Temperature prediction device, rotating electrical machine system and electric drive system using the same

Abstract

To provide a temperature prediction device and a rotary electric machine system and an electric drive system using the same to accurately predict temperature of an electric drive system without requiring much effort to construct an estimation model.SOLUTION: A temperature prediction device 300 for an electric drive system includes a rotary electric machine 101, learns by applying machine learning, and predicts temperature of each part of the electric drive system by using the learning model that predicts the temperature of each part of the electric drive system with state quantities during the operation of the electric drive system as input data, so as to output the temperature.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present invention relates to a temperature prediction device, a rotating electrical machine system using the same, and an electric drive system.

Background Art

[0002] If the operation is continued in a state where the temperatures of the stator and rotor constituting the rotating electrical machine exceed the limit temperature, demagnetization of the rotor core permanent magnet may occur and the output torque may decrease. Therefore, it is necessary to predict the temperature of the rotating electrical machine and appropriately control it.

[0003] For example, a temperature estimation device is disclosed that includes a temperature sensor provided inside a housing that houses a rotating electrical machine, a calorific value calculation unit that calculates the calorific value of components, an eddy current loss estimation unit that estimates the eddy current loss of a coil based on the rotational speed, torque, and applied voltage of the rotating electrical machine, and a component temperature calculation unit that calculates the temperature of components based on the calorific value of components and the eddy current loss of the coil (Patent Document 1).

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0005] In the conventional technology described in Patent Document 1, the designer constructs a thermal circuit model inside the electric drive system and estimates the flow of temperature in the device. However, it is considered necessary to determine the wiring method of the thermal circuit and the values of each parameter by trial and error, which may require labor.

Means for Solving the Problems

[0006] One aspect of the present invention is a temperature prediction device for an electric drive system including a rotating electrical machine, which is learned by applying machine learning, and predicts the temperature of each part of the electric drive system using a learning model that predicts the temperature of each part of the electric drive system with the state quantity during the operation of the electric drive system as input data and outputs the prediction.

[0007] Here, it is preferable that the learning model is machine-learned to output the temperature of each part of the electric drive system when the state quantity is input, using teacher-student learning data in which the state quantity during the operation of the electric drive system is combined with the temperature of each part of the electric drive system during the operation of the electric drive system as teacher data.

[0008] Further, it is preferable that the state quantity includes at least one of the rotational speed of the rotating electrical machine, the output torque of the rotating electrical machine, the input voltage of the rotating electrical machine, the input current of the rotating electrical machine, the input current density of the rotating electrical machine, the temperature of the coil end of the stator coil of the rotating electrical machine, the flow rate of the cooling fluid, the temperature of the cooling fluid, the ambient temperature, and the vehicle speed.

[0009] Further, it is preferable that the teacher-student learning data includes image data indicating the temperature at each part of the electric drive system.

[0010] Further, it is preferable that the image data indicates the temperature distribution at each part of the electric drive system by any one of brightness, color density, and color separation.

[0011] Further, it is preferable that the input data includes image data indicating the flow rate of the cooling fluid at each part of the electric drive system.

[0012] Further, it is preferable that the image data indicates the distribution of the flow rate of the cooling fluid at each part of the electric drive system by any one of brightness, color density, and color separation.

[0013] Further, it is preferable that the machine learning is performed using a convolutional neural network.

[0014] Another aspect of the present invention is a rotating electrical machine system characterized by predicting the temperature of each part of the electric drive system during operation using the temperature prediction device and controlling the operating conditions of the rotating electrical machine according to the predicted temperature.

[0015] Another aspect of the present invention is a rotating electrical machine system characterized by predicting the temperature of each part of the electric drive system during operation using the temperature prediction device and controlling the flow rate of the cooling fluid of the electric drive system according to the predicted temperature.

[0016] Another aspect of the present invention is an electric drive system characterized by utilizing the heat obtained by heat exchange with the cooling fluid of the electric drive system in the rotating electrical machine system for heating.

Advantages of the Invention

[0017] According to the present invention, it is possible to provide a temperature prediction device that can suppress trial and error in the wiring method of the thermal circuit and each parameter, a rotating electrical machine system using the same, and an electric drive system.

Brief Description of the Drawings

[0018]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Figure 20

Figure 21

Figure 22

Embodiments for Carrying Out the Invention

[0019] As shown in FIG. 1, a rotating electrical machine system 100 in an embodiment of the present invention includes a rotor 10, a shaft 12, a stator 14, bearings 16, a casing 18, an inverter 20, a battery 22, a control unit 24, an oil pump 26, a motor 28, and an oil pan 29. The rotor 10, the shaft 12, the stator 14, the bearings 16, and the casing 18 constitute a rotating electrical machine 101.

[0020] FIG. 2 shows an example of a vehicle 200 equipped with the rotating electrical machine 101. The vehicle 200 travels by transmitting the power output from the rotating electrical machine 101, which is a power source, to drive wheels (tires) 108 via a transmission 102, a differential gear 104, and a drive shaft 106.

[0021] Note that a configuration including the rotating electrical machine 101, the transmission 102, and the differential gear 104 is referred to as an electric drive system. However, the electric drive system only needs to include the rotating electrical machine 101 and may include other configurations. Hereinafter, temperature prediction and control based on the temperature prediction result will be mainly described for the rotating electrical machine 101, but the same can be applied to other configurations included in the electric drive system.

