Monitoring device for motor

Abstract

A monitoring device is provided with: a storage unit that stores a first model of a motor when the operating state of the motor is normal; and a model search unit that generates a second model of the motor when the motor operation state is abnormal. When the difference between the measured temperature of the temperature detector and the estimated temperature of the temperature detector based on the first model is out of the determination range, the model search unit generates a second model in which a coefficient related to heat transfer between the stator core model of the first model and the external air model is changed. The estimated temperature of the temperature detector based on the second model corresponds to the measured temperature of the temperature detector.

Description

Technical Field

[0001] The present invention relates to a monitoring device for an electric motor. Background Art

[0002] It is known that an electric motor is arranged in a machine in order to make the components move. When the electric motor is driven, the stator core, the coil fixed to the stator core, the bearings, etc. generate heat, and the temperature of the motor rises. When the temperature of the motor becomes too high, there are cases where the motor cannot move accurately, or the components of the motor are damaged, or the life of the motor is shortened. Therefore, it is preferable to cool the motor while the motor is driven. The electric motor can be cooled by the flow of natural air. Alternatively, the electric motor can be cooled by supplying air using a cooler or the like (for example, Japanese Patent Gazette No. 2015-167436).

[0003] The actual temperature when the motor is driven can be detected by a temperature detector installed on the motor components. Alternatively, there is a known simulation for estimating the motor temperature (for example, Japanese Patent Publication No. 11-262102). In particular, in order to estimate the motor temperature, there is a known method of using a thermal model that takes into account the thermal capacity of the motor components and the heat transfer between the components (for example, International Publication No. 2022 / 085543). In the thermal model, a heat transfer coefficient or thermal resistance can be set between each component to calculate the heat transfer between the components. The temperature of each component can be calculated in the thermal model.

[0004] Prior art literature

[0005] Patent Literature

[0006] Patent Document 1: Japanese Patent Application Publication No. 2015-167436

[0007] Patent Document 2: Japanese Patent Application Laid-Open No. 11-262102

[0008] Patent Document 3: International Publication No. 2022 / 085543

[0009] Patent Document 4: Japanese Patent Application Publication No. 2010-287971

[0010] Patent Document 5: Japanese Patent Application Publication No. 2006-311735 Summary of the invention

[0011] Problems to be solved by the invention

[0012] During the operation of the motor, the blower supplies air to the motor, thereby effectively cooling the motor. Here, due to the long-term use of the motor, the performance of cooling the motor may be reduced. For example, the blower may fail, or the air holes for circulating air may be clogged due to foreign matter such as dust, or sludge may accumulate on the surface of the motor. If the motor continues to operate in a state where the cooling performance of the motor is reduced, the motor may become hot.

[0013] As a method for detecting a blower failure, a temperature detector may be installed on the blower, or a magnetic sensor may be installed on the fan to detect the fan speed of the blower. As a method for detecting a blockage of the air hole of the blower, a flow sensor may be installed to detect the air flow rate flowing inside the blower.

[0014] Alternatively, the operator can visually inspect the motor regularly to detect a malfunction of the blower, blockage of the air holes of the blower, accumulation of sludge, etc. If the blower is abnormal, the blower can be repaired or replaced. If foreign matter is attached to the air holes or sludge is accumulated, the motor can be cleaned.

[0015] However, when using a sensor such as a magnetic sensor, a temperature sensor, or a flow sensor to detect the fan speed of the blower, there is a problem that a new sensor must be installed on the motor in order to detect the reduction in the cooling performance of the motor. Alternatively, in a method in which an operator regularly visually inspects the motor, it is necessary to disassemble the device including the motor, which causes the operator to waste energy.

[0016] Means for solving problems

[0017] The monitoring device of the motor of the disclosed embodiment includes: a state acquisition unit that acquires the motor operation state including the measured temperature detected by the temperature detector installed in the motor. The monitoring device includes: a temperature estimation unit that calculates the estimated temperature of the temperature detector based on the motor model. The monitoring device includes: a storage unit that stores a first model of the motor when the motor operation state is normal. The monitoring device includes: a model search unit that generates a second model of the motor when the motor operation state is abnormal. The monitoring device includes: a motor component model including a stator core and an external air model. A heat capacity is set for at least one component model. A coefficient related to heat transfer is set between the component models of the motor and between the stator core model and the external air model. When the difference between the measured temperature of the temperature detector and the estimated temperature of the temperature detector based on the first model is out of a predetermined determination range, the model search unit generates a second model in which the coefficient related to heat transfer between the stator core model and the external air model of the first model is changed so that the estimated temperature of the temperature detector based on the second model corresponds to the measured temperature. The performance estimating unit calculates a variable related to the cooling performance of the electric motor based on a coefficient related to heat transfer between the stator core model and the outside air model in the first model and a coefficient related to heat transfer between the stator core model and the outside air model in the second model.

[0018] Effects of the Invention

[0019] According to the monitoring device for an electric motor according to the aspect of the present disclosure, it is possible to provide information on the cooling performance of the electric motor. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a block diagram of a monitoring device for a machine and a motor in an embodiment.

[0021] Figure 2 It is a schematic cross-sectional view of the first electric motor in the embodiment.

[0022] Figure 3 This is a first model of the first electric motor in the embodiment.

[0023] Figure 4 This is a graph for explaining the motor operation pattern for setting parameters in the first model of the motor.

[0024] Figure 5 This is a first model of the second electric motor in the embodiment.

[0025] Figure 6 This is a graph of simulation results using the parameters set by the parameter calculation unit.

[0026] Figure 7 This is a graph showing the temperature change of the temperature detector when the motor is operating normally.

[0027] Figure 8 This is a graph showing the temperature change of the temperature detector when the motor operating state is abnormal.

[0028] Fig. 9 This is a second model of the first electric motor in the embodiment. DETAILED DESCRIPTION

[0029] Reference Figures 1 to 9 The motor monitoring device in the embodiment will be described. The motor monitoring device in the embodiment has a function of estimating the temperature of a predetermined component of the motor using a thermal model of the motor.

[0030] In this embodiment, the temperature actually measured by the temperature detector is referred to as the measured temperature. In addition, the temperature of the component estimated using the motor model is referred to as the estimated temperature. In this embodiment, an example of estimating the temperature of a temperature detector installed in a stator coil among the components of the motor is described. The motor monitoring device of this embodiment calculates a variable related to the cooling performance of the motor being cooled. In addition, the monitoring device displays information related to the cooling performance of the motor on a display unit or notifies other devices.

[0031] Figure 1 This is a block diagram of a monitoring device for a machine and a motor in this embodiment. The machine 1 in this embodiment has: a motor 10 that drives the components of the machine 1, and a machine control device 41 that controls the machine 1. The machine control device 41 in this embodiment has a processing device (computer). The machine control device 41 includes a CPU (Central Processing Unit) as a processor. The machine control device 41 has: a RAM (Random Access Memory) and a ROM (Read Only Memory) connected to the CPU via a bus.

[0032] The machine 1 of this embodiment is a numerical control type machine. The machine 1 is driven according to the command statements recorded in the pre-made action program 45. The machine control device 41 includes: a storage unit 42 that stores the action program 45, and an action control unit 43 that generates an action command for the motor 10 according to the action program 45. The machine 1 includes a drive device 44, and the drive device 44 includes: a circuit that supplies power to the motor 10 according to the action command generated by the action control unit 43. The motor 10 is driven by the power supplied by the drive device 44.

[0033] The storage unit 42 may be composed of a non-temporary storage medium capable of storing information. For example, the storage unit 42 may be composed of a storage medium such as a volatile memory, a non-volatile memory, a magnetic storage medium, or an optical storage medium. The action control unit 43 is equivalent to a processor driven according to the action program 45. The processor reads the action program 45 and implements the control specified in the action program 45, thereby functioning as the action control unit 43.

[0034] As such a machine 1, any machine having a motor 10 can be used. For example, a machine tool for processing a workpiece can be exemplified as the machine 1. As the motor 10, a spindle motor for rotating a tool or a feed axis motor for moving a table or a spindle head along a predetermined coordinate axis can be exemplified.

[0035] Figure 2 2 is a cross-sectional view of the first electric motor of the present embodiment. Figure 1 and Figure 2 The first motor 10 is a synchronous motor having a magnet 18 on the rotor 11. The motor 10 has a rotor 11 and a stator 12. The stator 12 includes a stator core 20 formed of a magnetic material and a coil 16 fixed to the stator core 20. The stator core 20 is formed of, for example, a plurality of magnetic steel plates stacked in the axial direction of the shaft 13. The coil 16 includes, for example, a winding wound on the stator core 20 and a resin portion that fixes the winding.

