequency ω, the d-axis current id or the second magnetic flux Φdr as variables.

[0178] Further, the polynomial operation section 50 outputs the operation result of the polynomial function H(ω, id), that is, the iron loss PIL, using the coefficients of each term of the polynomial function H(ω, id), the first frequency ω and the d-axis current id as inputs.

[0179] As an example, the first frequency ω and the d-axis current id are expressed by the polynomial function H(ω, id) and the iron loss PIL using the polynomial function H(ω, id) up to the second term by the following equation (16).

[0180] PIL = F(ω)

[0181] = CA- ω2- id2+ CB- ω2- id+ CC- ω- id2+ CD- ω- id+ CE- id2+ CF- id+ CG- ω2+ CH- ω+ CI (16)

[0182] Further, CA, CB, CC, CD, CE, CF, CG, CH and CI are the coefficients of each term of the polynomial function H(ω, id) output from the coefficient conversion section 20.

[0183] ​The coefficient conversion section 20 holds, in this case, nine pieces of coefficient information 20a, 20b, 20c, 20d, 20e, 20f, 20g, 20h, 20i, for each term of H(ω, id). Further, with the q-axis current iq as input, the coefficient conversion section 20 outputs the coefficients CA(iq), CB(iq), CC(iq), CD(iq), CE(iq), CF(iq), CG(iq), CH(iq), CI(iq) of each term of the polynomial function H(ω, id) with reference to each piece of coefficient information 20a to 20i. Further, CA(iq) to CI(iq) are the values of each coefficient CA to CI corresponding to the value of iq.

[0184] Each piece of coefficient information 20a to 20i held by the coefficient conversion section 20 is a function in which the q-axis current iq is the argument and each of the coefficients CA, CB, CC, CD, CE, CF, CG, CH, CI corresponds to one value. The function of each piece of coefficient information 20a to 20i is not limited to being expressed by an equation, but can be expressed by a data table or a data correspondence chart, and stored by collecting in advance the measured information of the iron loss of the target motor.

[0185] In this embodiment 5, too, the iron loss that varies in accordance with the operating condition (load condition) of the motor can be calculated with high precision, and the same effects as in the above-described embodiment 1 are obtained. Further, the polynomial function H(ω, id) involving two variables is used, so that the number of coefficients held by the coefficient conversion section 20 increases with an increase in the number of terms, and the data capacity or the amount of calculation increases, but the calculation precision of the iron loss PIL improves. Further, the d-axis current id is used for the calculation in advance, so that the correction section 4 is not needed.

[0186] Further, instead of the d-axis current id of the polynomial function H(ω, id), the second-order magnetic flux Φdr can be used in the induction motor.

[0187] In the case of the synchronous motor, if the d-axis magnetic flux (Φe + Ld · id) is used and the coefficient information of the coefficient conversion section 20 is prepared appropriately, the iron loss can be calculated with high precision as well.

[0188] Further, the iron loss resistance RIL can be calculated by a polynomial function in which the first-order frequency ω and the d-axis current id are two variables, and the iron loss PIL is generated from the iron loss resistance RIL.

[0189] Embodiment 6.

[0190] Figure 19 is a diagram showing the structure of the motor control device according to Embodiment 6.

[0191] In this embodiment, a motor control device that performs torque control of a motor is shown, which has the motor iron loss calculation device 1 related to Embodiment 1 described above.

[0192] As shown in the figure, a motor control device (hereinafter, simply referred to as a control device 32) controls the motor 30 by performing drive control of the inverter 31. In the control device 32, a control calculation process is performed, and voltage commands Vu*, Vv*, and Vw* are output to the inverter 31. In the inverter 31, a PWM process is performed, and based on the PWM process, switching elements are driven, and output voltages according to the voltage commands Vu*, Vv*, and Vw* are supplied to the motor 30. A current sensor 39 detects currents iu, iv, and iw of the motor 30, and sends signals of the detected currents to the control device 32.

[0193] The control device 32 has a current control section 33, a torque compensation section 34, a torque estimation section 35, coordinate conversion sections 36 and 37, an integrator 38, and the motor iron loss calculation device 1. The coordinate conversion sections 36 and 37 perform coordinate conversion between a three-phase stationary coordinate and a dq-axis rotating coordinate in which a rotor flux direction of the motor 30 is set as a d-axis.