[0022] The rotor 10 is a part that rotates in the rotating electrical machine 101. The rotor 10 is composed of a rotor core permanent magnet 10a and a rotor core electromagnetic steel plate 10b. The shaft 12 is connected to rotate together with the rotor 10. The shaft 12 is used to transmit the rotational torque output from the rotor 10 to the outside of the rotating electrical machine 101. The stator 14 is a part that is relatively stationary with respect to the rotor 10 in the rotating electrical machine 101. The stator 14 is composed of a stator coil 14a and a stator core electromagnetic steel plate 14b. The rotor 10, the shaft 12, and the stator 14 are housed in the casing 18. A bearing 16 is arranged between the shaft 12 and the casing 18, and the rotor 10 can rotate smoothly together with the shaft 12.

[0023] The power supplied from the battery 22 is adjusted in voltage and frequency by the inverter 20, and the current flowing through the stator coil 14a of the rotating electrical machine 101 is controlled. When a current appropriately controlled by the inverter 20 flows through the stator coil 14a, a rotating magnetic field is created in the stator 14. Due to the magnetic interaction between the rotating magnetic field and the rotor core permanent magnet 10a, a rotational torque is generated in the rotor 10, causing the rotor 10 to perform a rotational motion.

[0024] The control unit 24, also called an in-vehicle computer, is composed of an arithmetic unit that processes information, a storage device that stores information, etc. The control unit 24 calculates the required torque for the rotating electrical machine 101 based on information such as the accelerator pedal operation amount and vehicle speed operated by the driver. Then, the control unit 24 issues a command to the inverter 20 to control the current flowing through the stator coil 14a according to the required torque.

[0025] When current flows through the stator coil 14a, part of the power is lost and the stator coil 14a generates heat. Also, when the rotor 10 rotates, eddy currents flow in the rotor core electromagnetic steel plate 10b and generate heat. To cool the rotating electrical machine 101 heated by this heat generation, cooling oil is supplied from the oil pump 26. The oil pump 26 is driven by the motor 28. The cooling oil that has taken heat from the rotor 10 and the stator 14 is returned to the oil pan 29 and cooled by releasing heat to the outside air. A radiator (not shown) may be used to increase the efficiency of heat exchange with the outside air. Also, the heat of the cooling oil may be used to heat other components, for example, the battery 22 or the like. Note that the cooling oil also functions as a lubricating oil and smoothes the movement of sliding parts such as the bearing 16.

[0026] Here, in the operation of the electric drive system, it is extremely important to control the temperature of the rotating electrical machine 101 within an appropriate range. If the temperature of the rotor core permanent magnet 10a disposed in the rotor 10 exceeds the allowable value, there is a risk of irreversible demagnetization occurring where the magnetic force does not return even after the rotor core permanent magnet 10a is cooled and the temperature returns. When irreversible demagnetization occurs, the upper limit value of the torque that the rotating electrical machine 101 can output decreases, and the performance of the rotating electrical machine 101 deteriorates. Also, in the stator coil 14a, if the temperature rises to exceed the allowable value, the insulation coating may be damaged and a short circuit may occur, which may cause damage to the entire electric drive system.

[0027] [Temperature Prediction Device] Hereinafter, a temperature prediction device 300 for predicting the temperature of an electric drive system including the rotating electrical machine 101 will be described. As shown in FIG. 3, the temperature prediction device 300 can be configured to include a processing unit 30, a storage unit 32, an input unit 34, an output unit 36, and a communication unit 38.

[0028] The processing unit 30 includes means for performing arithmetic processing such as a CPU. The processing unit 30 performs machine learning for predicting the temperature of the electric drive system using the temperature prediction model stored in the storage unit 32, and predicts the temperature of the electric drive system using the machine-learned temperature prediction model. The storage unit 32 includes storage means such as a semiconductor memory and a memory card. The storage unit 32 is accessibly connected to the processing unit 30 and stores the temperature prediction model and the parameters learned in the model, as well as other information necessary for the processing in the temperature prediction device 300. The input unit 34 includes means for inputting information. The input unit 34 includes, for example, a keyboard, a touch panel, buttons, etc. for receiving input from an administrator. The output unit 36 includes means for outputting a processing result. The output unit 36 includes, for example, a display for outputting a processing result such as a user interface screen (UI). The communication unit 38 is configured to include an interface for communicating with an external device via an information communication network such as the Internet or a LAN. The communication by the communication unit 38 may be either wired or wireless.

[0029] FIG. 4 is a conceptual diagram showing the temperature prediction model applied to the temperature prediction device 300. For the temperature prediction model for predicting the temperature distribution inside the conduction drive system, it is preferable to apply a deep learning model including a convolutional neural network (CNN).

[0030] CNNs are used in a wide range of fields such as the recognition of still images and videos, object detection, etc. By utilizing not only the values of each point in the input data such as images but also the position information of the point being focused on, CNNs extract features more efficiently than conventional deep learning models. As a result, CNNs have the feature of having a high accuracy rate when applied to image recognition, etc. Deep learning models including CNNs are disclosed, for example, in Yann LeCun, Patrick Haffner, Leon Bottou and Yoshua Bengio, “Object Recognition with Gradient - Based Learning“, Shape, Contour and Grouping in Computer Vision, pp. 319 - 345, 1999.

[0031] Figure 5 shows an overview of a CNN model for predicting the temperature distribution inside an electric drive system. In Figure 5, the CNN model is composed of multiple layers from the input layer on the left end to the output layer on the right end. When input image data, which is a combination of image data with the number of pixels in the vertical x and horizontal y directions and the number of channels z1, is input into the CNN model, the input image data is compressed (reduced) in the vertical and horizontal directions by the convolution process, and the information in the input image data is gradually aggregated. In the example of the CNN model in Figure 5, the input image data is compressed to 1 / 8 in both the vertical and horizontal directions. At this time, the number of channels is changed as z2→z3→z4, which is equal to the number of kernels performing the convolution process. Then, while the number of channels is changed as z5→z6→z7 by the transposed convolution process, the information in the vertical and horizontal directions is restored, and finally, output image data with the same number of pixels in the vertical and horizontal directions as the input image data is output. The information of each pixel in this output image data is the predicted value of the temperature of each part inside the electric drive system. Note that if the number of channels is 1 channel, it is a monochrome image, and if it is 3 channels, it is a color image. The high or low temperature is represented by the shade or brightness of the pixels in each channel.