[0036] The rotor 11 is fixed to a rod-shaped shaft 13. The rotor 11 is fixed to the outer peripheral surface of the shaft 13 and includes a rotor core 17 formed of a magnetic material and a plurality of magnets 18 fixed to the rotor core 17. The magnets 18 of this embodiment are permanent magnets.

[0037] The shaft 13 is connected to other components to transmit rotational force. The shaft 13 rotates around the rotation axis RA. The shaft 13 is supported by the main shafts 14 and 15 as bearings. In this embodiment, in the motor 10, the side where the shaft 13 is connected to other components is called the front side. In addition, the side opposite to the front side is called the rear side. Figure 2 In the example shown, arrow 91 indicates the front side of the electric motor 10 .

[0038] The motor 10 includes a housing 21 at the front side and a housing 22 at the rear side. The stator core 20 of the stator 12 is supported by the housings 21 and 22. The housing 21 supports the main shaft 14. A main shaft support member 24 supporting the main shaft 15 is fixed to the housing 22. The housings 21 and 22 support the shaft 13 rotatably via the main shafts 14 and 15. A rear cover 23 that closes the internal space of the housing 22 is fixed to the rear end of the housing 22.

[0039] A rotation position detector 32 for detecting the rotation position or rotation speed of the shaft 13 is disposed at the rear end of the shaft 13. The rotation position detector 32 of the present embodiment is constituted by an encoder. A temperature detector 31 for detecting the temperature of the coil 16 is fixed to the coil 16 of the stator 12. The temperature detector 31 of the present embodiment is constituted by a thermistor. The outputs of the temperature detector 31 and the rotation position detector 32 are input to the mechanical control device 41.

[0040] As the components of the motor 10, there can be exemplified the rotor 11, the rotor core 17, the magnet 18, the stator 12, the stator core 20, the coil 16, the housing 21, 22, the shaft 13, the rear cover 23, the main shaft support member 24, the main shafts 14, 15, the temperature detector 31, and the rotation position detector 32. As the components of the motor 10, there is no limitation to this form, and any part constituting the motor 10 can be adopted. For example, as the component of the motor 10, a frame covering the stator can also be adopted.

[0041] The motor 10 of this embodiment has a cooling machine for cooling the main body of the motor. The cooling machine of this embodiment includes a blower 29 for supplying cooling air. The blower 29 of this embodiment is fixed to the stator core 20 via a cylindrical member 25. The cylindrical member 25 is fixed to the stator core 20. The inside of the cylindrical member 25 constitutes an air flow path.

[0042] The blower 29 includes a cooling fan 27, a frame 28, and a motor for rotating the cooling fan 27. The blower 29 of this embodiment is arranged so that the rotation axis of the cooling fan 27 coincides with the rotation axis RA of the shaft 13. That is, the rotor 11 and the cooling fan 27 are arranged coaxially. The cooling fan 27 is not limited to this mode, and can be arranged at any position so that air contacts the motor body.

[0043] The stator core 20 of this embodiment has a through hole 26a for cooling air to flow. The through hole 26a penetrates from one end face of the stator core 20 to the other end face along the axial direction of the rotor 11. The housing 21 has a through hole 26b for air to flow. The through hole 26b is connected to the through hole 26a. In addition, an air hole 28a for air to flow is formed in the frame 28 of the blower 29.

[0044] When the blower 29 is driven, the cooling fan 27 rotates, and the air flows in the direction indicated by the arrow 91. The cooling air flows into the interior of the frame 28 from the air hole 28a of the frame 28. The cooling air circulates inside the frame 28 and inside the tubular member 25. The cooling air circulates in the air flow path between the housing 22 and the tubular member 25. The cooling air circulates in the through hole 26a of the stator core 20 and the through hole 26b of the housing 21 as indicated by the arrow 92. The cooling air cools the housing 22 on the rear side, the stator core 20, and the housing 21 on the front side. The flow of the cooling air is not limited to this method, and the cooling air may also flow in the direction opposite to the arrow 91.

[0045] The cooling machine for cooling the motor body of the present embodiment is composed of a blower, but is not limited to this method. The cooling machine can be any device that cools the motor body. For example, a cooling machine that uses cooling water to cool the motor or an electronic cooling machine including a Peltier element or the like can also be used.

[0046] The motor monitoring device 2 of the present embodiment calculates an estimated temperature obtained by estimating the temperature output from the temperature detector 31 disposed in the motor 10. In the present embodiment, the temperature output from the temperature detector 31 disposed in the coil 16 of the stator 12 is estimated. In particular, the motor monitoring device 2 estimates the change in the temperature of the temperature detector 31 with respect to the passage of time.

[0047] The monitoring device 2 for the motor has: an operation processing device (computer) including a CPU as a processor. The monitoring device 2 includes: a storage unit 51 for storing information related to the monitoring of the motor 10. The storage unit 51 can be composed of a non-temporary storage medium capable of storing information. For example, the storage unit 51 can be composed of a storage medium such as a volatile memory, a non-volatile memory, a magnetic storage medium, or an optical storage medium. The monitoring device 2 includes: a display unit 52 for displaying information related to the motor 10. The display unit 52 can be composed of any display panel such as a liquid crystal display panel or an organic EL (Electro Luminescence) display panel.

[0048] The monitoring device 2 includes a temperature estimating unit 53 that calculates an estimated temperature of the temperature detector 31. The temperature estimating unit 53 includes a loss calculating unit 54 that calculates the amount of heat generated by the primary copper loss of the coil 16 and the amount of heat generated by the iron loss of the stator core 20 based on the operation command of the motor 10. The temperature estimating unit 53 includes a temperature calculating unit 55 that estimates the temperature of the temperature detector 31 using a motor model (thermal model). The temperature calculating unit 55 calculates the estimated temperature of the temperature detector 31 based on the amount of heat generated by the primary copper loss and the iron loss, the heat capacity of each component model, and the coefficient of heat transfer correlation between the component models.

[0049] The monitoring device 2 in this embodiment has a function of calculating parameters included in the motor model. The parameters include: the heat capacity set for the component model of the motor 10, and the coefficient of heat transfer between the component models. The monitoring device 2 includes a parameter calculation unit 63, which calculates the parameters included in the first model of the motor when the motor operation state is normal.

[0050] The monitoring device 2 includes a state acquisition unit 62, which acquires the operating state of the motor 10 when the motor 10 is actually driven. The operating state of the motor 10 includes: a measured temperature detected by a temperature detector 31 installed on the motor 10. The operating state of the motor 10 includes: an operating command of the motor 10 generated by actually driving the motor 10, and a rotation speed output from the rotation position detector 32. The operating command of the motor 10 can be obtained from the operation control unit 43. In addition, the state acquisition unit 62 can obtain the temperature of the outside air from the outside air temperature detector 33 that detects the ambient temperature of the machine 1. The outside air temperature detector 33 is configured to detect the temperature around the machine 1, for example.

[0051] The parameter calculation unit 63 of the present embodiment calculates parameters so that the change in the estimated temperature of the temperature detector calculated by the motor model corresponds to the change in the actual measured temperature. The parameter calculation unit 63 of the present embodiment can set the parameters of the motor model by machine learning.

[0052] The loss calculation unit 54 of the temperature estimation unit 53 calculates the heat generated by the coil 16 and the stator core 20 based on the operation command generated by the operation control unit 43 and the rotation speed detected by the rotation position detector 32. In addition, the temperature calculation unit 55 calculates the estimated temperature of the temperature detector based on the heat generated by the coil 16 and the stator core 20.

[0053] The parameter calculation unit 63 calculates the estimated temperature of the temperature detector 31 by using the temperature estimation unit 53. The parameter calculation unit 63 includes an evaluation unit 66 that evaluates the estimated temperature of the temperature detector 31 by comparing the estimated temperature of the temperature detector 31 with the measured temperature of the temperature detector 31 acquired by the state acquisition unit 62. The parameter calculation unit 63 includes a parameter change unit 67 that changes the value of the parameter according to the evaluation result of the evaluation unit 66.

[0054] The temperature estimation unit 53, the loss calculation unit 54, and the temperature calculation unit 55 are each equivalent to a processor driven according to a program. The state acquisition unit 62, the parameter calculation unit 63, the evaluation unit 66, and the parameter change unit 67 are each equivalent to a processor driven according to a program. The processor functions as each unit by implementing the control specified in the program.

