Method for predicting the rotor temperature for an electric motor

Abstract

A method for predicting the rotor temperature of an electric motor (12) for an electric vehicle, comprising: Measuring at least one operating parameter of the electric motor (12); Entering the at least one operating parameter of the electric motor (12) into a predetermined regression model in order to determine a rotor temperature (T rotor ) of the electric motor (12); and Transmitting a predicted rotor temperature (T rotor ) of the electric motor (12) to a control module for managing at least one thermal limit value of the electric motor (12), wherein the at least one operating parameter of the electric motor (12) comprises a torque output, a rotor speed and a stator temperature (T stator ) includes; wherein the at least one operating parameter of the electric motor (12) further comprises a coolant flow rate; wherein the predetermined regression model represents a mathematical relationship between a measured rotor temperature (T rotor ) of a reference electric motor (12) and at least one measured operating parameter of the reference electric motor (12) including a torque output, a rotor speed, a stator temperature (T stator ) and a coolant flow rate; wherein the regression model is developed by a procedure comprising: Operating the reference electric motor (12) at a plurality of predetermined rotor speeds; Operating the reference electric motor (12) at a plurality of predetermined torque levels at each of the plurality of predetermined rotor speed levels; Measuring a rotor temperature (T rotor ) and a stator temperature (T stator ) of the reference electric motor (12) at each of the plurality of predetermined torque levels at each of the plurality of predetermined rotor speeds; Measuring at least one operating parameter of the reference electric motor (12) at each of the plurality of predetermined torque levels at each of the plurality of predetermined rotor speed levels; and Using regression analysis to establish a mathematical relationship between a difference between the measured rotor temperature (T rotor ) and the measured stator temperature (T stator ) of the reference electric motor (12) and the at least one measured operating parameter of the reference electric motor (12).

Description

The present disclosure relates to a method for predicting the rotor temperature of an electric motor, in particular an induction motor.Electric motors are typically used to propel electric vehicles, including hybrid electric vehicles. Electric motors require accurate measurements of motor temperatures during operation to ensure optimum motor performance, avoid thermal over protection that could limit performance, and maximize life expectancy of the motor. It is important to measure or estimate the temperature of the rotor of the electric motor in order to determine the temperature conditions during operation. However, the known techniques for determining the temperature of an electric motor during operation do not allow direct and non-destructive temperature measurement of the rotor, in particular on the rotor surface in the air gap, since the rotor is not accessible.The known temperature measurements include a back-emf method which, in the case of permanent magnet motors, indirectly estimates the total temperature of the rotor magnet. However, this method is not applicable to three-phase asynchronous machines, also referred to as induction motors, since they do not have permanent magnets in the rotor and thus there is no back EMF. At present, there are no suitable methods for measuring the rotor temperature of induction motors.US 2022 / 0 052 633 A1 describes an induction motor control system which performs a dynamic on-board model to estimate rotor resistance and control torque output by the induction motor. The model includes equations for calculating stator and rotor temperatures and / or resistances based on combinations of voltage and current data, electrical frequency, rotor speed, switching patterns, and air flow rates during operation of the induction motor. The control system updates the model based on feedback collected during operation of the induction motor, including the difference between the actual observed stator temperature and the stator temperature predicted by the model.While current systems for measuring motor temperature serve their purpose, there is a need for a new and improved system and method for measuring or predicting rotor temperature of electric motors during operation of the motor.Accordingly, it is the object of the invention to provide a system and method that measure or predict the rotor temperature of electric motors during engine operation.The object is achieved with the subject matter of the independent claim.According to several embodiments, a method for predicting the rotor temperature of an electric motor for an electric vehicle is described.The method according to the invention comprises measuring at least one operating parameter of the electric motor; inputting the at least one operating parameter of the electric motor into a predetermined regression model in order to predict a rotor temperature of the electric motor; and transmitting the predicted rotor temperature of the electric motor to a control module for managing at least one thermal limit value of the electric motor. The at least one measured operating parameter of the electric motor comprises a torque output, a rotor rotational speed, a stator temperature and a