[0032] When predicting the temperature distribution inside an electric drive system, a deep learning technique using a learning model including a CNN model is utilized. In deep learning, in the learning stage of the learning model, teacher - supervised learning data, which combines the measured values of the temperatures of each part of the electric drive system in the state as teacher data (correct data or true value) with the input data indicating the state of the electric drive system, is input into the learning model, and learning is performed so that the output (prediction) value from the learning model approaches the teacher data. By advancing the learning while changing the teacher - supervised learning data, the parameters in the learning model are automatically corrected so that the difference between the predicted value of the temperature output from the learning model and the teacher data, which is the true value, becomes smaller, and the correct rate of the predicted value of the temperature is improved.

[0033] In this way, by training the learning model through machine learning, it becomes possible to predict the temperature of the electric drive system without constructing a circuit model or the like or determining the values of the model parameters through trial and error.

[0034] In the present embodiment, a configuration diagram showing the internal structure of the electric drive system is used as the input image data input to the CNN model. FIG. 6 shows an example of the internal configuration of the rotating electrical machine 101 included in the electric drive system. FIG. 6(a) shows a cross - sectional view when the rotating electrical machine 101 is cut along the line B - B in FIG. 6(b) by a plane perpendicular to the rotation axis. FIG. 6(b) shows a cross - sectional view when the rotating electrical machine 101 is cut along the line A - A in FIG. 6(a) in a direction parallel to the rotation axis. Note that, as the configuration diagram of the electric drive system, for example, CAD data created in the design of the electric drive system can be used.

[0035] For the CNN model, a configuration diagram in which each part of the electric drive system is color - coded is used as the input image data. By color - coding each part, the CNN model can automatically recognize through machine learning that the temperature characteristics such as the thermal resistance and heat capacity of each part are different, and can derive a predicted value of the temperature corresponding to the characteristics of each part.

[0036] In addition, in the present embodiment, a "part" refers to a component or a part thereof. For example, when the electric drive system is the rotating electric machine 101, it may be in units of components such as the rotor 10, the shaft 12, the stator 14, etc. Also, it may be in smaller units such as the rotor core permanent magnet 10a, the rotor core electromagnetic steel plate 10b, etc. Further, for example, it may be in even smaller units such as the center of the rotor core permanent magnet 10a, the center of the stator 14 (stator center), the coil end of the stator coil 14a, etc. How to divide each part may be appropriately designed according to the purpose. Furthermore, in the present embodiment, a "part" may also be a part formed by combining a plurality of components or parts thereof. For example, the rotor core permanent magnet 10a and the rotor core electromagnetic steel plate 10b may be combined and treated as one part, or a part of the rotor core permanent magnet 10a and a part of the rotor core electromagnetic steel plate 10b may be combined and treated as one part.

[0037] Examples of the input image data input to the CNN model include image data showing the supply state of the cooling oil for cooling the electric drive system. Since the supply condition of the cooling oil to each part of the electric drive system is important information when predicting the temperature of each part of the electric drive system, image data showing the supply path and flow rate of the cooling oil is given to the CNN model as the input image data.

[0038] FIG. 7 shows an example of the input image data when cooling oil is supplied from the axis of the shaft 12 to the side surface of the rotor 10. FIG. 8 shows an example of the input image data when cooling oil is supplied to the center and upper part of the stator 14. In FIGS. 7 and 8, the parts where the cooling oil is supplied are shown by hatching. Instead of or in addition to this, the parts where the cooling oil is supplied and the parts where it is not supplied may be distinguished by the brightness or color of the parts. Also, the flow rate of the cooling oil may be expressed by the brightness, the darkness of the color, etc. For example, it may be shown that the higher the brightness, the lower the flow rate of the cooling oil to the part, and the lower the brightness, the higher the flow rate of the cooling oil to the part. Also, for example, it may be shown that the lighter the color, the lower the flow rate of the cooling oil to the part, and the darker the color, the higher the flow rate of the cooling oil to the part. Note that the relationship between the brightness or the darkness of the color and the flow rate of the cooling oil may be reversed.

[0039] However, the flow rate of the cooling oil is not limited to being represented by the lightness or color density of the part, and any method can be used as long as it can represent the flow rate of the cooling oil in the configuration diagram of the electric drive system. For example, the flow rate of the cooling oil may be represented by color coding in the configuration diagram of the electric drive system.

[0040] In addition, examples of the state quantity indicating the operating state of the electric drive system include the rotational speed of the rotor 10, the output torque of the electric machine system 100, the input voltage of the electric machine system 100, the input current of the electric machine system 100, the input current density of the electric machine system 100, the temperature of the coil end of the stator coil 14a, the oil temperature of the cooling oil, the ambient temperature (outside air temperature), the vehicle speed, and the like. These state quantities can be collected by providing various sensors for measuring these state quantities in the electric drive system. Furthermore, even if they cannot be directly measured, the copper loss and iron loss inside the electric machine system 100 can also be obtained from the information of the above state quantities.