[0055] Figure 3 It represents the first model obtained by modeling the heat transfer of the first motor in the present embodiment. The first model of the motor is a thermal model that simulates the normal operating state of the motor. The thermal model of the motor includes a plurality of component models. The thermal model includes parameters such as the heat capacity of the components and the coefficients related to the heat transfer between the components. The first model 10a of the motor in the present embodiment includes models of the main components constituting the first motor 10. The motor model 10a includes: a rotor model 11a, a stator core model 20a, and a coil model 16a wound on the stator core. In addition, the motor model 10a includes: a temperature detector model 31a for detecting the temperature of the coil 16.

[0056] Reference Figure 2 , an air layer is interposed between the rotor 11 and the stator core 20. Furthermore, an air layer is interposed between the rotor 11 and the coil 16. The motor model 10a in this embodiment includes an air layer model 35a. In addition, the motor model 10a includes an external air model 36a as a model of the air surrounding the motor 10. In this way, in the motor model of this embodiment, the air layer and the external air are generated as component models of the motor.

[0057] The temperature detected by the temperature detector 31 is substantially equal to the temperature of the coil 16. However, under specified conditions, due to the small heat capacity of the temperature detector 31, the temperature detected by the temperature detector 31 may be different from the temperature of the coil 16. Therefore, in the present embodiment, a temperature detector model 31a is also generated as one of the component models for the temperature detector 31. In addition, the heat capacity of the temperature detector 31 may be set to zero, and the temperature of the temperature detector model may be calculated to be the same as the temperature of the component model on which the temperature detector is mounted.

[0058] In the motor model 10a, a plurality of parameters including heat capacity and coefficients related to heat transfer are set. Heat capacity is set for at least one component model. Temperatures T1, T2, T3, T4, T5 as variables and heat capacities C1, C2, C3, C4, C5 as constants are set for each of the coil model 16a, the stator core model 20a, the air layer model 35a, the rotor model 11a, and the temperature detector model 31a. In addition, temperature T is set as a variable for the external air model 36a. r .

[0059] The heat of one component of the motor 10 is transferred to the other components. Coefficients related to heat transfer are set between the component models of the motor 10. As the coefficient related to heat transfer, a heat transfer coefficient or a coefficient obtained by multiplying the heat transfer coefficient by the contact area between the components can be used. In the example here, a coefficient obtained by multiplying the heat transfer coefficient by the contact area is specified.

[0060] A heat transfer-related coefficient ha is set between the stator core model 20a and the coil model 16a. A heat transfer-related coefficient hc1 is set between the air layer model 35a and the coil model 16a. A heat transfer-related coefficient hc2 is set between the air layer model 35a and the stator core model 20a. A heat transfer-related coefficient hc3 is set between the air layer model 35a and the rotor model 11a. A heat transfer-related coefficient hd is set between the coil model 16a and the temperature detector model 31a. Furthermore, in order to simulate heat release from the stator core 20 to the outside air, a heat transfer-related coefficient hb is set between the stator core model 20a and the outside air model 36a.

[0061] In the motor model 10a of the present embodiment, the primary copper loss P generated in the coil 16 of the stator 12 is considered as the heat generated by the components. c1 The heat generated by the primary copper loss is input to the coil model 16a. In addition, the iron loss P of the stator core 20 generated by the magnetic force of the magnet 18 of the rotor 11 is considered. i The amount of heat generated by the iron loss is input to the stator core model 20a.

[0062] Heat moves between components such as the coil and the stator core depending on the magnitude of the coefficient related to heat transfer. In addition, the temperature of each component rises or falls according to the difference between the input heat and the output heat. Figure 3 The temperature change rate of each component of the first model 10a of the first electric motor shown can be expressed by the following equations (1) to (5). In each component, the temperature change rate can be calculated by dividing the difference between the input heat amount and the output heat amount by the heat capacity.

[0063] [Mathematical formula 1]

[0064]

[0065]

[0066]

[0067]

[0068]

[0069] T1: Coil temperature C1: Coil heat capacity

[0070] T2: Temperature of stator core C2: Heat capacity of stator core

[0071] T3: Temperature of the air layer C3: Heat capacity of the air layer

[0072] T4: rotor temperature C4: rotor heat capacity

[0073] T5: Temperature of the temperature detector C5: Heat capacity of the temperature detector

[0074] T r : Temperature of outside air

[0075] ha, hb, hc1, hc2, hc3, hd: coefficients related to heat transfer

[0076] P c1 :Primary copper loss

[0077] P i : Iron loss

[0078] The heat capacities C1, C2, C3, C4, and C5 of the components are constants and can be determined in advance. The coefficients ha, hb, hc1, hc2, hc3, and hd related to heat transfer are coefficients obtained by multiplying the heat transfer coefficient by the contact area. The coefficients ha, hb, hc1, hc2, hc3, and hd are constants and can be determined in advance. The loss calculation unit 54 of the temperature estimation unit 53 calculates the primary copper loss P in the coil 16 as described later. c1 and the iron loss P in the stator core i The temperature calculation unit 55 of the temperature estimation unit 53 can calculate the temperature change amount in the minute time dt based on the above-mentioned equations (1) to (5).

[0079] Next, the primary copper loss P included in equations (1) and (2) is c1 and iron loss P i The calculation method of the motor 10 is described below. The rotation speed of the motor 10 and the load factor (ratio to the maximum load) of the motor 10 can be set in advance by the operator according to the operation performed by the machine. The loss calculation unit 54 of the temperature estimation unit 53 calculates the primary copper loss P c1 and iron loss P i Table 1 shows the loss mapping used to calculate the loss.

[0080] [Table 1]

[0081] Table 1 Loss mapping

[0082]

[0083] Table 1 shows the loss at maximum output, the loss at no load, and the current at maximum output relative to the rotation speed (rotation speed) of the motor 10. m It is the loss when the motor load factor is 100%, and is a value determined by the motor rotation speed. n It is the loss when the motor load factor is zero and depends on the motor rotation speed. The current I at maximum output m It is the current value when the load factor is 100% at each rotation speed. The loss map shown in Table 1 can be created by actually driving the motor. This loss map can be stored in advance in the storage unit 51 of the monitoring device 2, for example.

[0084] The loss calculation unit 54 calculates the primary copper loss P c1 and iron loss P i The total loss P t Total loss P t It can be calculated by the following formula (6) and formula (7).

[0085] [Mathematical formula 2]

[0086] P t =k2·LF 2 +k1·LF+P n …(6)

[0087] k1=P m -P n -k2…(7)

[0088] P t : Total loss

[0089] P m : Loss at maximum output

[0090] P n : Loss at no load

[0091] LF: Motor load factor

[0092] k1, k2: constants

[0093] Total loss P t The loss P at maximum output can be m , loss at no load P n The motor's rotation speed and load factor are determined, so the loss P at maximum output can be calculated according to Table 1. m and the loss P when no load n The constants k1 and k2 can be determined in advance by the operator. Next, the primary copper loss P c1It can be calculated by the following formula (8) and formula (9).

[0094] [Mathematical formula 3]

[0095] P c1 = r1·I 2 …(8)

[0096] I=I m ·LF…(9)

[0097] P c1 :Primary copper loss

[0098] I: Current

[0099] r1: primary resistance

[0100] I m : Current at maximum output

[0101] Primary copper loss P c1 This corresponds to the Joule heat of the current flowing in the coil 16. In addition, the current I flowing in the coil 16 can be calculated by comparing the current I at the maximum output m Multiply by the motor load factor LF to calculate. Current I at maximum output m It can be obtained from Table 1. Here, the primary resistance r1 of the coil 16 is measured in advance. Next, the iron loss P i It can be calculated by the following formula (10). Iron loss P i It can be obtained by t Subtract the primary copper loss P c1 to calculate.

[0102] [Formula 4]

[0103] P i =P t -P c1 …(10)

[0104] P i : Iron loss

[0105] The temperature estimation unit 53 obtains the motor operation mode including the rotation speed and the load factor for driving the machine 1 from the state acquisition unit 62. The temperature calculation unit 55 of the temperature estimation unit 53 may first set the temperatures T1 to T5 of each component to an arbitrary temperature. For example, the temperature calculation unit 55 sets the temperatures T1 to T5 of the component to the normal temperature T1 of the outside air. r The outside air temperature T r It can be determined in advance according to the place where the machine 1 is arranged.

