Electric drive unit, method for operating an electric drive unit, and method for calculating temperature

Abstract

The invention relates to a method (10) for calculating the temperature at an electric drive unit (12), an electric drive unit (12) for a vehicle comprising an electric motor having a stator and a rotor, and a method for operating an electric drive unit (12) of said type.

Description

Technical Field

[0001] The invention relates to a method for calculating a temperature according to the preamble of claim 1. The invention also relates to an electric drive unit according to the preamble of claim 9 and to a method for operating an electric drive unit according to claim 10. Background Art

[0002] In vehicles with an electric drive unit, for example with an electric motor as traction drive, knowledge of the temperatures prevailing in the electric motor is required in order to ensure the most reliable possible mode of the electric motor. By knowing the exact temperature in the electric motor, the performance limits of the electric motor can be exploited in a more targeted manner. If the electric drive unit reaches critical temperatures, the permanent magnets in the electric motor may demagnetize and the insulating material in the stator may melt.

[0003] The most reliable way to determine the actual temperature is to directly measure and monitor the temperature at the sensing area of ​​interest in the electric drive unit. However, direct temperature measurement is not always possible. For example, if the temperature sensor used to accurately measure the temperature cannot be attached to the intended detection area due to available space or due to functionality.

[0004] If the temperature at the selected detection area cannot be measured directly, for example if only a single temperature sensor is used, a temperature model may be used to calculate and estimate the temperature at other locations.A number of different methods are known for calculating the temperature in an electric drive unit.

[0005] US 2011 / 0084638 A1 describes a temperature estimation module which calculates the motor temperature in an electric motor from a measured current in the stator and a measured oil temperature.

[0006] In JP H 0654572 A, the temperature in the coil of the electric motor is calculated using a thermal model of the electric motor. The temperature in the coil is calculated from the temperature-dependent resistance of the coil, which is determined by measuring voltage and current.

[0007] In WO 2015 101 107 A1, the temperature in the electric motor is estimated based on a motor temperature model from the temperature in power electronics that control the electric motor. Summary of the invention

[0008] The object of the present invention is to calculate the temperature on an electric drive unit in a more precise and reliable manner.

[0009] At least one of these objects is achieved by the method according to the invention for calculating the temperature. Thus, even if there is a fault in the temperature sensing element, the state variable can be calculated while taking into account the large heat input to the electric drive unit. During an interruption of the temperature sensing element, for example in the event of a fault in the temperature sensing element, a complete suspension of the temperature calculation may not occur. Thus, even if the measured variable is missing, the temperature calculation of the state variable can be performed during a fault in the temperature sensing element. The accuracy of the temperature calculation is improved.

[0010] The state space model may be a linear time-invariant (LTI) model between input variables and output variables. The LTI model may be described by the following two equations.

[0011]

[0012] Where x is the state variable, u is the input variable, y is the output variable, A is the system matrix, B is the input matrix, and C is the output matrix.

[0013] The state variable is preferably a non-measurable value, such as temperature and / or heat flow at an inaccessible location.

[0014] The system observer may be a Romberg observer and may be described by the following two equations.

[0015]

[0016] Among them, x m is the model state variable, u is the input variable, y is the measurement variable corresponding to the output variable, and y m is the model output variable, A is the system matrix, B is the input matrix, C is the output matrix, and L is the feedback matrix.

[0017] In a preferred embodiment of the present invention, the throughput mode is turned on when there is a detection status signal corresponding to a fault in the temperature detection element.

[0018] In a specific embodiment of the present invention, when there is a detection state signal corresponding to the normal mode of the temperature detection element, the control mode is turned on.

[0019] In another specific embodiment of the invention, the input variable indicates the heat input to the electric drive unit. The model state variable may be a temperature different from the model output variable, and the model output variable may be a temperature representing the measured variable. The input variable may be electrical power, voltage and / or heat flow.

[0020] In a preferred embodiment of the invention, the throughput mode corresponds to the control mode except that the model output variables are set to the difference of the measured variables and thus the output variable differences are set to zero.

[0021] In a specific embodiment of the invention, an error signal is outputted according to a calculated state value, which is dependent on the detected state signal. If the calculated state value corresponds to a value indicating a fault, the error signal may be outputted.

[0022] In another specific embodiment of the present invention, once the presence detection status signal corresponds to a fault in the temperature detection element for a first time period longer than a first specified time period, the calculated status value takes a value indicating a fault.

[0023] In a preferred embodiment of the present invention, the state of the temperature model can be described by a model state signal, and the model state signal indicates the accuracy of the temperature model according to the output variable difference. Once there is a fault in the temperature model corresponding to the model state signal within a second time period longer than the second specified time period, the calculated state value can take a value indicating the fault.

[0024] The calculated status value is preferably dependent on the first time period, the first specified time period, the second time period and the second specified time period.Depending on the calculated status value, the electric drive unit may enter an emergency mode.