[0194] A primary frequency ω is integrated by the integrator 38, and a phase θ for coordinate conversion is generated. The generated phase θ is input to the coordinate conversion sections 36 and 37. The primary frequency ω becomes an addition value of an estimated electrical angular velocity and a slip frequency or the addition value of the electrical angular velocity and the slip frequency in the case of an induction motor. In addition, in the case of a permanent magnet motor, the primary frequency ω becomes an electrical angular velocity or an estimated electrical angular velocity.

[0195] A torque command Te* is generated or a value of the torque command Te* that is set by using a control process is performed in a known method, so that the motor 30 becomes a desired rotational speed.

[0196] The torque compensation section 34 inputs the torque command Te* and the estimated torque Te, and performs a calculation process in a manner in which both are coincident, and outputs a torque current command, that is, a q-axis current command iq*. In addition, an excitation current command, that is, a d-axis current command id* is generated or a value of id* that is set by using a control process of a 2nd flux Φdr is performed in a known method.

[0197] The detected currents iu, iv, iw from the current sensor 39 are converted into currents id, iq on the dq-axis rotating coordinate by the coordinate conversion section 37, and input to the current control section 33. The current control section 33 inputs the detected currents id, iq on the dq-axis and the q-axis current command iq* and the d-axis current command id*, and performs arithmetic processing in such a manner that they coincide, and outputs the voltage commands Vd*, Vq* on the dq-axis. The voltage commands Vd*, Vq* are converted into the three-phase voltage commands Vu*, Vv*, Vw* by the coordinate conversion section 36, and output to the inverter 31.

[0198] The motor iron loss operation device 1 performs a presumptive operation of the iron loss PIL using the detected currents id, iq output from the coordinate conversion section 37 and the 1st frequency ω as inputs, and outputs the result. The torque presumptive section 35 performs a torque presumptive operation using the voltage commands Vd*, Vq*, the detected currents id, iq, the 1st frequency ω, the iron loss PIL, and motor constants, and operates the presumptive torque Te.

[0199] Through the above processing, the control device 32 achieves high-precision torque control.

[0200] As described above, the iron loss of the motor 30 changes with the q-axis current iq. Also, the torque of the motor 30 changes with the q-axis current iq, and also changes with the iron loss. Therefore, deriving a conversion table that uniquely associates the torque current iq and the torque requires a lot of work, but in this embodiment, a feedback loop is used after combining the torque presumption and the torque compensation, so the effect of not requiring derivation of the above conversion table is achieved.

[0201] The torque presumptive section 35 operates the presumptive torque Te by the following equation (17) in the case where the motor 30 is an induction motor. In this case, the dq-axis voltage commands Vd*, Vq* are used instead of the output voltage of the inverter 31.

[0202] Te = (P / ω) ((Vd* · id + Vq* · iq) - Rs · (id2+ iq2) - PIL)...(17)

[0203] where Vd*: d-axis voltage command, Vq*: q-axis voltage command, id: d-axis current of stator (1st), iq: q-axis current of stator (1st), Rs: 1st resistance, PIL: iron loss, P: number of pole pairs, ω: 1st frequency

[0204] This equation (17) is based on the input and output of energy, the 1st term becomes the input power to the motor 30, the 2nd term becomes the copper loss, and the 3rd term becomes the iron loss.

[0205] In particular, in the case where the motor 30 is an induction motor, the second-order copper loss added in the rotor is added to the frequency ω of 1 order. This will be described next. In the induction motor, the frequency ω of 1 order is the sum of the electrical angular frequency ωreand the slip frequency ωse, and the slip frequency ωseand the frequency ω of 1 order are given by the following equations (18) and (19), respectively. If these are substituted into the above (17), the following equation (20) is obtained, and the product of the mechanical speed ωrmand the estimated torque Te, that is, the mechanical output Pmech, is equivalent to a value obtained by subtracting the first-order copper loss, the second-order copper loss, and the iron loss from the input power. "M / Lr" in the 4th term of equation (20) is a coefficient for converting the primary side and the secondary side.

[0206] ωse= (M-Rr / Lr) - (iq / Φdr) = (Rr / Lr) - (iq / id) · · · (18)

[0207] ω = ωre+ ωse= P- ωrm+ ωse· · · (19)

[0208] Pmech= Te- ωrm

[0209] = (Vd* - id + Vq* - iq) - Rs- (id2+ iq2) - PIL- Rr- (M2 / Lr2) - iq2

[0210] = (Vd* - id + Vq* - iq) - Rs- (id2+ iq2) - PIL- Rr- iqr2· · · (20)

[0211] where iqr: q-axis current of rotor (2nd order), Lr: 2nd order inductance, Rr: 2nd order resistance, PIL: iron loss, P: number of pole pairs, ωrm: mechanical angular speed, ωse: slip frequency, Lr: 2nd order inductance, M: mutual inductance

[0212] In the calculation of the above equation (17) of the torque estimation section 35, only the first-order resistance Rs is used as the motor constant, and the motor inductance and the second-order resistance are not used. Therefore, there is an advantage that the variation with respect to the value of the other motor constants is robust. The first-order resistance Rs is easily determined based on the Ohm's law from the relationship between the voltage command and the detected current. Therefore, the higher the operation accuracy of the iron loss PIL is, the higher the torque estimation accuracy is.