coolant flow rate. The predetermined regression model is further a mathematical relationship between a measured rotor temperature of a reference electric motor and at least one measured operating parameter of the reference electric motor, including a torque output, a rotor speed, a stator temperature, and a coolant flow rate. The regression model is developed by: operating the reference electric motor at a plurality of predetermined rotor speeds; operating the reference electric motor at a plurality of predetermined torque levels at each of the plurality of predetermined rotor speed levels; measuring a temperature difference between a rotor temperature and a stator temperature of the reference electric motor at each of the plurality of predetermined torque levels at each of the plurality of predetermined rotor speeds; measuring at least one operating parameter of the reference electric motor at each of the plurality of predetermined torque levels at each of the plurality of predetermined rotor speed levels; and using regression analysis to develop a mathematical relationship between a difference in the measured rotor temperature and the measured stator temperature of the reference electric motor versus the at least one measured operating parameter of the reference electric motor.In another embodiment of the present specification, the reference electric motor is one of the following: (i) a reference device electric motor configured to have substantially the same performance and operating characteristics as the vehicle electric motor, and (ii) a modified reference series electric motor having means for measuring the rotor temperature of the modified vehicle electric motor.In another embodiment of the present description, the at least one measured operating parameter of the reference electric motor includes a stator temperature, a torque level output, a rotor speed, and a coolant flow rate.In another embodiment of the present description, the regression model expresses a causal relationship between a difference of the rotor temperature minus the stator temperature as a function of the at least one measured operating parameter of the reference electric motor.In another embodiment of the present description, the mathematical relationship is calibrated based on the measured differences between stator temperatures and rotor temperatures and the at least one measured operating parameter of the reference electric motor.According to one application, a method for developing a prediction model for predicting a rotor temperature of an electric motor in a vehicle according to the invention is described. The method includes operating a reference induction motor at a plurality of predetermined levels of coolant flow rate; operating the reference induction motor at a plurality of predetermined rotor speed levels at each of the plurality of predetermined levels of coolant flow rate; operating the reference induction motor at a plurality of predetermined torque levels at each of the plurality of predetermined rotor speed levels at each of the plurality of predetermined levels of coolant flow rate; measuring a rotor temperature, a stator temperature, and an operating parameter of the reference induction motor at each of the plurality of predetermined torque levels at each of the plurality of predetermined rotor speed levels for each of the plurality of predetermined levels of coolant flow rate; and developing a mathematical relationship between a difference between the measured stator temperature and the measured rotor temperature of the reference induction motor as a function of the measured operating parameter.According to a further application, a method for predicting a rotor temperature of an electric motor for an electric vehicle according to the invention is described. The method includes measuring a torque level output, a rotor speed, a stator temperature, or a coolant flow rate of the electric motor; inputting the measured torque level output, the rotor speed, the stator temperature, or the coolant flow rate of the electric motor into a predetermined prediction model to predict a rotor temperature of the electric motor; and communicating the predicted rotor temperature of the electric motor to a control module for managing at least one thermal limit of the electric motor.In another embodiment of the present specification, the predetermined prediction model is a regression model expressing a causal relationship between a difference between the rotor temperature minus the stator temperature and the at least one measured operating parameter of the reference electric motor.Further areas of applicability will become apparent from the description herein. It should be understood that the specification and specific examples are intended to be illustrative only and not to limit the scope of the present description.The drawings described herein are for illustrative purposes only and are not intended to limit the scope of the present description in any way. FIG. 1 is a front view of an induction motor including an apparatus and method for rotor temperature measurement according to an example aspect; FIG. 2 is a cross-sectional top view of the end taken in section 2 of FIG. 1 ; FIG. 3 is a flow block diagram of a method for developing a regression model for predicting the rotor temperature of an induction motor; FIG. 4 is an exemplary graph showing the relationship between the difference in measured rotor temperatures minus the measured stator temperatures