SYSTEM AND METHOD FOR ESTIMATING THE ENGINE TEMPERATURE OF AN ENGINE

DE502022006840D1Active Publication Date: 2026-02-19EBM PAPST MULFINGEN GMBH & CO KG
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
DE502022006840
Authority / Receiving Office
DE · DE
Patent Type
Patents
Current Assignee / Owner
Priority Date
2021-03-08
Filing Date
2022-02-18
Publication Date
2026-02-19
Estimated Expiration
2042-02-18

AI Technical Summary

Technical Problem

Existing methods for determining motor temperature without sensors face implementation challenges and inefficiencies, necessitating a reliable method for monitoring engine temperature to prevent overheating.

Method used

A system utilizing a thermal model, electrical input power, and shaft speed to calculate motor temperature through a system of differential equations, solved either numerically or analogously using RC circuits, without requiring physical sensors in the motor.

Benefits of technology

Accurately estimates motor temperature by calculating thermal power loss and integrating it with a reference temperature, effectively preventing engine overheating.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader
Need to check novelty before this filing date? Find Prior Art

Description

[0001] The present invention relates to a system and method for determining or estimating the engine temperature of an engine, in particular for protecting against engine overheating. The invention further relates to an engine equipped with such a system.

[0002] Known methods for measuring motor temperature include, for example, installing sensors in the winding, determining the winding temperature via state estimators (Kalman filter, gradient method) from the motor differential equations, directly measuring the winding resistance during operation by coupling a measuring voltage, or calculating the motor temperature using thermal network models of the motor.

[0003] Recently, there has been an increasing effort to determine the temperature inside an electric motor without having to install a temperature sensor or similar measuring device in the motor, i.e., in the electromechanical part of the motor.

[0004] From DE 10 2008 040 725 A1, for example, a method for determining a rotor temperature of a permanent excitation synchronous machine is known, in which an estimated value for the rotor temperature is determined by means of a Kalman filter containing a thermal model of the synchronous machine.

[0005] From EP 2977733 A1, a further method for determining the winding temperature of an excitation winding of a separately excited synchronous machine using a Kalman filter is known. In this method, an initial temperature estimate of the winding temperature is specified for a starting time. At various measurement times following the starting time, an electrical excitation voltage applied to the excitation winding and an electrical excitation current flowing through the excitation winding are recorded. For each measurement time, an updated temperature estimate is recursively generated as a function of the excitation voltage recorded at that measurement time, the excitation current recorded at that measurement time, and a previous temperature estimate.

[0006] From DE 102019105081 A1, a device for determining the temperature of a winding of a rotating field machine with a stator and a rotor is known, comprising at least a first Kalman filter (K1) for calculating the temperature and a rotating field control for vector control of the rotating field machine with at least one id current controller, which is configured to impose a high-frequency voltage signal (HF) on the winding voltage in order to obtain a winding current with a high-frequency superimposed current component for the winding, which is supplied to the Kalman filter (K1), wherein the Kalman filter (K1) has a high-frequency model of the motor in order to determine the high-frequency resistance of the winding and from this, in turn, the winding temperature by means of a calculation function.

[0007] However, the methods known in the prior art each have specific disadvantages in terms of implementation or application, so there remains a need to determine a reliable monitoring of the motor temperature inside an electric motor without having to install a temperature sensor or similar measuring device in the motor, i.e., in the electromechanical part of the motor.

[0008] Further prior art in the present technical field is disclosed in documents EP 3 107 204 A1, US 2020 / 341 062 A1, US 5 600 575 A, US 6 349 023 B1 and US 2020 / 400 355 A1. In particular, document EP 3 107 204 A1 discloses a motor power conversion device equipped with an electronic thermal function for overload protection of a motor, which includes a total loss time integration counter that, when the time integration of a total loss of the motor becomes constant, outputs an integrated value simulating the winding temperature of the motor, and when the integrated value reaches a predetermined threshold, outputs a signal to stop the operation of the motor.

[0009] The present invention therefore aims to overcome the aforementioned disadvantages and to propose an efficient system and method for detecting the engine temperature of an engine, in particular for protecting against engine overheating.

[0010] The basic idea of ​​the invention is to design a sensor in such a way that, based on a thermal model which represents the thermal behavior of the motor, the electrical input power and preferably also the shaft speed as well as a reference temperature, it determines the temperature inside the motor.

[0011] According to the invention, a system for monitoring the temperature of a motor with an electromechanically driven motor shaft is proposed, comprising a "virtual" sensor which, based on a thermal model describing the thermal behavior of the motor, a measuring means for determining the electrical input power of the motor and a measuring means for detecting the shaft speed of the motor shaft, wherein furthermore an evaluation device is provided which is configured to determine the current motor temperature of the motor from the input power, a reference temperature T Ref, the determined shaft speed n and the thermal model.