[0025] At least one of the objects specified above is also achieved by an electric drive unit for a vehicle having an electric motor with a stator and a rotor, wherein at least one model state variable of the electric drive unit is calculated using a method having at least one of the features specified above.

[0026] At least one of the objects indicated above is also achieved by a method for operating an electric drive unit of this type, in which the electrical energy introduced into the electric drive unit is regulated as a function of a model state variable.

[0027] Further advantages and advantageous embodiments of the invention emerge from the description of the figures and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The present invention is described in detail below with reference to the accompanying drawings. Specifically:

[0029] Figure 1 A circuit block diagram of a method for temperature calculation in a specific embodiment of the present invention is shown.

[0030] Figure 2 The time course of the detection state signal and the model state signal in the method for temperature calculation in another specific embodiment of the present invention is shown. DETAILED DESCRIPTION

[0031] Figure 1A block diagram of a method 10 for temperature calculation in a specific embodiment of the present invention is shown. In particular, the method 10 for temperature calculation is performed on an electric drive unit 12 fed by electrical energy. The electric drive unit 12 can be arranged in a vehicle and can have an electric motor with a stator and a rotatable rotor.

[0032] The temperature of the electric drive unit 12 as an output variable depends on the input variable u by means of the system environment 14 of the electric drive unit 12. The system environment 14 can be described by a temperature model 16 through a state space model, through which the relationship between the input variable u and the output variable can be specified. The input variable u indicates the thermal input to the electric drive unit 12. For example, the input variable u can be electric power, voltage and / or heat flow.

[0033] The actual temperature of at least one location of the electric drive unit 12 is measured as a measured variable y by a temperature detection element. The temperature detection element is, for example, a temperature sensor arranged at a predetermined location of the electric drive unit 12.

[0034] The temperature model 16 can also be used to calculate state variables x of the electric drive unit 12 that cannot be measured directly. State variables x are, for example, unmeasurable temperatures and / or heat flows at inaccessible locations.

[0035] The temperature model 16 built on the state space model is assigned to the system observer 18, which uses the temperature model 16 to calculate the state variable x representing the actual state variable x from the input variable u. m and the model output variable y representing the measured variable y m The state space model is preferably a linear time-invariant (LTI) model between the input variables u and the measured variables y. The LTI model can be described by the following two equations.

[0036]

[0037] Where A is the system matrix, B is the input matrix, and C is the output matrix.

[0038] The system observer 18 calculates the measured variable y and the model output variable y m The output variable difference Δy between them is used to adjust the state variable x m And the state variable x m The system observer 18 may be a Romberg observer and may be described using the following two equations.

[0039]

[0040] Where A is the system matrix, B is the input matrix, C is the output matrix, and L is the feedback matrix.

[0041] The system observer 18 provides the state variable x according to the output variable difference Δy m Therefore, the temperature model 16 is supported by the measured variable y and the calculated state variable x m The deviation from the actual state variable x decreases.

[0042] On the other hand, if the measured variable y cannot be obtained in the event of a failure of the temperature detection element, the system observer 18 lacks the measured variable y as an input value. As a possible result, a failure of the temperature detection element may be indicated and the temperature calculation is stopped. However, the electric drive unit 12 operating in the absence of the temperature calculation still experiences a thermal input that is no longer considered in the suspended temperature model 16. If the temperature calculation is started again after the failure in the temperature detection element has been eliminated, the calculated state variable x is not considered due to the information that was not considered during the failure. m There may be a large difference between the actual state variable x.

[0043] Therefore, in the method 10 for temperature calculation, the fault and normal modes of the temperature detection element can be identified by detecting the state signal c. According to the detected state signal c, the system observer 18 can switch between the control mode and the throughput mode. In the control mode, the state variable x is controlled according to the output variable difference Δy. m , in throughput mode, the state variable x is calculated independently of the measured variable y m When there is a detection status signal corresponding to a failure of the temperature detection element, the throughput mode is turned on, and when there is a detection status signal corresponding to a normal mode of the temperature detection element, the control mode is turned on.

[0044] Therefore, even if the temperature sensing element fails, the state variable x can be calculated m At the same time, a large energy input to the electric drive unit 12 is taken into account. During a temperature sensing element failure, a complete suspension of the temperature calculation does not occur. Thus, even if the measured variable y is missing, the state variable x can be calculated during a temperature sensing element failure. m Temperature calculation.

[0045] The calculation 20 of the output variable difference Δy is preceded by a state selection 22, which selects the measured variable y or the model output variable y according to the detection state signal s1. m Forwarded to the calculation of the output variable difference Δy 20. If the throughput mode is turned on according to the detection status signal s1 - in particular, if the detection status signal s1 indicates a fault in the temperature detection element, then the condition selection 22 will model the output variable y m Output to the calculation 20 of the output variable difference Δy. Since the calculation 20 converts the model output variable y mAssuming the second difference element, the output variable difference Δy is therefore zero in the continuous mode and thus corresponds to the open loop mode of the temperature calculation. This means that despite the presence of a fault in the temperature sensing element and the lack of the measured variable y as a support point for the temperature calculation, the temperature can still be calculated by the throughput mode.