[0106] The loss calculation unit 54 of the temperature estimation unit 53 calculates the primary copper loss and iron loss based on the rotation speed and the load factor of the motor in the operation mode. Then, the temperature calculation unit 55 can calculate the change in the temperature T5 of the temperature detector 31 in the small time dt by solving the above equations (1) to (5). In this way, the operator can define the motor operation mode and estimate the change in the temperature of the temperature detector over time when the motor is operated in the operation mode.

[0107] However, in the motor model 10a of the present embodiment, it is sufficient that the temperature of one of the multiple components of the motor can be estimated with high accuracy. The temperature of components other than the one component may deviate from the actual temperature. In the example here, it is sufficient that the temperature T5 of the temperature detector model 31a can be estimated with high accuracy. The temperature T1 of the coil model 16a, the temperature T2 of the stator core model 20a, the temperature T3 of the air layer model 35a, and the temperature T4 of the rotor model 11a may deviate from the actual temperature.

[0108] Furthermore, the heat capacities C1 to C5 set for the motor model 10a and the heat transfer-related coefficients ha, hb, hc1, hc2, hc3, and hd set between the components have inherent values ​​depending on the material, shape, and arrangement of the components. However, in the motor model 10a of the present embodiment, at least some of the parameters of the plurality of heat capacities and the plurality of heat transfer-related coefficients may be set to values ​​that deviate from the actual heat capacities or the actual heat transfer-related coefficients.

[0109] Each parameter is set so that the change of the temperature T5 of the temperature detector model 31a corresponds to the change of the actual temperature. For example, even if the temperature of the coil and the stator core deviates from the actual temperature, the parameters of the motor model can be set so that the temperature of the temperature detector represents a value close to the actual temperature. In addition, the coefficients related to heat capacity and heat transfer are calculated, and as a result, the total heat capacity of the components and the total heat transfer-related coefficients can correspond to the actual heat capacity and the actual heat transfer-related coefficients with high accuracy. And when the temperature estimation unit estimates the temperature of the components, the temperature of all the components can correspond to the actual temperature of the components with high accuracy.

[0110] The motor monitoring device 2 of the present embodiment is configured to be implemented by switching between the following modes: a normal modeling mode for calculating the parameters of the first model of the motor when the motor operation state is normal, and an abnormal modeling mode for generating the second model of the motor when the motor operation state is abnormal. In the normal modeling mode, parameters including coefficients related to heat transfer and thermal capacities of components are set in the first model 10a of the motor.

[0111] Reference Figure 1 , the parameter calculation unit 63 of the present embodiment generates a first model of the motor in the normal modeling mode. The parameter calculation unit 63 sets the heat capacity and heat transfer-related coefficients contained in the motor model 10a and the constants k1 and k2 in equations (6) and (7). The operator actually drives the motor 10 according to a predetermined operation mode. The state acquisition unit 62 acquires the load factor of the motor 10, the rotation speed of the motor 10, and the temperature output from the temperature detector 31 as the state of the motor 10. In addition, the state acquisition unit 62 acquires the temperature of the outside air from the outside air temperature detector 33.

[0112] Figure 4 A graph showing an operation pattern when the motor is driven in order to set the parameters included in the first model of the motor according to the present embodiment. Figure 4 1 shows an operation mode when there is no load. In this operation mode, no load is applied to the motor 10, and the rotation speed of the motor 10 is gradually increased. The load factor of the motor is temporarily increased at a predetermined time interval, thereby increasing the rotation speed of the motor 10.

[0113] The temperature detected by the temperature detector 31 gradually increases. At times t1 to t7, the rotation speed of the motor 10 is increased by temporarily increasing the load factor of the motor 10. The state acquisition unit 62 acquires the operating state of the motor 10 and the temperature output from the temperature detector 31 during the period of gradually increasing the rotation speed of the motor 10. More specifically, the state acquisition unit 62 acquires the load factor of the motor 10, the rotation speed of the motor 10, and the temperature output from the temperature detector 31 at predetermined short time intervals, and stores them in the storage unit 51. In the present embodiment, a fixed outside air temperature is used, but it is not limited to this method. The state acquisition unit 62 may also acquire the outside air temperature from the outside air temperature detector 33 at short time intervals.

[0114] Reference Figure 1 The state acquisition unit 62 acquires the torque command included in the motion command generated by the motion control unit 43 of the mechanical control device 41. The torque command corresponds to the load factor of the motor 10, so the state acquisition unit 62 can calculate the load factor based on the torque command.

[0115] The parameter calculation unit 63 calculates the parameters of the motor model 10a based on the variables obtained by the state acquisition unit 62. The parameter calculation unit 63 of the present embodiment calculates parameters including heat capacity C1, C2, C3, C4, C5 and heat transfer related coefficients ha, hb, hc1, hc2, hc3, hd based on the heat generated in the coil 16 and the stator core 20 and the temperature detected by the temperature detector 31. In addition, the parameter calculation unit 63 calculates constants k1 and k2 in equations (6) and (7) as parameters. The parameter calculation unit 63 calculates the parameters so that the estimated temperature change of the temperature detector model 31a during the simulation is close to the actual measured temperature change.

[0116] The parameter calculation unit 63 sets the initial value of each parameter. The initial value of the parameter can be set by any method. The parameter calculation unit 63 uses the loss calculation unit 54 to calculate the heat generated by the primary copper loss of the coil 16 and the heat generated by the iron loss of the stator core 20. The loss calculation unit 54 calculates the primary copper loss P based on the rotation speed of the motor 10 and the load factor of the motor 10 obtained by the state acquisition unit 62 using Table 1 and equations (6) to (10). c1 and iron loss P i .

[0117] Calculate the copper loss P c1 and iron loss P i The constants k1 and k2 are included in the equations (6) and (7). Furthermore, the loss calculation unit 54 calculates the loss of the predetermined micro-time dt, that is, the heat generated in the micro-time. In this way, the loss calculation unit 54 calculates the primary copper loss P in the equations (1) and (2) based on the measured value including the motor operation command (load factor) and the output of the rotation position detector 32. c1 And iron loss P i .

[0118] The parameter calculation unit 63 estimates the temperature of the components using the temperature calculation unit 55. The temperature calculation unit 55 calculates the estimated temperature of the temperature detector 31 based on the first model 10a of the motor using the various parameters and the loss calculated by the loss calculation unit 54. That is, the temperature of the temperature detector model 31a is estimated by simulation.

[0119] The temperature calculation unit 55 can calculate the change in the estimated temperature detected by the temperature detector 31 over time after the start of the drive of the motor 10 based on the temporarily set parameters. The temperature of each component model of the motor 10 can be calculated using the differential equations of the above-mentioned equations (1) to (5). The initial temperature value of each component model can be set to, for example, the outside air temperature when the drive of the motor 10 is started, that is, the room temperature.

[0120] The evaluation unit 66 of the parameter calculation unit 63 compares the temperature (estimated temperature) of the temperature detector model 31a calculated by the temperature calculation unit 55 with the measured temperature actually measured by the temperature detector 31, thereby evaluating the parameters temporarily set in the first model 10a of the motor. In the example here, the evaluation unit 66 does not evaluate variables other than the temperature of the temperature detector model 31a, and only evaluates the temperature of the temperature detector model 31a. As long as the temperature change of the temperature detector model 31a is close to the actual temperature change, at least part of the temperatures of other components are not evaluated.

[0121] Next, the parameter changing unit 67 of the parameter calculating unit 63 changes the parameters according to the evaluation result of the evaluating unit 66. Then, according to the changed parameters, the loss calculation by the loss calculating unit 54, the estimated temperature calculation of the temperature detector model by the temperature calculating unit 55, the evaluation by the evaluating unit 66, and the parameter change by the parameter changing unit 67 are repeatedly performed by the same calculation as above: the loss calculation by the loss calculating unit 54, the estimated temperature calculation of the temperature detector model by the temperature calculating unit 55, the evaluation by the evaluating unit 66, and the parameter change by the parameter changing unit 67. When the evaluation by the evaluating unit satisfies the predetermined conditions, the parameters can be determined as the final parameters.

[0122] Here, the number of multiple parameter combinations in the motor model 10a is very large. Multiple parameters can be determined by a machine learning method. For example, multiple parameters can be set by a Bayesian optimization method. In Bayesian optimization, an objective function as an evaluation object is generated for an explanatory variable including a parameter as an input. And, a parameter predicted to be the minimum or maximum of the objective function is searched and set. By repeatedly searching for this parameter, the optimal value of the parameter can be set. In addition, the range of setting each parameter can be predetermined.