[0012] In a particularly advantageous embodiment of the invention, the thermal model (MOD) is or comprises a system of differential equations that simulates the thermal behavior of the motor. Thus, by solving the system of differential equations and inputting the required values, the motor temperature can be calculated. In this respect, the sensor based on the system of differential equations can also be considered a virtual sensor.

[0013] The differential equation system used is also preferably characterized by a specific property: it allows for several alternative solution methods for determining the temperature, in particular by means of numerical integration, or by means of an analog calculation method based on an analog computer, or by means of an analog calculation method using a model comprising passive components. Other solution methods would also be conceivable in principle.

[0014] Another aspect of the present invention relates, in addition to the temperature monitoring system, to a method for monitoring the temperature. According to the invention, a method is provided for this purpose, in particular with a system as described above using a system of differential equations (DEs) that describes the thermal behavior of the motor, comprising the following steps: a. Determining the electrical input power of the motor; b. Calculating the mechanical shaft power from the determined rotational speed of the motor and determining the corresponding torque of the motor; c. Calculating the thermal power loss that leads to an increase in the motor temperature, in particular by subtracting the mechanical shaft power from the electrical input power; d. The determined thermal power loss is fed into the system of differential equations as an input for solving the system of differential equations, and a temperature increase is calculated from it, in particular using an additional reference temperature Tref, which is added to the temperature increase to determine the motor temperature.

[0015] In a first advantageous embodiment of the invention, it is provided that the said method is characterized in that the solution of the system of differential equations for determining the motor temperature is carried out by numerical integration (e.g. according to Euler,..., Runge Kutta etc.).

[0016] In two alternative embodiments, the solution to the system of differential equations for determining the motor temperature is calculated analogously, either by using a corresponding circuit similar to an analog computer with appropriate computer components, or alternatively, based on suitable RC components and passive components. The current and / or voltage sources can be operational or transistor circuits when using passive RC systems.

[0017] In the aforementioned embodiments, it is advantageous if the electrical input power of the motor is determined by measuring a DC link voltage and a DC link current and multiplying the two measured values, preferably using a microcomputer, or by deriving it from the DC link voltage. Possible implementations for this will be explained in more detail in the exemplary embodiments described later.

[0018] According to the invention, the rotational speed is determined by measuring the frequency of the phase voltage and determining the corresponding torque from a correspondingly stored characteristic map, or by means of the q-component of the current iq or by means of a strain gauge, particularly in the case of the embodiment in numerical integration.

[0019] According to the invention, an analog frequency-to-voltage conversion is used to generate a voltage proportional to the rotational speed from a measured phase voltage of the motor, in order to determine the shaft power. Alternative methods for determining the shaft power are also conceivable.

[0020] Other advantageous embodiments of the invention are characterized in the dependent claims or are described in more detail below together with the description of the preferred embodiment of the invention with reference to the figures.

[0021] They show: Fig. 1 a schematic representation of an embodiment of the invention; Fig. 2 an exemplary implementation of the calculation method with the software Matlab Simulink (without actual data acquisition) for automatic code generation; Fig. 3 an exemplary implementation of the calculation method with the software Matlab Simulink (with actual data acquisition) for automatic code generation; Fig. 4 an RC chain circuit to illustrate a realization design using differential equations; Fig. 5 a realization approach of a CRCR network with a voltage source; and Fig. 6 a realization approach of a CRCR network with a voltage source.

[0022] The invention is described below using an exemplary embodiment with reference to the Figures 1 to 3 described in more detail, whereby identical reference symbols indicate identical functional and / or structural features.

[0023] Three alternative design variants are presented first, with reference to the Fig. 1described. In the Figure 1 The following procedure steps for determining the engine temperature are shown: Step a: Measurement of the motor's electrical input power; Step b: Calculating the mechanical shaft power from the measured rotational speed of the motor and determining the corresponding torque of the motor; Step c: Calculating the thermal power loss that leads to an increase in motor temperature, in particular by subtracting the mechanical shaft power from the electrical input power and Step d: The determined thermal power loss is fed into the system of differential equations as an input variable for solving the system of differential equations, and a temperature increase is calculated from it, in particular using an additional reference temperature T ref, which is added to the temperature increase in order to determine the motor temperature.

[0024] In a first implementation using numerical integration to solve the differential equations, steps a) to d) are characterized as follows.