and the measured stator temperature for multiple predetermined motor speeds at 100% torque output; FIG. 5A is an exemplary regression model for predicting the rotor temperature of an induction motor based on the relationship of the operating parameters shown in the graph of FIG. 4; FIG. 5B is an exemplary general regression model for predicting the rotor temperature of an induction motor based on the relationship of the operating parameters shown in the graph of FIG. 4; and FIG. 6 is a block flow diagram of a method for predicting the rotor temperature of an induction motor for an electric vehicle.The term "control module" refers to one or more combinations of one or more processors, associated memory, and other components that can execute software, firmware, program, instruction, routine, code, or algorithm to provide the described functions. The processors include, but are not limited to, application specific integrated circuits (ASICs), electronic circuits, central processing units, microprocessors, and microcontrollers. The associated memory includes, among other things, read-only memory (ROM), random-access memory (RAM), and electrically programmable read-only memory (EPROM). The functions of a control module as described in this disclosure may be performed in a distributed control architecture among multiple networked control modules. A control module may include a variety of communication interfaces, including point-to-point or discrete lines and wired or wireless interfaces to other control modules.Software, firmware, programs, instructions, routines, code, algorithms, and similar terms refer to all executable instruction sets for control modules including methods, calibrations, data structures, and look-up tables. A control module has a series of control routines that are executed to provide the described functions. The routines are executed by, for example, a central processing unit and may monitor inputs from sensor devices and other networked control modules and execute control and diagnostic routines to control operation of actuators. The routines may be executed periodically during ongoing vehicle operation. Alternatively, routines may be executed in response to the occurrence of an event, software calls, or requests via user interface inputs or requests.U.S. Patent Application Serial No. 17 / 110,865, filed on 3,12,2020, entitled "Apparatus and Method for Rotor Temperature Measurement" (apparatus and method for measuring rotor temperature) (hereinafter "U.S. Appl. No. '865") teaches a method for measuring the rotor temperature of an electric motor in motor vehicles. Instead of implementing in situ rotor temperature measurement methods in electric vehicles, as described in U.S. Pat. No. Appl. No. '865, a novel method for estimating or predicting rotor temperatures in electric vehicles has been developed that is based on measurable, determinable, or otherwise determinable operating parameters of the electric motor, such as torque output, rotor speed, stator temperature, and coolant flow rate. This new method is a physical prediction model or a mathematical regression model based on rotor temperature measurements at all relevant engine operating conditions of a reference electric motor. The measurements of the rotor temperature of the reference electric motor can be based on the techniques as described in U.S. Pat. No. Appl. No. '865. U.S. Patent No. which is incorporated herein by reference in its entirety.The method for predicting the rotor temperature of an electric motor for an electric vehicle will be described. The electric vehicle may include fully electric or hybrid land vehicles, watercraft, and aircraft that use at least one electric motor for propulsion. The method uses a mathematical regression model to predict the rotor temperature of the electric motor by inputting one or more of the following: torque output, rotor speed, stator temperature, and coolant flow rate of the electric motor. The method is particularly applicable to electric induction motors because there is no back EMF typically used to estimate the rotor temperatures of most electric motors. The mathematical regression model can be stored as a software routine or a look-up table on non-transitory computer readable storage media and executed by a control module to control the performance of the induction motor and control thermal protection to avoid exceeding the thermal limits of the induction motor.The mathematical regression model, also referred to as "the regression model" for brevity, is developed from experimental data collected from a reference motor. In one embodiment, the reference motor may be a fixed-position laboratory motor, also known as a machine motor, configured to have substantially the same performance and operating characteristics as a series motor. Having substantially the same power and operating characteristics means that torque, rotor speed, stator temperature, and coolant flow rate are within + / -10% for a given power input of the induction series motor. In another embodiment, the reference motor may be a modified series motor configured such that torque output, rotor speed, stator temperature, and coolant flow rate may be directly measured, estimated, or otherwise determined. The reference device motor and the modified reference series motor may be an electric induction motor.FIG. 1 shows an example referencing device motor 12 configured to have substantially the same operating and performance characteristics as an