[0123] In the example here, regarding the temperature of the temperature detector 31, the difference between the temperature (estimated temperature) of the temperature detector model 31a estimated by the first model 10a of the motor and the measured temperature detected by the actual temperature detector 31 is set as the objective function. That is, regarding the temperature of the temperature detector 31, the objective function can use the difference between the predicted value calculated from equations (1) to (5) based on the temporarily set parameters and the actual measured value detected by the temperature detector 31. As the objective function, for example, the average value of the difference in a small time period can be used. Then, the parameter changing unit 67 searches for the next parameter so that the objective function becomes smaller.

[0124] In Bayesian optimization, parameter search and parameter evaluation can be repeated. If the objective function is within a predetermined judgment range, the evaluation unit 66 can adopt the parameter value at that time. On the other hand, if the objective function deviates from the predetermined judgment range, the next parameter search can be performed. In the Bayesian optimization method, the search is performed while predicting the area where the solution exists, so the amount of calculation processing can be suppressed.

[0125] The parameters included in the motor model 10a can be set by any method other than setting the parameters based on Bayesian optimization. For example, the range of setting each parameter can be predetermined. The parameter changing unit 67 of the parameter calculation unit 63 randomly sets a plurality of parameters within the parameter range. The temperature calculation unit 55 estimates the temperature of the temperature detector model 31a based on the set parameters. The evaluation unit 66 can evaluate the set parameters based on the actual measured value of the temperature obtained from the temperature detector 31. Such a parameter setting method is called a random search method.

[0126] Alternatively, the parameter changing unit 67 may set the parameter at a predetermined interval within the range of the set parameter. The temperature calculating unit 55 estimates the temperature of the temperature detector model 31a using the set parameter. The evaluating unit 66 evaluates all combinations of discretely set parameters. This method is called a grid search method.

[0127] In the random search method or the grid search method, as in the Bayesian optimization method, the evaluation unit 66 can set the temperature of the temperature detector 31 as the evaluation object. If the objective function is within a predetermined determination range, the evaluation unit 66 can adopt the parameter value at that time. Alternatively, the evaluation unit 66 can adopt the parameter that is optimal for the objective function. The evaluation unit 66 can determine the parameter that is consistent with the estimated temperature of the temperature detector 31 and the measured temperature detected by the actual temperature detector 31 as the parameter in the motor model 10a.

[0128] In this embodiment, the parameters are set so that the temperature change detected by the temperature detector 31 can be estimated with high accuracy. In this embodiment, the temperature of the components other than the temperature detector 31 may deviate from the actual temperature. Therefore, in the parameter evaluation, only the temperature of the temperature detector that detects the coil temperature can be evaluated. Therefore, the parameters can be set in a short time with a small amount of calculation. The storage unit 51 can store the generated first model 10a of the motor in advance.

[0129] In the above-mentioned embodiment, as the operation mode of driving the motor 10 in order to set the parameters of the first model 10a of the motor, the operation under no load is shown, but it is not limited to this mode. When specifying the parameters of the first model 10a of the motor, it is preferable to operate the motor 10 in various operation modes to obtain the operating state of the motor 10. For example, an operation mode in which the load factor of the motor 10 is repeatedly increased and decreased can be adopted. The load factor of the motor 10 can be greatly changed to change the rotation speed of the motor. The temperature detected by the temperature detector 31 rises or falls sharply. In this way, an operation mode including a sharp temperature change of the motor can be adopted.

[0130] In the above embodiment, the coil including the winding is used as an example of the component of the motor for estimating the temperature, but the present invention is not limited to this method. As the component for calculating the estimated temperature, any component of the motor can be used. In addition, a temperature detector can be installed in the component for calculating the estimated temperature.

[0131] In the above embodiment, a synchronous motor having a permanent magnet in the rotor is described, but the present invention is not limited to this embodiment. The monitoring system in this embodiment can be applied to any motor. For example, the motor model in this embodiment can also be applied to an induction motor having no permanent magnet in the rotor.

[0132] Figure 5 A second motor model according to this embodiment is shown. Figure 5 The first model 30a is when the second motor is in a normal operating state. Here, the second motor is an induction motor. The rotor of the induction motor includes a cage conductor formed of stainless steel or copper. The cage conductor is fixed to the shaft and rotates integrally with the shaft. In the induction motor, an induced current flows inside the cage conductor by the magnetic force generated by the stator coil. A magnetic field is generated around the cage conductor, and the rotor rotates.

[0133] In an induction motor, the current flows in the cage conductor of the rotor, and thus secondary copper loss P is generated as a secondary loss. c2 The secondary loss is equivalent to the Joule heat caused by the current flowing in the cage conductor. In the second motor model 30a, heat is generated in the rotor due to the secondary copper loss. The heat capacity in the components of the second motor and the coefficients related to heat transfer between the components are the same as those in the first model 10a of the first motor.

[0134] Among the differential equations for temperature changes of the components in the second motor model 30a, the differential equation for calculating the temperature change of the rotor is different from that in the first model 10a of the first motor. The differential equation expressing the temperature change of the rotor is the following equation (11).

[0135] [Formula 5]

[0136]

[0137] P c2 : Secondary copper loss

[0138] In equation (11), the secondary copper loss P is added to equation (4) of the rotor model 11a of the first motor. c2 The differential equations that express the temperature changes of other coils, the stator core, the air layer, and the temperature detector are the same as those in the thermal model of the first motor.

[0139] The loss calculation unit 54 calculates the heat generated by the secondary copper loss generated in the rotor conductor. The loss calculation unit 54 estimates the current flowing in the cage conductor. The loss calculation unit 54 can calculate the secondary copper loss based on the current flowing in the conductor, the secondary resistance of the conductor, the inductance of the conductor, and the mutual inductance of the conductor, the stator, and the coil. The inductance, mutual inductance, and secondary resistance of the conductor can be determined in advance.

[0140] Total losses in induction motors P t and primary copper loss P c1 It can be calculated in the same way as the total loss and primary copper loss in a synchronous motor. In addition, the secondary copper loss P can be considered. c2 , the iron loss P is calculated by the following formula (12): i .

[0141] [Mathematical formula 6]

[0142] P i =P t -P c1 -P c2 …(12)

[0143] P i : Iron loss

[0144] In this way, the primary copper loss, the iron loss, and the secondary copper loss are calculated in the first model 30a of the second motor. The parameter calculation unit 63 can also calculate the parameters included in the first model 30a of the second motor when the motor operation state is normal by the same control as the first model 10a of the first motor in the first model 30a of the second motor. In addition, the temperature estimation unit 53 can calculate the estimated temperature of the components when the motor operation state is normal using the first model 30a of the second motor.

[0145] Figure 6 A graph showing the estimated temperature of the temperature detector estimated by the temperature estimating unit using the parameters calculated by the parameter calculating unit of the present embodiment. Here, an example of the second electric motor is shown. Figure 6 2 shows a graph when a simulation is performed using parameter group A and parameter group B having different values. Parameter group A and parameter group B are calculated by parameter calculation unit 63. Table 2 shows the parameters included in parameter group A and parameter group B.

[0146] [Table 2]

[0147] Table 2 Parameters

[0148] Unit of heat transfer related coefficient: [W / K]

[0149] Unit of mass: [kg]

[0150]

[0151]

[0152] Parameter group A and parameter group B are obtained by driving the second motor in different operation modes. Table 2 shows coefficients related to heat transfer obtained by multiplying the heat transfer coefficient between the components of the motor by the contact area. In addition, the heat capacity is calculated by multiplying the specific heat of the material of each component by the mass. The specific heat of each material can be determined in advance, so the mass m of the component used to calculate the heat capacity is shown in Table 2. If parameter group A is compared with parameter group B, it can be seen that some parameters such as the heat transfer-related coefficients hc2 and hd and the mass m4 of the rotor have greatly different values ​​between the two parameter groups A and B.

[0153] On the other hand, referring to Figure 6 It can be seen that the estimated temperature of the temperature detector calculated using parameter group B is very consistent with the estimated temperature of the temperature detector calculated using parameter group A. In particular, the temperature changes are very consistent during the period when the temperature rises and during the period when the temperature changes within the specified range. Figure 6 The temperature changes shown are very consistent with the temperature changes detected by the temperature detector 31 when the motor 10 is actually driven.