[0025] The following is an exemplary implementation concept for an RC chain circuit according to the Fig. 4 In the RC chain circuit, three parallel RC loops are implemented with resistors R1, R2, R3, RL and capacitors C1, C2, C3. Furthermore, the nodes K1, K2, and K3 are specified, at which the currents i1, i2, and i3, or ic1, ic2, and ic3, respectively, divide. The current i1 flows to the first node K1 and then divides into the currents i2 and ic1. From the Figure 4It is evident how the currents are distributed across the loops accordingly. The voltage UR is applied between the terminals. The voltage drop across capacitors C1, C2, and C3 is denoted by UC1, UC2, and UC3, respectively. Ue denotes the input voltage, and Ua denotes the output voltage.

[0026] The following relationships apply: U R = i 1 R 1 + U C 1 K1: i 1 = i c1 + i 2 → (i 1 - i 2 ) • 1 / C1 = dU C1 / dt K2: i 2 = i c2 + i 3 → (i 2 - i 3 ) • 1 / C2 = dU C2 / dt K2: i 3 = i c3 + i 4 → (i 3 - i 4 ) • 1 / C3 = dUc3 / dt

[0027] For i 4, i 4 = 0 (without load) i 4 = U C3 / RL (with load)

[0028] The currents each represent thermal power, and the voltages represent temperatures. The thermal model is thus calculated analogously to the electrical model shown. Therefore, the electrical resistances are analogous to the thermal resistances, and the capacitors represent the heat capacities.

[0029] Regarding the constraints M1: (U e - U C1 ) • 1 / R 1 = i 1 M2: (U C1 - U C2 ) • 1 / R 2 = i 2 M3: (U C2 - U C3 ) • 1 / R 3 = i 3

[0030] When supplied with a current source, i 1 is predetermined or impressed, which makes the equation M1 obsolete (as previously explained in the analogy of the calculation).

[0031] The following relationships then apply to the state space: dU C 1 dt = 1 C 1 ⋅ U e R 1 − U C 1 R 1 − U C 1 R 2 − U C 2 R 2 dU C 2 dt = 1 C 2 ⋅ U C 1 R 2 − U C 2 R 2 − U C 2 R 3 − U C 3 R 3 dU C 3 dt = 1 C 3 ⋅ U C 2 R 3 − U C 3 R 3 − i 4 where i 4 = U C 3 R L dU C 1 dt dU C 2 dt dU C 3 dt = − 1 R 1 C 1 − 1 R 2 C 1 1 R 2 C 1 0 1 R 2 C 2 − 1 R 2 C 2 − 1 R 3 C 2 1 R 3 C 2 0 1 R 3 C 3 − 1 R 3 C 3 − 1 R L C 3 ⋅ U C 1 U C 2 U C 3 + U R 1 R 1 C 1 0 0

[0032] The electrical input power (for supplying the thermal ODE system) can be measured, for example, by measuring the DC link voltage and DC link current (two AD channels) and by multiplying the two values ​​in a microcomputer implemented for this purpose.

[0033] The mechanical shaft power can be calculated from the rotational speed (via measurement of the phase voltage frequency using timers) and the corresponding torque from a stored characteristic map. Alternatively, the torque can be calculated from the current iq (which can be calculated using the rotor position and the three-phase phase current or is provided by the commutation program). Another option is to measure the torque using a strain gauge.

[0034] The mechanical shaft power is subtracted from the electrical input power, and this difference yields the thermal power (i.e., power loss), which leads to the motor heating up. The magnetic power in the input power, which is reversible reactive power, is only present when the phase current (the RMS value) changes and can be included if necessary for frequent speed and load changes, e.g., via a correspondingly dependent constant, characteristic map, or as an input value from a computational circuit for the precise calculation of the power balance. The heating caused by the current (through the winding resistances) to store the magnetic power (reactive power) is, however, automatically taken into account.

[0035] The calculated thermal power (power loss) is fed into the system of differential equations as input, and the temperature increase resulting from this power is calculated. The temperature increase calculated using the differential equation system and the input power is then added to the measured reference temperature to obtain the estimated temperature inside the engine.

[0036] The temperature can then be output, for example, as an analog signal via a DA converter of any design, via a data bus, or as a binary threshold and alarm value on a pin.

[0037] In a second embodiment, analog calculation is used. The system of differential equations is thus calculated analogously by building the corresponding circuit on the printed circuit board (PCB) in the manner of an analog computer. This requires the necessary operational amplifiers, adders, subtractors, constant potentiometers (for multiplication by constants), function generators (for the torque characteristic(s)), and integrators. Steps a) to d) are then carried out with these components as follows: The electrical input power is preferably measured from the DC link using suitable voltage dividers for the DC link voltage and shunts for current measurement, as well as operational amplifiers (op-amps) for signal amplification.The multiplication for calculating the input power is then performed using suitable operational circuits (multipliers) or special (analog or digital) multiplier components (modulation methods or calculations). If the DC link voltage is constant and its stability is monitored, it is sufficient to convert the measured current into the input power with a constant gain.

