CN-115250044-B - Magnet temperature estimation device

CN115250044BCN 115250044 BCN115250044 BCN 115250044BCN-115250044-B

Abstract

Provided is a magnet temperature estimation device which acquires parameters related to the rotation of a motor (2) measured for each fixed time, and calculates a moving average value of the parameters for each fixed interval. The calculated moving average is input to a learning model that has been learned so as to output the temperature of the magnet provided to the rotor (7) of the motor (2) when the moving average of the parameter relating to the rotation of the motor (2) is input, and the estimated value of the magnet temperature output from the learning model is obtained. Then, the obtained estimated value of the magnet temperature is outputted.

Inventors

Yono knows

Assignees

丰田自动车株式会社

Dates

Publication Date: 20260505
Application Date: 20220318
Priority Date: 20210426

Claims (4)

1. A magnet temperature estimation device is provided with: a parameter acquisition unit that acquires parameters including a stator coil temperature, a motor rotation speed, an oil temperature, and an oil pump rotation speed, which are measured for each fixed time, in relation to the rotation of the target motor; a calculation unit that calculates a moving average value of the parameter for each fixed section; A temperature obtaining part for inputting the moving average value calculated by the calculating part to a learning model which is learned in such a way that the temperature of the magnet mounted on the rotor of the target motor is outputted when the moving average value of the parameter is inputted, obtaining the estimated value of the magnet temperature outputted from the learning model, and An output unit configured to output the estimated value of the magnet temperature obtained by the temperature obtaining unit, The learning model is a one-dimensional convolutional neural network, and performs a convolutional process by a filter having the same number of parameters as the number of the parameters by shifting the filter little by little for each moving average value of the stator coil temperature, the motor rotation speed, the engine oil temperature, and the engine oil pump rotation speed, thereby obtaining an output value composed of the product of the value of each parameter and the value of the filter, and learning the one-dimensional convolutional neural network using the output value.
2. The magnet temperature estimation device according to claim 1, wherein, The parameters include motor torque, motor current, and inverter frequency in addition to stator coil temperature, motor speed, oil temperature, and oil pump speed.
3. The magnet temperature estimation device according to claim 1, wherein, The calculation section calculates an exponentially smoothed moving average as the moving average.
4. The magnet temperature estimation device according to claim 1, wherein, The interval at which the estimated value of the magnet temperature is obtained by the temperature obtaining unit is changed according to the driving load of the vehicle, and the interval at which the estimated value of the magnet temperature is obtained when the driving load of the vehicle is low is made longer than the interval at which the estimated value of the magnet temperature is obtained when the driving load of the vehicle is high.

Description

Magnet temperature estimation device Technical Field The present invention relates to a magnet temperature estimation device. Background In a rotary electric motor comprising a rotor and a stator having permanent magnets, when the temperature of the permanent magnets increases, the magnetic force of the permanent magnets decreases, so that the electric motor cannot output a predetermined output. Therefore, it is necessary to detect the temperature of the permanent magnet. However, in a commercially available vehicle, it is difficult to detect the temperature of the permanent magnet mounted to the rotor. Accordingly, a magnet temperature estimating device is known which estimates the temperature of a permanent magnet attached to a rotor (for example, refer to japanese patent application laid-open No. 2014-93867). In the magnet temperature estimation device, the temperature rise per unit time of the rotor is calculated from the difference between the heat dissipation per unit time from the rotor to the engine oil calculated based on the estimated temperature difference between the rotor and the temperature of the engine oil flowing around the rotor, and the heat dissipation per unit time from the rotor to the engine oil calculated based on the temperature rise of the engine oil, and the calculated temperature rise per unit time of the rotor is integrated to estimate the temperature of the rotor, that is, the temperature of the magnet. Disclosure of Invention However, when estimating the magnet temperature by accumulating the temperature rise per unit time of the rotor calculated in this way, there is a problem in that errors are accumulated during the period of accumulating the temperature rise per unit time, and the estimated value of the magnet temperature often deviates greatly from the actual magnet temperature. Therefore, according to the present invention, there is provided a magnet temperature estimation device comprising: a parameter acquisition unit that acquires parameters related to the rotation of the target motor measured for each fixed time; a calculation unit that calculates a moving average value for each fixed interval of the parameter; A temperature obtaining part for inputting the moving average value calculated by the calculating part to a learning model which is learned in such a way that the temperature of the magnet mounted on the rotor of the motor is outputted when the moving average value of the parameter related to the rotation of the motor is inputted, obtaining the estimated value of the magnet temperature outputted from the learning model, and And an output unit configured to output the estimated value of the magnet temperature obtained by the temperature obtaining unit. According to the present invention, the magnet temperature can be accurately estimated. Drawings Fig. 1 is an overall view of a magnet temperature estimation device. Fig. 2 is a timing chart for explaining a method of estimating the magnet temperature. Fig. 3 is a diagram for explaining one-dimensional convolution processing. Fig. 4A and 4B are diagrams for explaining one-dimensional convolution processing. Fig. 5 is a diagram showing the construction of a one-dimensional convolutional neural network. Fig. 6 is a diagram for explaining one-dimensional convolution processing. Fig. 7 is a diagram for explaining one-dimensional convolution processing. Fig. 8A and 8B are diagrams for explaining one-dimensional convolution processing. Fig. 9 is a timing chart for explaining a method of estimating the magnet temperature. Fig. 10A and 10B are diagrams showing a list of acquisition parameters and a list of moving averages, respectively. Fig. 11A and 11B are diagrams for explaining the simple moving average process and the exponential smoothing moving average process, respectively. Fig. 12 is a flowchart for estimating the magnet temperature. Fig. 13 is a functional configuration diagram of the present invention. Fig. 14 is a diagram showing a relationship between the estimation error of the magnet temperature and the frequency distribution. Fig. 15 is a diagram showing a relationship between an operation time interval and a driving load of a vehicle. Fig. 16 is a flowchart for estimating the magnet temperature. Fig. 17A and 17B are diagrams showing a relationship between the magnet temperature and the upper limit of the drive current and a flowchart for performing drive current control, respectively. Fig. 18A and 18B are diagrams showing a relationship between a magnet temperature and a driving current of an oil pump driving motor and a flowchart for executing driving control of the oil pump, respectively. Detailed Description Referring to fig. 1, 1 denotes a housing of a transaxle (transaxle) of a hybrid vehicle, 2 denotes a motor for driving the vehicle, 3 denotes an oil pump, and 4 denotes an oil cooler. The motor 2 includes a rotary shaft 6 rotatably supported by a bearing 5, a rotor 7 fixed to the rotary shaft 6, an