[0213] The motor iron loss operation device 1 according to the above-described embodiment 1 can operate with high accuracy in accordance with the operation state of the motor, and therefore the accuracy of the torque estimation by the torque estimation section 35 can be improved.

[0214] In addition, while the motor 30 is limited to an induction motor, the torque estimation can be performed using the following Equation (21) that includes the iron loss resistance Rm to calculate the second-order flux. In this case, the torque estimation shown in Equation (23) is performed by adding a delay through a low-pass filter shown in Equation (22) to the calculated second-order flux.

[0215] [Equation 1]

[0216]

[0217]

[0218] Te = P - (M / Lr) - (Φdr(LPF) - iq - Φqr(LPF) - id)... (23)

[0219] where Φdr: rotor (second-order) d-axis flux, Φqr: rotor (second-order) q-axis flux, σ: leakage coefficient, Ls: first-order inductance, Φdr(LPF): rotor (second-order) d-axis flux after low-pass filter processing, Φqr(LPF): rotor (second-order) q-axis flux after low-pass filter processing, s: Laplace variable

[0220] It is known that the second-order flux of the induction motor has a delay of a second-order time constant (Rr-Lr) with respect to the current on the stator side. By simulating this, torque estimation along with transient changes in the torque of the induction motor can be performed. In other words, even in the case of an operating condition of the motor in which the second-order flux dynamically changes, such as field weakening control, when the inverter output voltage is saturated, high-precision torque estimation that follows changes in the torque can be performed.

[0221] In addition, in the above Equation (23), in the case where the delay characteristics added through Equation (22) are ignored, that is, in the case where Φdr(LPF) = Φdr and Φqr(LPF) = Φqr, if the right side of Equation (21) is substituted into the right side of Equation (23) and is transformed, the above Equation (17) is obtained, which is equivalent to the energy equation.

[0222] The calculation of the above Equation (23) requires the calculation of the second-order flux and the iron loss resistance Rm, and the amount of calculation increases, but as described above, it has the effect of enabling torque estimation corresponding to the transient characteristics of the induction motor.

[0223] In the torque compensation section 34, the torque command Te* and the estimated torque Te are input, and an operation process shown in the following equation (24) is performed to generate the q-axis current command iq*. Equation (24) is an equation in the case where the motor 30 is an induction motor, and the first term on the right side is a known theoretical formula for converting the torque command Te* to the q-axis current command iq*. In this case, it becomes a feedforward term, and the response of the torque compensation control is improved. The second and third terms are feedback control terms for implementing PI control, and the accuracy of the torque compensation control is achieved.

[0224] iq* = (1 / (P-(M / Lr)-Φdr)) - Te* + Kp-(Te* - Te) + (Ki / s) -(Te* - Te)...(24)

[0225] However, Kp: proportional gain of the torque compensation control, Ki: integral gain of the torque compensation control

[0226] By the operation process of this equation (24), the torque command Te* and the estimated torque Te are made to coincide, and high-accuracy control of the output torque of the motor 30 can be achieved.

[0227] On the other hand, in the case where the motor 30 is a synchronous motor, the torque estimation section 35 performs an operation on the estimated torque Te by the following equation (25). In the case of the synchronous motor, since there is no slip frequency, the first-order frequency ω is equivalent to the electrical angular frequency ωre of the motor.

[0228] Te = (P / ω) ((Vd* - id + Vq* - iq) - R-(id2 + iq2) - PIL)...(25)

[0229] where Vd*: d-axis voltage command, Vq*: q-axis voltage command, id: d-axis current of the stator, iq: q-axis current of the stator, R: winding resistance, PIL: iron loss, P: number of pole pairs, ω: first-order frequency (electrical angular frequency)

[0230] In the calculation of the above equation (25), the motor constant used is only the winding resistance R, and as in the case of the calculation using the above equation (17), robustness with respect to variations in the values of other motor constants can be achieved. In addition, the motor iron loss operation device 1 can perform the operation with high accuracy in accordance with the operating conditions of the motor, and thus the accuracy of the torque estimation can be improved.