[0041] The information of these state quantities is important information when predicting the temperature of each part of the electric drive system, so it is preferably given to the learning model. However, it is not necessary to input all of the rotational speed of the rotor 10, the output torque of the electric machine system 100, the input voltage of the electric machine system 100, the input current of the electric machine system 100, the input current density of the electric machine system 100, the temperature of the coil end of the stator coil 14a, the oil temperature of the cooling oil, the ambient temperature (outside air temperature), and the vehicle speed into the learning model, and they may be selectively used according to the accuracy of the predicted temperature value output from the learned learning model.

[0042] The information of these state quantities is preferably input into the CNN model as image data indicating the state quantities of each part in the configuration diagram of the electric drive system. Fig. 9 shows an example of the input image data indicating these state quantities. As shown in Fig. 9, the value of each state quantity may be indicated by the overall brightness of the electric drive system. That is, as shown in Fig. 9(a), the higher the overall brightness of the configuration diagram of the electric drive system, the smaller the value of the state quantity, and as shown in Figs. 9(b) and 9(c), the lower the overall brightness of the configuration diagram of the electric drive system, the larger the value of the state quantity. Also, instead of brightness, the value of the state quantity may be indicated by the darkness of the color. That is, it may be shown that the lighter the color, the smaller the value of the state quantity, and the darker the color, the larger the value of the state quantity. Note that the relationship between brightness or color darkness and the state quantity may be reversed.

[0043] However, it is not limited to being represented by brightness or color darkness, and any method that can represent the value of each state quantity in the configuration diagram of the electric drive system is acceptable. For example, the value of each state quantity may be represented by color coding in the configuration diagram of the electric drive system.

[0044] In this embodiment, the state quantity during the operation of the electric drive system is input into the CNN model as input image data. However, when using a learning model that combines the CNN model with other neural networks, the state quantity indicating the operating state of the electric drive system may be numerically input into a neural network other than the CNN model.

[0045] For the input data indicating the state quantities indicating the supply state of the cooling oil to the electric drive system and the operating state of the electric drive system, the measured values of the temperature of each part of the electric drive system are combined as teacher data to form supervised learning data. Then, the supervised learning data is input into the learning model, and learning is performed so that the predicted value of the temperature output from the learning model approaches the teacher data.

[0046] The measured values of the temperatures of the respective parts of the electric drive system are preferably image data indicating the temperatures of the respective parts in the configuration diagram of the electric drive system. The temperature may be expressed by, for example, brightness or color density. For example, it may be shown that the higher the brightness, the higher the temperature of the part, and the lower the brightness, the lower the temperature of the part. Also, for example, it may be shown that the lighter the color, the lower the temperature of the part, and the darker the color, the higher the temperature of the part. Note that the relationship between brightness or color density and temperature may be reversed.

[0047] However, it is not limited to being represented by brightness or color density, and any method that can represent temperature in the configuration diagram of the electric drive system is acceptable. For example, temperature may be represented by color coding in the configuration diagram of the electric drive system.

[0048] When predicting the temperatures of the respective parts inside the electric drive system using a learning model to which a CNN model is applied, as shown in FIG. 10, it can be predicted as a regression problem of predicting the temperatures of the respective parts inside the electric drive system as continuous numerical data. In FIG. 10, an example of predicting as continuous numerical data for the rotor core permanent magnet 10a, the coil end of the stator coil 14a, and the center (central portion in the axial length direction) of the stator core electromagnetic steel sheet 14b of the stator 14 is shown. Also, as shown in FIG. 11, it can also be predicted as a classification problem of dividing the temperature into several classes from low temperature to high temperature and predicting to which class the temperature of the part of interest belongs. In the case of the example in FIG. 11, the temperature is classified into three classes: 110°C or more and less than 115°C, 115°C or more and less than 120°C, and 120°C or more and less than 125°C, and an example of predicting to which class the temperatures of the rotor core permanent magnet 10a, the coil end of the stator coil 14a, and the center (central portion in the axial length direction) of the stator core electromagnetic steel sheet 14b of the stator 14 belong is shown.

[0049] FIG. 12 shows an example in which, using the learned learning model, the results of predicting the temperatures of the respective parts of the electric drive system are output as image data of the temperature distribution in the configuration diagram of the electric drive system. In FIG. 12, the temperature range of each part is shown by the level of brightness of the region corresponding to each part of the electric drive system.

[0050] As described above, according to the temperature prediction device 300 in the present embodiment, the temperature of the electric drive system can be accurately predicted without requiring great labor for constructing the temperature estimation model.

[0051] [Control of Rotating Electrical Machine Based on Temperature Prediction] FIG. 13 shows a flowchart for controlling the rotating electrical machine 101 based on the result of predicting the temperature distribution inside the electric drive system using the temperature prediction device 300.

[0052] In step S10, data necessary for predicting the temperature distribution inside the electric drive system is acquired using the temperature prediction device 300. Data necessary for predicting the temperature distribution inside the electric drive system, such as the supply path and flow rate of the cooling oil, the rotational speed of the rotor 10, the output torque of the rotating electrical machine system 100, the input voltage of the rotating electrical machine system 100, the input current of the rotating electrical machine system 100, the input current density of the rotating electrical machine system 100, the temperature of the coil end of the stator coil 14a, the oil temperature of the cooling oil, the ambient temperature (outside air temperature), and the vehicle speed, is acquired by various sensors.

[0053] In step S12, the temperature distribution of each part inside the electric drive system is predicted using the temperature prediction device 300. That is, by inputting the state quantity of the electric drive system acquired in step S10 as input data into the learned learning model of the temperature prediction device 300, the prediction result of the temperature distribution of each part inside the electric drive system is output from the temperature prediction device 300.