induction series motor for an electric vehicle. The motor 12 includes a rotor 14 for rotating an output shaft 16 with respect to a shaft rotation axis 18. For the precise and direct determination of a rotor temperature T rotor at a rotor surface 22, a first through-opening 24 is provided through the stator 20. The first through hole 24 provides line-of-sight access to the rotor surface 22 To accurately and directly determine the stator temperature T stator a second through hole 26 is provided through a housing 28 that supports the stator 20. The first through-opening 24 and the second through-opening 26 are aligned coaxially on a common axis 30. To determine the rotor temperature T rotor on the rotor surface 22 and the stator temperature T stator on the stator surface 32, a sensor 34, e.g., an infrared sensor, is attached to the housing 28 with a sensor mount 36.To control the temperature of the engine 12, an internal passage 50 is provided for a flow of coolant oil through the engine 12. The cooling oil may be an engine oil or a hydraulic oil such as an automatic transmission oil (ATF). The output shaft 16 of the motor of the apparatus 12 may be connected to sensors for measuring the output of the torque level and the rotor speed.FIG. 2 is a cross-sectional view of the motor 12 taken along section line 2- 2. The stator 20 includes a plurality of stator laminations 38 each having an outer ring 40 and a plurality of inboard lamination teeth 42. A plurality of stator slots 44 are individually disposed between successive lamination teeth 42. Stator windings 46, for example made of copper wire, are arranged in the stator slots 44. The first through-opening 24 is produced, for example, by central drilling through one of the sheet metal teeth 42'. In the example shown, a first diameter of the first through-hole 24 is less than a second diameter of the second through-hole 26 that allows for line-of-sight access to the rotor surface 22 and line-of-sight access to the stator surface 32 through the sensor 34.Note that the first through hole 24, the second through hole 26, the sensor 34, and the sensor holder 36 of the referencing device motor 12 are provided in the referencing device motor 12 to obtain rotor temperature data for the development of the regression model. However, the first through hole 24, the second through hole 26, the sensor 34, and the sensor holder 36 are not required in the series motor because the regression model is used to predict the rotor temperature of the series motor.FIG. 3 shows a flow block diagram for a method for generating the regression model for predicting a rotor temperature of a series electric motor of an electric vehicle. The method begins at block 302, where the referencing device motor is configured to have substantially the same, if not the same, performance and operating characteristics as the series electric motor.At block 304, the reference device motor is operated at a plurality of predetermined rotor speeds, also referred to as motor speed, measured in revolutions per minute (U / min). Each of the predetermined rotor speeds is at a particular speed, from 0 + rpm to the maximum operating speed, which together represent the entire operating range of the series motor. At each of the predetermined rotor speeds, the reference device motor is operated at a plurality of predetermined torque levels ranging from 0 + to 100% torque output. The rotor temperature, the stator temperature, and the coolant flow rate are measured at each of the plurality of predetermined torque levels within each of the plurality of predetermined rotor speeds.In block 306, the difference of the measured rotor temperature minus the measured stator temperature is plotted against the measured stator temperature at the various predetermined rotor speeds for each of the predetermined torque levels. An exemplary graph is shown in FIG. 4 that shows the relationship between the difference in measured rotor temperature minus the measured stator temperature (y-axis) and the measured stator temperature (x-axis) at 100% torque level for each of the illustrated predetermined rotor speeds. Preferably, the predetermined rotor speeds and torques meet the extreme conditions expected for the vehicle electric motor.At block 308, a regression model is developed using regression analysis for the causal relationship between the difference in rotor temperature minus the stator temperature as a function of the stator temperature, rotor speed, torque level, and coolant flow rate of the motor of the device. Regression analysis is a series of statistical methods for estimating the relationships between a dependent variable, in this case the difference between rotor temperature and stator temperature, and multiple independent variables, in this case the stator temperature, the torque level, the rotor speed, and the coolant flow rate. Regression analysis is used to evaluate the strength of the relationship between these variables and model the relationship between them. An example regression model is shown in FIG. 5A for predicting the rotor temperature of a series electric motor using the measured data from the reference device model. FIG. 5B shows an example general regression model for predicting the rotor temperature of a series electric motor using the measured data from the reference model of the apparatus. In FIGS. 5A and 5B, a-g are constants specific to the motor that the equations represent.At block 310, the rotor temperature predicted from the