[0231] Furthermore, in the case where the motor 30 is a synchronous motor, in the torque compensation section 34, the torque command Te* and the estimated torque Te are input, and an operation process shown in the following equation (26) is performed to generate the q-axis current command iq*.

[0232] iq* = (1 / (P-Φe)) - Te* + Kp- (Te* - Te) + (Ki / s) - (Te* - Te) • • • (26)

[0233] However, Φe: Induced voltage constant of the rotor

[0234] As described above, in this embodiment, the motor control device 32 is configured with the motor iron loss operation device 1 related to the above-described embodiment 1. Also, the iron loss value PIL calculated by the motor iron loss operation device 1 is utilized in the operation of estimating the torque Te, and the torque compensation control is performed based on the estimated torque Te. Therefore, high-precision torque control corresponding to the load (torque) condition of the motor 30 can be achieved. In particular, the high precision of the iron loss operation achieved by the motor iron loss operation device 1 contributes to the improvement of the precision of the torque control.

[0235] Further, in this embodiment, the motor iron loss operation device 1 related to the above-described embodiment 1 is applied to the motor control device 32, but the motor iron loss operation devices 1A to 1D related to the other embodiments 2 to 5 can also be applied similarly, to obtain the same effects.

[0236] Embodiment 7.

[0237] Figure 20 is a diagram showing the structure of the motor control device related to the embodiment 7.

[0238] In this embodiment, a motor control device that has the motor iron loss operation device 1 related to the above-described embodiment 1 and performs overload protection of the motor is shown. The same reference numerals are attached to the same structural elements as those of the above-described embodiment 6, and the description is appropriately omitted.

[0239] As shown in the figure, the motor control device (hereinafter, simply referred to as the control device 32A) has a current control section 33, coordinate conversion sections 36, 37, an integrator 38, a copper loss operation section 40, a protection section 41 that performs overload protection, and the motor iron loss operation device 1.

[0240] In this case, the torque compensation is not performed, and therefore the q-axis current command iq* for the normal operation of the motor 30 is generated with a generation section not shown.

[0241] The copper loss operation section 40 operates the copper loss PCL based on the detected currents id, iq on the dq axes. In the case where the motor 30 is an induction motor, the copper loss PCL is operated using the following equation (27), and in the case where the motor 30 is a synchronous motor, the copper loss PCL is operated using the following equation (28).

[0242] PCL = Rs■ (id2+ iq2) + Rr■ iqr2... (27)

[0243] PCL = R■ (id2+ iq2)... (28)

[0244] Further, in this case, the copper loss PCL is used for over-load protection to prevent the stator from burning, and thus the second term in the above equation (27) representing the rotor copper loss can be omitted.

[0245] The iron loss PIL calculated by the motor iron loss calculation device 1 and the copper loss PCL are added together as the motor loss PL, which is input to the protection section 41.

[0246] In the protection section 41, the motor loss PL is input, and the motor loss PL is compared with a threshold value set in advance, and if the threshold value is exceeded, a protection process for preventing over-load is performed. The protection process performs a process of reducing the current of the motor 30 or stopping energization of the motor 30.

[0247] Further, the protection section 41 can input the motor loss PL to a thermal circuit network model of the motor 30 to estimate the motor temperature, and if the estimated motor temperature exceeds a set value, the protection process is performed.

[0248] As described above, in this embodiment, the motor control device 32A is configured with the motor iron loss calculation device 1 described in Embodiment 1. Further, the motor loss PL is calculated using the iron loss value PIL calculated by the motor iron loss calculation device 1, and over-load protection is performed. Therefore, the motor loss PL can be calculated with high precision, and the reliability of the over-load protection is improved.

[0249] Further, in this embodiment, the motor iron loss calculation devices 1A to 1D described in Embodiments 2 to 5 can also be applied similarly to obtain the same effects.

[0250] In addition, the over-load protection described in this embodiment can also be applied to Embodiment 6 described above to obtain both the improvement in the precision of the torque control and the improvement in the reliability of the over-load protection.

[0251] Various features, modes, and functions described in one or more embodiments are not limited to the application of a specific embodiment, and can be applied individually or in various combinations to the embodiments.

[0252] Therefore, numerous modifications are contemplated within the scope of the technology disclosed. For example, cases where at least one structural element is modified, added, or omitted, and cases where at least one structural element is extracted and combined with structural elements of other embodiments are included.