[0054] In step S14, the difference value between the highest temperature t pred in the predicted temperature distribution and the limit allowable temperature value t tol in the electric drive system is obtained, and it is determined whether or not the difference value is less than a preset temperature threshold ε1. The part having the highest temperature t pred may be a part having a temperature within a predetermined reference temperature range from the highest temperature t pred . The reference temperature range may be set based on the allowable temperature of each part of the electric drive system, the cooling capacity of the cooling oil, and the like. The limit allowable value ttol - The maximum temperature t pred If it is less than the temperature threshold ε1, it is determined that the thermal limit is approaching and there is no thermal margin in the electric drive system, and the process is transferred to step S16. The limit tolerance value t tol - The maximum temperature t pred If it is greater than or equal to the temperature threshold ε1, it is determined that there is a thermal margin in the electric drive system, and the control of the rotary electric machine 101 based on temperature is terminated. The limit tolerance value t tol The limit tolerance value t and the temperature threshold ε1 may be appropriately set according to the characteristics of the electric drive system to be controlled.

[0055] In step S16, the maximum torque T that the rotary electric machine 101 can output max is temporarily limited to suppress the temperature rise inside the electric drive system.

[0056] FIG. 14 shows a flowchart of another control method for controlling the rotary electric machine 101 based on the result of predicting the temperature distribution inside the electric drive system using the temperature prediction device 300.

[0057] In step S10, the data necessary for predicting the temperature distribution inside the electric drive system is acquired. Also, in step S12, the temperature distribution of each part inside the electric drive system is predicted using the temperature prediction device 300. The processing in these steps is the same as the processing already described, so the description is omitted.

[0058] In step S18, the area S pred of the part having the highest temperature t pred in the predicted temperature distribution is obtained. FIG. 15 shows an example of the part having the highest temperature t pred in the predicted temperature distribution. The part having the highest temperature t pred and the part having a temperature within a predetermined reference temperature range from the highest temperature t pred may be used. The reference temperature range may be set based on the allowable temperature of each part of the electric drive system, the cooling capacity of the cooling oil, etc. In FIG. 15, the hatched part is the area S pred of the part having the highest temperature t predis shown. The processing unit 30 determines the area S pred of the part having the highest temperature t in the configuration diagram showing the predicted temperature distribution pred .

[0059] Subsequently, it is determined whether or not the area S pred is larger than the area threshold ε2. If the area S pred is larger than the area threshold ε2, it is determined that the thermal limit is approaching and there is no thermal margin in the electric drive system, and the process proceeds to step S16. If the area S pred is less than or equal to the area threshold ε2, it is determined that there is a thermal margin in the electric drive system, and the control of the rotating electrical machine 101 based on the temperature is terminated. The limit tolerance value t tol and the area threshold ε2 may be appropriately set according to the characteristics of the electric drive system to be controlled.

[0060] In step S16, the maximum torque T max that the rotating electrical machine 101 can output is temporarily limited to suppress the temperature rise inside the electric drive system.

[0061] Note that in this embodiment, the process is performed using the area S pred of the part having the highest temperature t in the configuration diagram of the electric drive system, but the process may be performed using the volume of the part having the highest temperature t pred in the electric drive system. In this case, the volume of the part having the highest temperature t pred in the configuration diagram of the electric drive system may be calculated by integration. pred

[0062] As described above, by controlling the rotating electrical machine 101 based on the result of predicting the temperature distribution inside the electric drive system using the temperature prediction device 300, the temperature rise of the electric drive system can be suppressed.

[0063] [Supply Control of Cooling Oil Based on Temperature Prediction] FIG. 16 shows a flowchart for controlling the supply of the cooling oil supplied to the electric drive system based on the result of predicting the temperature distribution inside the electric drive system using the temperature prediction device 300.

[0064] ​ In step S10, data necessary for predicting the temperature distribution inside the electric drive system is acquired. Also, in step S12, the temperature distribution of each part inside the electric drive system is predicted using the temperature prediction device 300. The processing in these steps is the same as the processing already described, so the description is omitted.

[0065] In step S20, the difference value between the highest temperature t pred in the predicted temperature distribution and the limit allowable temperature t tol in the electric drive system is obtained, and it is determined whether the difference value is less than a preset temperature threshold ε3. Limit allowable temperature t tol - highest temperature t pred is less than the temperature threshold ε3, it is determined that the thermal limit is approaching and there is no thermal margin in the electric drive system, and the process proceeds to step S22. Limit allowable temperature t tol - highest temperature t pred is greater than or equal to the temperature threshold ε3, it is determined that there is a thermal margin in the electric drive system, and the process proceeds to step S24.

[0066] In step S22, a process of increasing the flow rate Q cool of the cooling oil supplied to the electric drive system is performed. Specifically, the rotation speed of the motor 28 shown in FIG. 2 is controlled, etc., to increase the flow rate Q cool of the cooling oil supplied from the oil pan 29.

[0067] On the other hand, when limit allowable temperature t tol - highest temperature t pred is greater than or equal to the temperature threshold ε3, in step S24, it is determined whether limit allowable temperature t tol - highest temperature t pred is greater than a preset temperature threshold ε4. The temperature threshold ε4 is set to a value greater than the temperature threshold ε3, and it is preferably set to a value that can determine that there is still a thermal margin even if the electric drive system reduces the flow rate of the cooling oil. Limit allowable temperature t tol - highest temperature t predIf it is greater than the temperature threshold ε4, it is determined that there is sufficient thermal margin even if the flow rate of the cooling oil is decreased in the electric drive system, and the process proceeds to step S26. Limit tolerance value t tol - Maximum temperature t pred If it is less than or equal to the temperature threshold ε4, it is determined that there is a thermal margin in the electric drive system but there is no margin to decrease the flow rate of the cooling oil, and the supply control of the cooling oil based on the temperature is terminated.