regression model is compared to the measured rotor temperature of the reference device motor operating at the same rotor speed, torque level, and coolant flow rate as were input to the regression model. The regression model is considered to be within an acceptable error if the predicted rotor temperature deviates plus or minus 10°C from the measured rotor temperature of the reference device motor.At block 312, the developed regression model is stored as a software routine or a look-up table on non-transitory computer readable storage media and may be executed by a control module to control the performance of the induction motor and control thermal protection to avoid exceeding the thermal limits of the induction motor.FIG. 6 shows a functional block flow diagram of a method 600 for predicting a rotor temperature of a series electric motor for an electric vehicle using the developed regression model. In block 602, the rotor speed of the series electric motor is determined. In block 604, the torque output of the series electric motor is determined. In block 606, the stator temperature of the series electric motor is determined. At block 608, the coolant flow rate through the series electric motor is determined.At block 610, the determined torque output, rotor speed, stator temperature, and coolant flow rate are input to the regression model to predict the rotor temperature of the electric motor.At block 612, the predicted rotor temperature from block 610 is communicated to a vehicle or motor controller 614 along with the determined speed, torque level, stator temperature, and coolant flow rate. The vehicle controller 614 is configured to process the information for managing the electric motor. The regression model may be stored as a software routine or a look-up table on a non-transitory computer readable storage medium 616 that is accessible by the control module 614.

Claims

A method for predicting the rotor temperature of an electric motor (12) for an electric vehicle, comprising: measuring at least one operating parameter of the electric motor (12); inputting the at least one operating parameter of the electric motor (12) into a predetermined regression model to predict a rotor temperature (T rotor) of the electric motor (12); and transmitting a predicted rotor temperature (T rotor) of the electric motor (12) to a control module for managing at least one thermal threshold of the electric motor (12), wherein the at least one operating parameter of the electric motor (12) comprises a torque output, a rotor speed, and a stator temperature (T stator) ; wherein the at least one operating parameter of the electric motor (12) further comprises a coolant flow rate; wherein the predetermined regression model is a mathematical relationship between a measured rotor temperature (T rotor) of a reference electric motor (12) and at least one measured operating parameter of the reference electric motor (12) including a torque output, a rotor speed, a stator temperature (T stator) and a coolant flow rate; wherein the regression model is developed by a method comprising: operating the reference electric motor (12) at a plurality of predetermined rotor speeds; operating the reference electric motor (12) at a plurality of predetermined torque levels at each of the plurality of predetermined rotor speed levels; measuring a rotor temperature (T rotor) and a stator temperature (T stator) of the reference electric motor (12) at each of the plurality of predetermined torque levels at each of the plurality of predetermined rotor speeds; measuring at least one operating parameter of the reference electric motor (12) at each of the plurality of predetermined torque levels at each of the plurality of predetermined rotor speed levels; and using regression analysis to develop a mathematical relationship between a difference between the measured rotor temperature (T rotor) and the measured stator temperature (T stator) of the reference electric motor (12) and the at least one measured operating parameter of the reference electric motor (12).The method of claim 1, wherein the at least one operating parameter of the electric motor (12) is selected from the group of operating parameters consisting of a torque output, a rotor speed, a stator temperature (T stator) and a coolant flow rate.The method of claim 1, wherein the reference electric motor (12) is one of the following motors: (i) a reference device electric motor configured to have substantially the same performance and operating characteristics as the electric motor for the electric vehicle; and (ii) a modified reference series electric motor having means for measuring the rotor temperature (T rotor) of the modified series electric motor.The method of claim 1, wherein the at least one measured operating parameter of the reference electric motor (12) comprises a stator temperature (T stator), a torque level output, a rotor speed, and a coolant flow rate.The method of claim 1, wherein the mathematical relationship is calibrated from the differences between the measured stator temperatures (T stator) and the measured rotor temperatures (T rotor) and the at least one measured operating parameter of the reference electric motor (12).The method of claim 1, wherein the regression model expresses a causal relationship between a difference of the rotor temperature (T rotor) minus the stator temperature (T stator) as a function of the at least one measured operating parameter of the reference electric motor (12).

Images