[0154] There are parameters with greatly different values ​​between parameter group A and parameter group B. Therefore, it is known that the value of at least one of parameter group A and parameter group B is different from the parameter group in the actual motor. In particular, it is known that at least some of the parameters of the plurality of heat capacities and the plurality of heat transfer-related coefficients are set to values ​​different from the actual heat capacity or the actual heat transfer-related coefficients. For example, it is known that at least one of the heat transfer-related coefficients hc2 of parameter group A and hc2 of parameter group B deviates from the actual heat transfer-related coefficient.

[0155] Thus, in the motor monitoring device of this embodiment, even if at least some of the parameters among the plurality of parameters are different from the actual values, the temperature of the temperature detector can be estimated with high accuracy. In addition, the parameter calculation unit of this embodiment can also set the parameters of such a motor model.

[0156] In addition, in the present embodiment, one temperature detector is installed on the motor, but the present invention is not limited to this embodiment. A plurality of temperature detectors may be installed on a plurality of components of the motor. The evaluation unit of the parameter calculation unit may compare the measured temperatures of the plurality of temperature detectors with the estimated temperatures calculated by simulation. The parameter changing unit may set the parameters of the first model of the motor so that the estimated temperatures of the plurality of components are close to the measured temperatures detected by the actual temperature detectors.

[0157] The more temperature detectors are installed in the motor, the closer the values ​​of the plurality of parameters of the first model of the motor can be to the actual values. In addition, the estimated temperature of the components of each motor can be made close to the actual measured temperature. When a plurality of temperature detectors are installed, the coefficients related to the total heat capacity and the total heat transfer can be substantially the same as the actual heat capacity and the actual heat transfer. In this case, when the temperature estimation unit estimates the temperature of the components, the temperature of all the components corresponds to the actual temperature of the components with high accuracy.

[0158] Next, the abnormal model generation mode for generating the second model of the motor when the motor operation state is abnormal is described. The motor monitoring device 2 of this embodiment creates the second model when the cooling performance of the motor changes. The monitoring device 2 provides information related to the cooling performance of the motor to the operator or other devices based on the second model of the motor. In addition, the monitoring device 2 determines whether the cooling state of the motor is reduced.

[0159] In the description of the abnormal model making mode, Figure 2 The first motor 10 and Figure 3 The first model 10a of the first motor shown is taken as an example, but is not limited to this method. Figure 5 The second motor model 30a shown can also be controlled in the same manner as the first motor. In addition, as in the normal modeling mode, the temperature detected by the temperature detector 31 attached to the coil 16 is used as an example for explanation.

[0160] Reference Figure 1 The motor monitoring device 2 of the present embodiment includes a model search unit 71 for generating a second model of the motor when the motor operation state is abnormal. The model search unit 71 includes a temperature determination unit 72 for determining whether the difference between the measured temperature of the temperature detector 31 and the estimated temperature of the temperature detector 31 based on the first model 10a is within a predetermined determination range.

[0161] The model search unit 71 includes a setting unit 73, which creates a second model 10b based on the first model 10a when the difference between the measured temperature of the temperature detector 31 and the estimated temperature of the temperature detector 31 based on the first model 10a is out of a predetermined determination range. When the difference between the estimated temperature of the temperature detector and the measured temperature is large, the setting unit 73 changes the coefficient related to heat transfer of a part of the first model 10a, and generates a second model 10b in which the estimated temperature corresponds to the measured temperature with high accuracy. In the present embodiment, the setting unit 73 generates the second model 10b in which the coefficient related to heat transfer between the stator core model 20a and the outside air model 36a of the first model 10a is changed so that the estimated temperature of the temperature detector 31 based on the second model 10b corresponds to the measured temperature.

[0162] The monitoring device 2 in this embodiment includes a performance estimating unit 74 for estimating the cooling performance of the motor 10. The performance estimating unit 74 calculates a variable CA related to the cooling performance of the motor based on a coefficient hb related to heat transfer between the stator core model 20a and the external air model 36a in the first model 10a and a coefficient hbx related to heat transfer between the stator core model 20a and the external air model 36a in the second model 10b.

[0163] The monitoring device 2 includes a notification unit 75 that notifies other devices of information on the cooling performance of the motor. The notification unit 75 notifies other devices of information on the second model 10b. In addition, the notification unit 75 of this embodiment notifies other devices of information on a decrease in the cooling performance of the motor.

[0164] The above-mentioned model search unit 71, temperature determination unit 72 and setting unit 73 are each equivalent to a processor driven according to a pre-made program. In addition, the performance estimation unit 74 and notification unit 75 are each equivalent to a processor driven according to a pre-made program. The processor reads the program and implements the control specified in the program, thereby functioning as each unit.

[0165] If the performance of cooling the motor is abnormal, the temperature of the motor may rise. For example, the blower 29 may fail, or the air hole 28a for air circulation may be clogged by foreign matter such as dust, or sludge may accumulate on the surface of the motor 10. In this embodiment, when the measured temperature of the temperature detector 31 installed on the motor deviates from the predetermined determination range based on the estimated temperature when the motor is in a normal operating state, it is determined that the cooling performance of the cooling motor is abnormal. In other words, it is determined that the motor is in an abnormal operating state.

[0166] Figure 7 A graph showing temperature changes when the first motor is in normal operation. Figure 1 , Figure 3 as well as Figure 7 The vertical axis of the graph is the temperature of the temperature detector 31 attached to the coil 16. The horizontal axis is the elapsed time. The solid line represents the measured temperature actually detected by the temperature detector 31. That is, it represents the measured temperature acquired by the state acquisition unit 62.

[0167] In the operation example here, power is supplied at time t0, and the load of the motor 10 increases sharply. As the current increases, the temperature of the temperature detector 31 increases sharply. At time t1, the current supplied to the motor 10 is reduced. The speed of the motor 10 is maintained at a substantially constant speed under no load. After time t1, the temperature of the temperature detector 31 gradually decreases over time.

[0168] exist Figure 7 In FIG. 1 , the estimated temperature of the temperature detector 31 estimated by the first model 10a of the first motor is indicated by a dotted line. The first model 10a of the motor is pre-generated in the normal model creation mode. It can be seen that when the motor operation state is normal, the estimated temperature of the temperature detector 31 estimated by the first model 10a of the motor corresponds to the measured temperature detected by the temperature detector 31 with high accuracy.

[0169] Figure 8 A graph showing temperature changes when the cooling performance of the first electric motor decreases. Figure 8 The operating conditions and Figure 7 The operating conditions are the same. Figure 1 , Figure 3 as well as Figure 8 , the load is increased from time t0 to time t1. At time t1, the load is set to zero. In the interval from time t0 to time t2, the cooling performance of the cooling motor is normal. That is, the motor operation state is normal. The measured temperature detected by the temperature detector 31 corresponds to the estimated temperature calculated by the first model 10a of the motor with high accuracy.

[0170] However, after time t2, the cooling state of the motor 10 becomes abnormal and the measured temperature rises. Here, an example is shown in which the blower 29 fails and the cooling fan 27 stops. The blower 29 fails, resulting in insufficient heat removal from the stator core 20 and the housings 21 and 22, and the temperature of the entire motor 10 rises. The temperature of the temperature detector 31 attached to the coil 16 also rises.

[0171] The model search unit 71 of the monitoring device 2 detects that an abnormality has occurred in the operating state of the motor 10 at time t2. The model search unit 71 sets an interval EP in which the abnormality has occurred in the motor 10. The model search unit 71 searches for a coefficient hbx obtained by changing the coefficient hb related to heat transfer between the stator core model 20a and the external air model 36a in the first model 10a so that the estimated temperature in the interval EP corresponds to the measured temperature with high accuracy. That is, the model search unit 71 generates a second model 10b including the coefficient hbx related to heat transfer.

[0172] The performance estimation unit 74 of the monitoring device 2 calculates a variable CA related to the cooling performance of the motor based on the coefficient hb related to heat transfer in the first model 10a and the coefficient hbx related to heat transfer in the second model 10b. The monitoring device 2 displays the information related to the cooling performance of the motor 10 on the display unit 52 to notify the operator. Alternatively, the notification unit 75 of the monitoring device 2 transmits the information related to the cooling performance of the motor to other devices. Next, each process is described in detail.

[0173] While the motor 10 is being driven, the state acquisition unit 62 of the monitoring device 2 acquires the motor operation state including the operation command of the motor 10 and the rotation speed output from the rotation position detector 32 at a predetermined time interval. The temperature estimation unit 53 calculates the estimated temperature of each component of the motor based on the operation state acquired by the state acquisition unit 62 and the first model 10a of the motor. In particular, in the present embodiment, the temperature estimation unit 53 calculates the estimated temperature of the temperature detector 31 using the first model 10a at a predetermined time interval.