[0068] In step S26, a process of decreasing the flow rate Q of the cooling oil supplied to the electric drive system is performed. cool Specifically, by controlling the rotational speed etc. of the motor 28 shown in Fig. 2, the flow rate Q of the cooling oil supplied from the oil pan 29 is decreased. cool is decreased.

[0069] Fig. 17 shows a flowchart of another control method for controlling the supply of the cooling oil supplied to the electric drive system based on the result of predicting the temperature distribution inside the electric drive system using the temperature prediction device 300.

[0070] In step S10, data necessary for predicting the temperature distribution inside the electric drive system is acquired. Also, in step S12, the temperature distribution of each part inside the electric drive system is predicted using the temperature prediction device 300. Since the processing in these steps is the same as the processing already described, the description is omitted.

[0071] In step S28, the area S of the part having the highest temperature t among the predicted temperature distributions is obtained. Subsequently, it is determined whether or not the area S is greater than the area threshold ε5. When the area S is greater than the area threshold ε5, it is determined that the thermal limit is near and there is no thermal margin in the electric drive system, and the process proceeds to step S22. When the area S is less than or equal to the area threshold ε5, it is determined that there is a thermal margin in the electric drive system, and the process proceeds to step S30. Limit tolerance value t pred and the area threshold ε5 may be appropriately set according to the characteristics of the electric drive system to be controlled. pred is obtained. Subsequently, it is determined whether or not the area S pred is greater than the area threshold ε5. When the area S pred is greater than the area threshold ε5, it is determined that the thermal limit is near and there is no thermal margin in the electric drive system, and the process proceeds to step S22. When the area S pred is less than or equal to the area threshold ε5, it is determined that there is a thermal margin in the electric drive system, and the process proceeds to step S30. Limit tolerance value t tol and the area threshold ε5 may be appropriately set according to the characteristics of the electric drive system to be controlled.

[0072] In step S22, a process of increasing the flow rate Q of the cooling oil supplied to the electric drive system is performed. Specifically, the rotational speed of the motor 28 shown in FIG. 2 and the like are controlled to increase the flow rate Q of the cooling oil supplied from the oil pan 29. cool to increase it. cool

[0073] On the other hand, when the area S pred is less than or equal to the area threshold value ε5, in step S30, it is determined whether the area S pred is less than a preset area threshold value ε6. The area threshold value ε6 is set to a value smaller than the area threshold value ε5, and it is preferably set to a value that can determine that there is a thermal margin even if the electric drive system reduces the flow rate of the cooling oil. When the area S pred is less than the area threshold value ε6, it is determined that there is sufficient thermal margin even if the flow rate of the cooling oil is reduced in the electric drive system, and the process proceeds to step S26. When the area S pred is greater than the area threshold value ε6, it is determined that there is a thermal margin in the electric drive system but there is not enough margin to reduce the flow rate of the cooling oil, and the supply control of the cooling oil based on the temperature is terminated.

[0074] In step S26, a process of decreasing the flow rate Q of the cooling oil supplied to the electric drive system is performed. Specifically, the rotational speed of the motor 28 shown in FIG. 2 and the like are controlled to decrease the flow rate Q of the cooling oil supplied from the oil pan 29. cool to decrease it. cool

[0075] As described above, by controlling the supply of the cooling oil to the electric drive system based on the result of predicting the temperature distribution inside the electric drive system using the temperature prediction device 300, it is possible to suppress the temperature rise of the electric drive system. When there is sufficient thermal margin in the electric drive system, by reducing the flow rate of the cooling oil, the loss due to the dragging of the cooling oil by the rotating parts of the electric drive system can be reduced, and the electricity cost can be improved.

[0076] Note that in the temperature prediction device 300 according to the present embodiment, since the temperature of each part inside the electric drive system can be predicted by using the CNN model, the part where the cooling oil is supplied and its flow rate may be controlled so that the cooling effect of the part with a high temperature becomes higher.

[0077] FIG. 18 shows a configuration in which a direction switching valve 40 is provided in the supply path of the cooling oil in the rotating electrical machine system 100 to control the part where the cooling oil is supplied so that the cooling effect of the part with a high temperature becomes higher. As the path of the cooling oil, a first supply path 42a for supplying the cooling oil to the coil end of the stator coil 14a, a second supply path 42b for supplying the cooling oil to the axial center of the stator 14, and a third supply path 42c for supplying the cooling oil to the axial centers of the shaft 12 and the rotor 10 are provided. However, the supply path of the cooling oil is not limited to the above example, and other supply paths may be provided. The control unit 24 controls the direction switching valve 40 according to the temperature of each part of the rotating electrical machine 101 predicted by the temperature prediction device 300, and selects at least one of the coil end of the stator coil 14a, the axial center of the stator 14, and the axial centers of the shaft 12 and the rotor 10 to control the individual supply of the cooling oil.

[0078] FIG. 19 shows a configuration in which flow control valves 44 (44a to 44c) are provided in each of the supply paths of the cooling oil in the rotating electrical machine system 100 to control the part where the cooling oil is supplied and its flow rate so that the cooling effect of the part with a high temperature becomes higher. The flow control valve 44a is provided in the first supply path 42a for supplying the cooling oil to the coil end of the stator coil 14a. The flow control valve 44b is provided in the second supply path 42b for supplying the cooling oil to the axial center of the stator 14. The flow control valve 44c is provided in the third supply path 42c for supplying the cooling oil to the axial centers of the shaft 12 and the rotor 10. The control unit 24 controls the flow control valves 44 (44a to 44c) according to the temperature of each part of the rotating electrical machine 101 predicted by the temperature prediction device 300, and individually controls the flow rate of the cooling oil to the coil end of the stator coil 14a, the axial center of the stator 14, and the axial centers of the shaft 12 and the rotor 10.