[0046] If the control mode is turned on according to the detection state signal s1, in particular, the control mode is turned on when the detection state signal s1 indicates the normal mode of the temperature detection element, the condition selection 22 outputs the measured variable y to the calculation of the output variable difference Δy. This means that the measured variable y can be used as a reference point for the temperature calculation, and the temperature calculation can be performed more accurately.

[0047] Figure 2 The time course of the detection state signal s1 and the model state signal s2 in the method for temperature calculation in another specific embodiment of the present invention is shown. The detection state signal s1 is 1 when the temperature detection element fails and is 0 when the temperature detection element is in normal mode.

[0048] The model state signal s2 indicates the accuracy state of the temperature model and depends on the output variable difference and a predetermined maximum output variable difference. When the predetermined maximum output variable difference is exceeded, the value of the model state signal s2 is 1, and when the predetermined maximum output variable difference is not exceeded, the value of the model state signal s2 is zero.

[0049] When both the detection state signal s1 and the model state signal s2 are 0, the calculation state value c is 0. However, if one of the two signals is 1, the calculation state value c does not immediately take the value 1. The calculation state value c is set to the value 1 only when the detection state signal s1 has the value 1 in a first time period t1 that is longer than the first specified time period T1. In addition, when the model state signal s2 takes the value 1 in a second time period t2 that is longer than the second specified time period T2, the calculation state value c is changed to the value 1.

[0050] The calculated state value c of 1 indicates a fault and thus triggers the output error signal. For example, if the calculated state value c has a value of 1, ie an output error signal is output, the electric drive unit may enter an emergency mode.

[0051] List of attached pictures

[0052] 10 Method 12 Electric drive unit 14 System environment 16 Temperature model 18 System observer 20 Calculation 22 Condition selection A System matrix B Input matrix C Output matrix c Calculation state value s1 Detection state signal s2 Model state signal T1 First specified time period t1 First time period T2 Second specified time period t2 Second time period u Input variable x State variable x mModel state variable Δy Output variable difference y Measurement variable y m Model output variables.

Claims

1. A method (10) for calculating the temperature of an electric drive unit (12), the temperature of the electric drive unit being derived as an output variable from an input variable (u) via a system environment (14) of the electric drive unit (12), The temperature model (16) describes the system environment (14) via a state space model, and with the aid of the temperature model a model output variable (y) representing the output variable is calculated from the input variable (u). m ), the temperature model is assigned to a system observer (18), which compares the actual temperature at at least one measurement location of the electric drive unit (12) as a measurement variable (y) corresponding to the output variable measured by at least one temperature detection element with the model output variable (y m ) to adjust and output the model state variable (x) of the system environment (14) m ), It is characterized in that indicating the fault and normal modes of the temperature sensing element by means of a sensing status signal (s1), and The system observer (18) is capable of switching between a control mode and a throughput mode according to the detection state signal (s1), wherein in the control mode, the model state variable (x) is controlled according to the output variable difference (Δy). m ), in the throughput mode, the model state variable (x) is calculated independently of the measured variable (y) m ); wherein an error signal is outputted based on a calculated state value (c) dependent on the detection state signal (s1); the state of the temperature model (16) is described by a model state signal (s2), the model state signal indicates the accuracy of the temperature model (16) based on the output variable difference (Δy), and once there is an inaccuracy of the model state signal (s2) corresponding to the temperature model (16) within a second time period (t2) longer than a second specified time period (T2), the calculated state value (c) takes a value indicating a fault.

2. The method (10) according to claim 1, characterized in that: When there is a detection status signal (s1) corresponding to a fault in the temperature detection element, the throughput mode is turned on.

3. The method (10) according to claim 1 or 2, characterized in that: When there is a detection state signal (s1) corresponding to the normal mode of the temperature detection element, the control mode is turned on.

4. The method (10) according to claim 1 or 2, characterized in that The input variable (u) indicates the heat input into the electric drive unit (12), and the model state variable (x m ) is different from the model output variable (y m ) temperature, and the model output variable (y m ) is the temperature representing the measured variable (y).

5. The method (10) according to claim 1 or 2, characterized in that In addition to the model output variables (y m ) is set to the measured variable (y) and thus the output variable difference (Δy) is set to zero. The throughput mode corresponds to the control mode.

6. The method (10) according to claim 1, characterized in that: Once the presence of the detection status signal (s1) corresponds to a fault in the temperature detection element for a first time period (t1) longer than a first specified time period (T1), the calculated status value (c) takes a value indicating a fault.

7. An electric drive unit (12) for a vehicle, the electric drive unit having an electric motor with a stator and a rotor, It is characterized in that At least one model state variable (x m ) is calculated by a method (10) according to any of the preceding claims.

8. A method for operating an electric drive unit (12) according to claim 7, It is characterized in that The electric energy introduced into the electric drive unit (12) is based on the model state variable (x m ) to set.

Images