[0253] Explanation of Reference Signs

[0254] 1, 1A-D motor iron loss operation device, 2 coefficient conversion section, 2a, 2b, 2c coefficient information, 3 polynomial operation section, 4 correction section, 5 coefficient conversion section, 5a, 5b, 5c coefficient information, 6, 7 function operation section, 20 coefficient conversion section, 20a-20i coefficient information, 30 motor, 32, 32A motor control device, 34 torque compensation section, 35 torque estimation section, 36, 37 coordinate conversion section, 41 protection section, 50 polynomial operation section, A0, A1, A2 coefficients, B0, B1, B2 coefficients, CA, CB, CC, CD, CE, CF, CG, CH, CI coefficients, F(ω), G(ω), H(ω, id) polynomial functions, id d-axis current, iq q-axis current, PIL, PILα iron loss value, RIL iron loss resistance, Te estimated torque, Te* torque command, Vd*, Vq* dq-axis voltage commands, ω1st frequency, Φdr 2nd magnetic flux, θ phase.

Claims

1. An electric motor iron loss operation device that operates an iron loss value of an electric motor based on a fundamental frequency of the electric motor and a q-axis current component in a dq-axis rotating coordinate in which a rotor magnetic flux direction of the electric motor is set as a d-axis, comprising: a polynomial operation section that sets a polynomial function with the fundamental frequency as a variable in advance, outputs a result of operation of the polynomial function with a coefficient of each term of the polynomial function and the fundamental frequency as inputs; and a coefficient conversion section that holds coefficient information for each term of the polynomial function, outputs the coefficient for each term of the polynomial function with reference to the coefficient information with the q-axis current component as an input, the polynomial operation section outputs the result of operation with the coefficient output from the coefficient conversion section and the fundamental frequency substituted into the polynomial function, and an iron loss value of the electric motor is operated based on the result of operation.

2. The electric motor iron loss operation device according to claim 1, wherein the coefficient information held for each term of the polynomial function is a function that generates one of the coefficients corresponding to each value of the q-axis current component with one set value of a d-axis current component.

3. The electric motor iron loss operation device according to claim 1 or 2, wherein the polynomial operation section outputs an iron loss resistance as the result of operation.

4. The electric motor iron loss operation device according to any one of claims 1 to 3, wherein an iron loss value operated is correction-operated in correspondence with a d-axis current component or an electric motor magnetic flux.

5. The electric motor iron loss operation device according to claim 1, wherein the polynomial function is a two-variable function that further sets a d-axis current component or a second-order magnetic flux as a variable, and the polynomial operation section outputs the result of operation with the d-axis current component or the second-order magnetic flux further substituted into the polynomial function.

6. The electric motor iron loss operation device according to claim 5, wherein the polynomial operation section outputs an iron loss resistance as the result of operation.

7. An electric motor iron loss operation device that operates an iron loss value of an electric motor based on a fundamental frequency of the electric motor and a q-axis current component in a dq-axis rotating coordinate in which a rotor magnetic flux direction of the electric motor is set as a d-axis, wherein a correction operation is performed with respect to an iron loss value operated by adding an iron loss correction amount calculated using a d-axis current component or an electric motor magnetic flux.

8. The electric motor iron loss operation device according to claim 7, wherein a function operation section that sets and holds a function with the fundamental frequency and the q-axis current component as two variables in advance outputs a result of operation of the function with the fundamental frequency and the q-axis current component as inputs, and an iron loss value of the electric motor is operated based on the result of operation.

9. The electric motor iron loss operation device according to claim 8, wherein the function held by the function operation section is generated with one set value of a d-axis current component. ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ ​ 10. The motor iron loss operation device according to claim 8 or 9, wherein the function operation section outputs the iron loss resistance as the operation result.

11. A motor control device that detects a current in a dq-axis rotating coordinate in which a rotor flux direction of a motor is set as a d-axis, generates a voltage command in a manner that the current follows a current command value, in the motor control device, a phase of the dq-axis rotating coordinate that is operated based on the imparted 1st frequency is used for coordinate conversion between the dq-axis rotating coordinate and a three-phase stationary coordinate in both the detected current and the voltage command, operates an iron loss value of the motor with the motor iron loss operation device described in any one of claims 1 to 10.

12. The motor control device according to claim 11, wherein has: a torque estimation section that operates an estimated torque based on the voltage command, the detected current, the 1st frequency, and the operated iron loss value; and a torque compensation section that operates a q-axis current command in the command value in a manner that the estimated torque coincides with an imparted torque command.

13. The motor control device according to claim 11 or 12, wherein has a protection section that performs overload protection based on the operated iron loss value.

Images