[0174] In addition, the state acquisition unit 62 acquires the measured temperature output from the temperature detector 31 at a predetermined time interval. The storage unit 51 stores the time, the estimated temperature of the temperature detector 31 calculated by the first model 10a, and the measured temperature of the temperature detector 31. Such sampling of the operating state can be performed at intervals of, for example, 1 second.

[0175] The temperature determination unit 72 determines whether the difference between the estimated temperature of the temperature detector 31 and the measured temperature of the temperature detector 31 is out of a predetermined determination range. For example, the temperature determination unit 72 can calculate the average movement of the difference between the measured temperature and the estimated temperature. The temperature determination unit 72 can, for example, calculate the average movement over the past minute. And, when the average movement is out of a predetermined determination range, it can be determined that the difference between the measured temperature and the determination temperature is out of the determination range. The temperature determination unit 72 can determine that an abnormality has occurred in the motor operating state at the moment in the center of the average movement interval. Here, at the moment t2 when the measured temperature is substantially higher than the estimated temperature, it is determined that an abnormality has occurred in the motor.

[0176] The control of the temperature determination unit for determining the measured temperature is not limited to this method, and any control may be used to determine whether the difference between the measured temperature and the estimated temperature is out of the determination range. For example, the temperature determination unit determines whether the absolute value of the difference between the measured temperature and the estimated temperature is greater than a predetermined determination value. When the absolute value is greater than the predetermined determination value, it can be determined that the difference between the measured temperature and the actual temperature is out of the determination range.

[0177] Reference Figure 8 In order to create the second model 10b based on the first model 10a and the measured temperature of the temperature detector 31, the temperature determination unit 72 sets an interval EP of a predetermined time length from the time t2 when the abnormality occurs. The interval EP of this embodiment is an interval in which the cooling of the motor is insufficient. In the interval EP, the difference between the measured temperature and the estimated temperature is out of the determination range.

[0178] Fig. 9 The second model of the electric motor of the present embodiment is shown. In the second model 10b, the coefficient hbx related to the heat transfer between the stator core model 20a and the external air model 36a is changed according to the coefficient hb related to the heat transfer between the stator core model 20a and the external air model 36a in the first model 10a. The other parameters of the second model 10b are the same as those of the first model 10a.

[0179] Reference Figure 1 , Figure 3 , Figure 8 as well as Fig. 9 , the model search unit 71 generates Fig. 9 The second model 10b of the first motor shown. The setting unit 73 of the model search unit 71 sets the coefficient hbx related to heat transfer in the second model 10b in the interval EP so that the estimated temperature of the temperature detector model 31a of the second model 10b corresponds to the measured temperature of the temperature detector 31. In the present embodiment, only the coefficient hbx related to heat transfer is changed. The parameters of the components specified in the other first model 10a are used without change. In the present embodiment, the numerical range of the coefficient hbx related to heat transfer is predetermined.

[0180] The initial value of the coefficient hbx related to heat transfer between the stator core model 20a and the outside air model 36a can be any value. For example, an initial value obtained by subtracting a predetermined relative value from the coefficient hb related to heat transfer used in the first model can be used. The loss calculation unit 54 of the temperature estimation unit 53 obtains the motor operation state in the interval EP from the storage unit 51 to calculate the loss. The temperature calculation unit 55 of the temperature estimation unit 53 uses the second model 10b to calculate the estimated temperature of the temperature detector model 31a in the interval EP at a predetermined time interval.

[0181] The setting unit 73 can search for the coefficient hbx related to heat transfer by the same method as the parameter setting of the first model 10a. The setting unit 73 of this embodiment calculates the coefficient hbx related to heat transfer using the evaluation unit 66 and the parameter change unit 67 of the parameter calculation unit 63. The setting unit 73 can set the coefficient hbx related to heat transfer by a machine learning method, similar to the setting of multiple parameters of the first model 10a.

[0182] The evaluation unit 66 obtains the measured temperature of the temperature detector 31 in the interval EP from the storage unit 51. The evaluation unit 66 calculates the difference between the measured temperature of the temperature detector 31 and the estimated temperature in the interval EP. The evaluation unit 66 may calculate the difference between the measured temperature of the temperature detector 31 and the estimated temperature based on the second model 10b at the sampling time interval. The storage unit 51 stores the coefficient hbx related to the heat transfer of the second model 10b and the difference between the measured temperature of the temperature detector 31 and the estimated temperature in the interval EP.

[0183] The coefficient hbx related to heat transfer in the second model 10b can be set by the Bayesian optimization method. The difference between the temperature (estimated temperature) of the temperature detector model 31a estimated by the second model 10b of the motor and the measured temperature detected by the actual temperature detector 31 is set as the objective function. The evaluation unit 66 evaluates the coefficient hbx related to heat transfer temporarily set in the second model 10b of the motor. The parameter changing unit 67 searches for the next parameter to reduce the objective function. When the evaluation of the evaluation unit 66 satisfies a predetermined condition, it can be determined as the final parameter.

[0184] Alternatively, the parameter changing unit 67 may implement a grid search method in which the coefficient hbx related to heat transfer between the stator core model 20a and the outside air model 36a is changed at a predetermined interval within the range of the coefficient hbx related to heat transfer between the stator core model 20a and the outside air model 36a.

[0185] The parameter changing unit 67 generates a second model 10b in which the coefficient hbx related to heat transfer is set to a single value within a predetermined range. The temperature estimating unit 53 calculates the estimated temperature of the temperature detector model 31a for each predetermined time interval in the interval EP based on the second model 10b. Furthermore, the evaluating unit 66 calculates the difference between the measured temperature and the estimated temperature of the temperature detector for each predetermined time interval in the interval EP. The storage unit 51 stores the coefficient hbx related to heat transfer and the difference between the measured temperature and the actual temperature.

[0186] Next, the parameter changing unit 67 changes the coefficient hbx related to heat transfer at a predetermined interval. In addition, the temperature estimating unit 53 and the evaluating unit 66 repeatedly perform the same calculation in the interval EP and store the calculation results. The parameter changing unit 67 changes the coefficient hbx related to heat transfer little by little at a predetermined interval. The evaluating unit 66 calculates the difference between the measured temperature and the estimated temperature after the coefficient hbx related to heat transfer is changed. In this way, the calculation is performed using all the coefficients hbx related to heat transfer. For each coefficient hbx related to heat transfer, the storage unit 51 stores the difference between the measured temperature and the actual temperature in the interval EP.

[0187] The setting unit 73 selects the second model with the smallest difference between the measured temperature of the temperature detector 31 and the estimated temperature from among the plurality of calculation results stored in the storage unit 51. For example, in the interval EP, the second model 10b with the smallest square average of the difference between the measured temperature of the temperature detector 31 and the estimated temperature obtained at a predetermined time interval is used. The setting unit 73 obtains a parameter including a coefficient hbx related to heat transfer of the second model.

[0188] In the above-mentioned grid search method, all heat transfer-related coefficients hbx for each predetermined interval are set for calculation, but it is not limited to this method. For example, the allowable range of the difference between the estimated temperature and the measured temperature of the temperature detector can be predetermined. In addition, the setting unit 73 can also end the search of the second model when the difference between the estimated temperature and the measured temperature of the temperature detector is within the allowable range. For example, in the interval EP, when the square average of the difference between the estimated temperature and the measured temperature obtained at a predetermined time interval is less than a predetermined judgment value, the setting unit 73 can determine that the difference between the estimated temperature and the measured temperature of the temperature detector is within the allowable range. The setting unit 73 can use the second model that includes the coefficients related to heat transfer at this time.

[0189] In addition, the model search unit 71 can generate a second model in which the estimated temperature of the temperature detector corresponds to the measured temperature with high accuracy through arbitrary control. For example, the coefficient of heat transfer between the stator core model and the external air model can be calculated by the least squares method to reduce the error of the estimated temperature relative to the measured temperature.

[0190] Next, the performance estimation unit 74 calculates variables related to the cooling performance of the motor 10 based on the first model 10a and the second model 10b generated by the model search unit 71. The performance estimation unit 74 obtains the coefficient hb related to heat transfer in the first model 10a and the coefficient hbx related to heat transfer in the second model 10b. The performance estimation unit 74 calculates the heat transfer coefficient between the stator core model 20a and the outside air model 36a based on the respective heat transfer-related coefficients hb and hbx.