[0079] Figure 20 shows a flowchart for obtaining the flow rate of the cooling oil required for cooling. The flow rate of the cooling oil required for cooling can be obtained from the difference between the ideal temperature distribution inside the electric drive system and the temperature distribution inside the electric drive system output from the temperature prediction device 300. Figure 21 shows an example of the ideal temperature distribution inside the electric drive system. Note that an example of the temperature distribution inside the electric drive system output from the temperature prediction device 300 is shown in FIG. 12.

[0080] In step S10, the data necessary for predicting the temperature distribution inside the electric drive system is acquired. Also, the ideal temperature distribution inside the electric drive system as shown in FIG. 21 is acquired. In step S12, the temperature distribution of each part inside the electric drive system is predicted using the temperature prediction device 300.

[0081] In step S32, the amount of heat U that must be removed at each part inside the electric drive system is calculated. The processing unit 30 of the temperature prediction device 30 calculates the amount of heat U that must be removed for each part by multiplying the difference (temperature difference) between the ideal temperature distribution inside the electric drive system and the predicted temperature distribution inside the electric drive system by the volume of each part.

[0082] In step S34, the flow rate Q of the cooling oil required for the amount of heat U that must be removed at each part inside the electric drive system cool is calculated. The processing unit 30 divides the amount of heat U to be removed from each part inside the electric drive system calculated in step S32 by the amount of heat ΔU taken away by the cooling oil per unit flow rate (for example, 1 L / min) to calculate the flow rate Q of the cooling oil required to cool each part. cool When controlling the supply path of the cooling oil and its flow rate, control may be performed so that the required flow rate Q of the cooling oil cool is supplied to each part.

[0083] Note that control of the rotating electrical machine based on the prediction result of the temperature inside the electric drive system and control of the supply of the cooling oil may be combined.

[0084] FIG. 22 shows the configuration of an electric drive system 110 that utilizes the heat of the cooling oil for warming up the battery. The electric drive system 110 utilizes the amount of heat taken away from an electric drive system including a rotating electrical machine system 100 for warming up a battery mounted on a vehicle or the like.

[0085] The electric drive system 110 is configured to include a rotating electrical machine system 100 and a battery system 112. The rotating electrical machine system 100 includes a direction switching valve 46 and cooling oil passages 48 (48a, 48b) in addition to the configuration shown in FIG. 18. Further, the battery system 112 includes a battery 50, a heat exchanger 52, a cooling water passage 54, a cooling water pump 56, a pump 58, and a cooling water pan 60.

[0086] The control unit 24, when the area S pred is less than the area threshold value ε6 and the amount of heat taken away from the rotating electrical machine system 100 by cooling is small, that is, when performing a process of decreasing the flow rate Q cool of the cooling oil supplied to the rotating electrical machine system 100 in step S26 above, switches the direction switching valve 46 to the first cooling oil passage 48a that does not pass through the heat exchanger 52. In this state, the heat of the cooling oil of the rotating electrical machine system 100 is not used for warming up the battery 50.

[0087] On the other hand, when the area S pred is larger than the area threshold value ε5 and the amount of heat taken away from the rotating electrical machine system 100 by cooling is large, that is, when performing a process of increasing the flow rate Q cool of the cooling oil supplied to the rotating electrical machine system 100 in step S22 above, switches the direction switching valve 46 to the second cooling oil passage 48b on the heat exchanger 52 side. In this state, in the heat exchanger 52, heat exchange is performed from the cooling oil flowing through the second cooling oil passage 48b to the cooling water flowing through the cooling water passage 54.

[0088] The cooling water pump 56 is driven by a pump 58. Cooling water is supplied from a cooling water pan 60 to the battery 50 via a cooling water passage 54 by the cooling water pump 56. The battery 50 is warmed up by supplying the cooling water heated by the cooling oil in the heat exchanger 52 to the battery 50. Such a warm-up process can be used for early warm-up of the battery 50.

[0089] Note that although the object to be heated in the electric drive system 110 is the battery 50, it is not limited thereto, and any other component other than the electric drive system can be the object to be heated.