[0191] The performance estimation unit 74 calculates a variable CA related to the cooling performance of the motor based on the heat transfer coefficient in the first model 10a and the heat transfer coefficient in the second model 10b. The variable CA in this embodiment is a ratio obtained by dividing the heat transfer coefficient between the stator core model 20a and the external air model 36a in the second model 10b by the heat transfer coefficient between the stator core model 20a and the external air model 36a in the first model 10a. That is, the variable CA related to the cooling performance is the ratio of the heat transfer coefficient in the second model 10b to the heat transfer coefficient in the first model 10a. It can be determined that the smaller the ratio is, the lower the cooling performance in the second model 10b. The performance estimation unit 74 determines that the cooling performance of the motor is reduced when the ratio of the heat transfer coefficients is less than a predetermined determination value.

[0192] Reference Figure 1 The display unit 52 of the present embodiment displays information related to the second model 10b produced by the model search unit 71. For example, the display unit 52 displays the information of the second model 10b corresponding to the measured temperature with the highest accuracy of the estimated temperature of the temperature detector 31 on the display unit 52. Alternatively, the display unit 52 displays the coefficient hbx related to heat transfer between the stator core model 20a and the external air model 36a, the heat transfer coefficient, or the variable CA related to the cooling performance. The operator can observe the information related to the second model 10b to estimate the cooling performance of the motor. For example, the operator can determine that the smaller the heat transfer coefficient of the second model 10b, the lower the cooling performance. Alternatively, the display unit 52 can also display information on the reduction in the cooling performance of the motor estimated by the performance estimation unit 74.

[0193] When it is estimated that the cooling performance of the motor is reduced, the operator can perform maintenance before the motor fails. For example, the operator can clean, repair or replace the motor. Alternatively, the operator can analyze countermeasures before the motor fails. For example, the load of the machine can be reduced before the motor is replaced, delaying the time of the motor failure.

[0194] In addition, the display unit 52 may also display information indicating that the motor is normal. For example, the display unit may display that the cooling performance of the motor is sufficient when the variable CA related to the cooling performance is greater than a predetermined determination value. The display unit may display the determination result each time the performance estimation unit 74 makes a determination.

[0195] The notification unit 75 of the monitoring device 2 can notify other devices of information related to the cooling performance of the motor. For example, the notification unit 75 sends a message to the mechanical control device 41 indicating that the cooling performance of the motor has decreased. The mechanical control device 41 can implement any control based on the notification from the notification unit 75. For example, the mechanical control device 41 can change the operating state of the machine 1. The mechanical control device 41 can reduce the speed of the motor drive, or change the rate of increase and rate of decrease of the speed. Alternatively, the mechanical control device 41 can limit the motor operation such as reducing the upper limit value of the rotation speed of the motor.

[0196] The variable used by the performance estimation unit to determine the cooling performance can be any variable based on the coefficient related to the heat transfer between the stator core and the outside air. For example, the variable obtained by subtracting the heat transfer coefficient between the stator core model and the outside air model in the first model from the heat transfer coefficient between the stator core model and the outside air model in the second model can be used.

[0197] The first model of the motor in the above-mentioned embodiment is composed of a coil model, a stator core model, a temperature detector model, an air layer model, a rotor model, and an external air model, but is not limited to this method. The first model of the motor may also include models of other components. For example, the first model of the motor may also include a housing model supporting the stator and the rotor, a bearing model, and a shaft model supporting the rotor. Alternatively, the first model of the motor may not include a part of the model. For example, the motor model may not include an air layer model.

[0198] The motor monitoring device of this embodiment is configured to be implemented by switching between a normal modeling mode and an abnormal modeling mode, but is not limited to this method. The motor monitoring device may not have the function of the normal modeling mode. The first model of the motor when the motor operation state is normal may also be generated by other devices.

[0199] In addition, in this embodiment, the monitoring device of the motor is composed of a computing device different from the mechanical control device, but the present invention is not limited to this embodiment. The mechanical control device may also have the function of the monitoring device of the motor. That is, the processor of the mechanical control device may also function as a unit of the monitoring device such as the model search unit, the temperature estimation unit, and the parameter calculation unit.

[0200] In each of the above-mentioned controls, the order of the steps can be appropriately changed within the scope of not changing the function and effect. The above-mentioned embodiments can be appropriately combined. In the above-mentioned figures, the same symbols are marked for the same or equivalent parts. In addition, the above-mentioned embodiments are examples and do not limit the invention. In addition, in the embodiments, changes to the embodiments shown in the claims are included.

[0201] Explanation of symbols

[0202] 2 Monitoring device

[0203] 10. Electric Motor

[0204] 10a, 10b motor model

[0205] 11 Rotor

[0206] 11a Rotor model

[0207] 16 Coil

[0208] 16a Coil Model

[0209] 20 stator core

[0210] 20a Stator core model

[0211] 29 Blower

[0212] 30a Electric Motor Model

[0213] 31a Temperature detector model

[0214] 35a Air layer model

[0215] 36a Outside Air Model

[0216] 51 Storage

[0217] 52 Display

[0218] 53. Temperature estimation unit

[0219] 54 Loss Calculation Department

[0220] 55 Temperature calculation unit

[0221] 62 Status acquisition unit

[0222] 63 Parameter calculation unit

[0223] 66 Evaluation Department

[0224] 67 Parameter Change Section

[0225] 71 Model Search Department

[0226] 72 Temperature determination unit

[0227] 73 Setting Department

[0228] 74 Performance Estimation Department

[0229] 75 Notification Department

[0230] hb, hbx heat transfer related coefficients

[0231] CA variable.

Claims

1. A monitoring device for an electric motor, characterized in that: The monitoring device comprises: a state acquisition unit that acquires an operating state of the motor including a measured temperature detected by a temperature detector attached to the motor; a temperature estimating unit that calculates an estimated temperature of the temperature detector based on a motor model; A storage unit storing a first model of the motor when the motor is in a normal operating state; a model search unit that generates a second model of the motor when the motor operation state is abnormal; and a performance estimating unit for estimating cooling performance of the electric motor, The motor model includes: a model of the motor components including the stator core, an external air model, A heat capacity is set for at least one component model, The coefficients related to heat transfer are set between the motor component models and between the stator core model and the external air model. When the difference between the measured temperature of the temperature detector and the estimated temperature of the temperature detector based on the first model is out of a predetermined determination range, the model search unit generates a second model in which a coefficient related to heat transfer between a stator core model and an external air model of the first model is changed so that the estimated temperature of the temperature detector based on the second model corresponds to the measured temperature. The performance estimating unit calculates a variable related to cooling performance of the motor based on a coefficient related to heat transfer between a stator core model and an outside air model in the first model and a coefficient related to heat transfer between the stator core model and the outside air model in the second model.

2. The motor monitoring device according to claim 1, characterized in that: The variable related to the cooling performance of the motor is: the ratio of the heat transfer coefficient between the stator core model and the external air model in the second model divided by the heat transfer coefficient between the stator core model and the external air model in the first model, The performance estimating unit determines that the cooling performance of the electric motor is deteriorated when the ratio is smaller than a predetermined determination value.

3. The motor monitoring device according to claim 1 or 2, characterized in that: The monitoring device for the electric motor includes a notification unit that notifies other devices of information related to the cooling performance of the electric motor.

4. The motor monitoring device according to any one of claims 1 to 3, characterized in that: The electric motor monitoring device includes a display unit that displays information related to the cooling performance of the electric motor.

5. The motor monitoring device according to any one of claims 1 to 4, characterized in that: The monitoring device includes: a parameter calculation unit that calculates parameters of a first model of the motor; The state acquisition unit acquires the motor operation command generated by actually driving the motor. The parameters include: heat capacity set for the component models, coefficients related to heat transfer between the component models, The temperature estimation unit includes: a loss calculation unit that calculates the heat generated by the primary copper loss of the coil and the heat generated by the iron loss of the stator core according to the operation instruction; a temperature calculation unit that calculates the estimated temperature of the temperature detector using a first model of the motor based on the heat generated by the coil and the heat generated by the stator core. The parameter calculation unit includes: an evaluation unit that evaluates the estimated temperature of the temperature detector by comparing the estimated temperature of the temperature detector with the measured temperature of the temperature detector; and a parameter changing unit that changes the value of the parameter according to the evaluation result of the evaluation unit so that the estimated temperature of the temperature detector approaches the measured temperature. The mode is configured to be switchable between a mode in which the parameter calculation unit calculates the parameters of the first model of the electric motor and a mode in which the model search unit generates the second model of the electric motor.

Images