[0090] [Configuration of the present invention] Configuration 1: A temperature prediction device for an electric drive system including a rotating electric machine, which is learned by applying machine learning, and predicts and outputs the temperature of each part of the electric drive system using a learning model that predicts the temperature of each part of the electric drive system with the state quantity during the operation of the electric drive system as input data. Configuration 2: The temperature prediction device according to Configuration 1, wherein the learning model is machine-learned to output the temperature of each part of the electric drive system when the state quantity is input, using teacher-student learning data in which the temperature of each part of the electric drive system during the operation of the electric drive system is combined with the state quantity during the operation of the electric drive system as teacher data. Configuration 3: The temperature prediction device according to Configuration 1 or 2, wherein the state quantity includes at least one of the rotational speed of the rotating electric machine, the output torque of the rotating electric machine, the input voltage of the rotating electric machine, the input current of the rotating electric machine, the input current density of the rotating electric machine, the temperature of the coil end of the stator coil of the rotating electric machine, the flow rate of the cooling fluid, the temperature of the cooling fluid, the ambient temperature, and the vehicle speed. Configuration 4: The temperature prediction device according to Configuration 2, The temperature prediction device is characterized in that the supervised learning data includes image data indicating the temperature at each part of the electric drive system. Configuration 5: The temperature prediction device according to Configuration 4, wherein the image data indicates the temperature distribution at each part of the electric drive system by any one of brightness, color density, and color separation. Configuration 6: The temperature prediction device according to Configuration 3, wherein the input data includes image data indicating the flow rate of the cooling fluid at each part of the electric drive system. Configuration 7: The temperature prediction device according to Configuration 6, wherein the image data indicates the distribution of the flow rate of the cooling fluid at each part of the electric drive system by any one of brightness, color density, and color separation. Configuration 8: The temperature prediction device according to any one of Configurations 1 to 7, wherein the machine learning is performed using a convolutional neural network. Configuration 9: The temperature prediction device according to any one of Configurations 1 to 8, wherein the temperature at each part of the electric drive system is predicted as a continuous value. Configuration 10: The temperature prediction device according to any one of Configurations 1 to 8, wherein the temperature at each part of the electric drive system is classified into a plurality of temperature classes and predicted. Configuration 11: Using the temperature prediction device according to any one of Configurations 1 to 10, the temperature at each part of the electric drive system during operation is predicted, and the operating conditions of the rotating electrical machine are controlled according to the predicted temperature. Configuration 12: The rotating electrical machine system according to Configuration 11, A rotating electrical machine system, characterized by predicting the temperature distribution of each part of the electric drive device during operation and controlling the operating conditions of the rotating electrical machine according to the area of the part where the predicted temperature is equal to or higher than a reference value. Configuration 13: Using the temperature prediction device according to any one of Configurations 1 to 10, predicting the temperature of each part of the electric drive system during operation, and controlling the flow rate of the cooling fluid of the electric drive system according to the predicted temperature. A rotating electrical machine system characterized by this. Configuration 14: The rotating electrical machine system according to Configuration 13, Predicting the temperature distribution of each part of the electric drive system during operation, and controlling the flow rate of the cooling fluid of the electric drive system according to the area of the part where the predicted temperature is equal to or higher than a reference value. A rotating electrical machine system characterized by this. Configuration 15: In the rotating electrical machine system according to any one of Configurations 11 to 14, the heat obtained by heat exchange with the cooling fluid of the electric drive system is used for heating. An electric drive system characterized by this.

Explanation of Signs

[0091] 10 Rotor, 10a Rotor Core Permanent Magnet, 10b Rotor Core Electromagnetic Steel Sheet, 12 Shaft, 14 Stator, 14a Stator Coil, 14b Stator Core Electromagnetic Steel Sheet, 16 Bearing, 18 Casing, 20 Inverter, 22 Battery, 24 Control Unit, 26 Oil Pump, 28 Motor, 29 Oil Pan, 30 Processing Unit, 32 Storage Unit, 34 Input Unit, 36 Output Unit, 38 Communication Unit, 40 Direction Switching Valve, 44 (44a~44c) Flow Control Valve, 46 Direction Switching Valve, 48 (48a, 48b) Cooling Oil Circuit, 50 Battery, 52 Heat Exchanger, 54 Cooling Water Circuit, 56 Cooling Water Pump, 58 Pump, 60 Cooling Water Pan, 100 Rotating Electrical Machine System, 101 Rotating Electrical Machine, 102 Transmission, 104 Differential Gear, 106 Drive Shaft, 108 Driving Wheel (Tire), 110 Electric Drive System, 112 Battery System, 200 Vehicle, 300 Temperature Prediction Device.

Claims

1. A temperature prediction device for an electric drive system including a rotating electrical machine, which is trained by applying machine learning, and predicts and outputs the temperature of each part of the electric drive system using a learning model that predicts the temperature of each part of the electric drive system with the state quantity during operation of the electric drive system as input data, wherein the learning model is machine-learned to output the temperature of each part of the electric drive system when the state quantity is input, using teacher-student learning data in which image data indicating the flow rate of the cooling fluid in each part of the electric drive system as the state quantity during operation of the electric drive system is combined with image data indicating the temperature of each part of the electric drive system during operation of the electric drive system as teacher data. The temperature prediction device is characterized by this.

2. The temperature prediction device according to claim 1, wherein the state quantity further includes at least one of the rotational speed of the rotating electrical machine, the output torque of the rotating electrical machine, the input voltage of the rotating electrical machine, the input current of the rotating electrical machine, the input current density of the rotating electrical machine, the temperature of the coil end of the stator coil of the rotating electrical machine, the temperature of the cooling fluid, the ambient temperature, and the vehicle speed. The temperature prediction device is characterized by this.

3. The temperature prediction device according to claim 1, wherein the image data indicates the temperature distribution in each part of the electric drive system by any one of brightness, color density, and color separation. The temperature prediction device is characterized by this.

4. The temperature prediction device according to claim 1, wherein the image data indicates the distribution of the flow rate of the cooling fluid in each part of the electric drive system by any one of brightness, color density, and color separation. The temperature prediction device is characterized by this.

5. The temperature prediction device according to claim 1, wherein the machine learning is performed using a convolutional neural network. The temperature prediction device is characterized by this.

6. A rotating electrical machine system, characterized in that the temperature of each part of the electric drive system during operation is predicted using the temperature prediction device according to claim 1, and the operating conditions of the rotating electrical machine are controlled according to the predicted temperature.

7. A rotating electrical machine system, characterized in that the temperature of each part of the electric drive system during operation is predicted using the temperature prediction device according to claim 1, and the flow rate of the cooling fluid of the electric drive system is controlled according to the predicted temperature.

8. An electric drive system, characterized in that in the rotating electrical machine system according to claim 6 or 7, heat obtained by heat exchange with a cooling fluid of the electric drive system is used for heating.

Images