[0038] Using an analog frequency-to-voltage converter, a voltage proportional to the rotational speed can be generated from the motor's phase voltage. Multiplying this voltage by a voltage proportional to the torque using an analog multiplier or a controlled amplifier (gain control) yields the shaft power.

[0039] Such a voltage proportional to the torque can be obtained, for example, using an analog or digital function generator (the rotational speed being the input variable). The characteristic curve of the fan wheel driven by the motor is stored in the function generator for this purpose.

[0040] Subtracting the mechanical shaft power from the electrical input power again yields the thermal power (power loss) which leads to the heating of the motor. This thermal power is then fed into the system of differential equations as an input variable for calculating the motor temperature, just as in the previously mentioned embodiment.

[0041] If the system of differential equations is supplied with a voltage whose value corresponds to the thermal power dissipation, the system produces the desired temperature increase at its last capacitor. If the system of differential equations is supplied with the thermal power dissipation as the current source, the voltage across the first capacitor produces the desired temperature increase.

[0042] The temperature increase calculated using the analog differential equation system and the input power is then added to the measured reference temperature, yielding the "estimated" temperature inside the motor. The aforementioned adding circuit then provides an analog output voltage proportional to the motor temperature, which can, for example, be easily displayed and / or passed on, or used with a comparator to switch off the motor.

[0043] In a third embodiment of the Figure 1The use of analog calculations is also employed. This utilizes the fact that solutions developed in electronic analog computers address problems by creating direct (electrical) analogies using passive elements such as resistors, capacitors, and inductors. The underlying paths can be modeled with a network of resistors. For example, if an alternating voltage is applied or a pulsed voltage is applied between the input and / or capacitors and inductors and the output, the shortest path can be found by following the path of greatest current flow at each node. The basic principle of the solution relies on current distribution through a resistor network with parallel and series elements. Such systems are also known as analog cellular automata. Using such a modified RC network (resistors and capacitors), it is possible, after calibration (i.e.,Determining the R and C values ​​allows the actual temperature behavior of the motor to be simulated. The cellular automaton used can also be a simple RC chain circuit.

[0044] In this embodiment, steps a) to d) are carried out as follows: The electrical input power is measured using appropriately dimensioned voltage dividers for the DC link voltage and shunts for current measurement, as well as operational amplifiers (op-amps) for signal amplification. The multiplication to calculate the input power is then performed using suitable op-amp circuits (multipliers) or using special (analog or digital) multiplication components (modulation method or calculation).

[0045] The shaft power is determined analogously to the steps described in the second embodiment. Subtracting the mechanical shaft power from the electrical input power yields the thermal power (power loss) that leads to the heating of the motor. Operational amplifier circuits are used for the subtraction and for supplying the RC network with controlled sources, as well as for the other mathematical operations and / or transistor circuits mentioned in this approach. The actual RC network is considered passive here. If the RC network is powered by a voltage source ( Figure 6 If the RC network is powered by a current source whose value corresponds to the thermal power dissipation, the end (voltage across the load resistor or the last capacitor) of the corresponding RC chain delivers the desired temperature increase. However, if the RC network is powered by the thermal power dissipation as a current source ( Figure 5) supplied, the beginning (voltage at the power source or at the first capacitor) of the RC chain delivers the desired temperature increase.

[0046] In the Figure 5 The diagram shows a CRCR network and a power supply with a controlled current source. The control signal of the current source represents the calculated thermal power dissipation. The current source can be designed, for example, with operational amplifiers according to Tietze Schenk.

[0047] In the Figure 6 A CRCR network and a power supply with a controlled voltage source are shown. The control signal of the voltage source represents the calculated thermal power dissipation. The components for R and C are determined using a parameter fit based on the measured values ​​of the actual motor. The voltage source can be designed, for example, according to Tietze Schenk, using operational amplifiers.

[0048] Measuring the respective voltages across the RC network (op-amp circuit) when the RC network is powered with voltage or current sources yields the temperature increase. The measured ambient temperature, in the form of an analog voltage, is then added to this temperature increase (analog voltage).

[0049] The thermal model can be implemented and calculated in any embodiment, depending on the requirements. This concept corresponds to the modular system or solution according to the invention.

[0050] The invention is not limited in its implementation to the preferred embodiments specified above. Rather, a number of variants are conceivable which make use of the solution presented even in fundamentally different designs.

Claims

1. A system (1) for monitoring the temperature of a motor (M) with an electromechanically drivable motor shaft, comprising a sensor (10) that is based on a thermal model (MOD), which describes the thermal behavior of the motor (M), a measuring means for determining the electrical input power (P) of the motor (M), and a measuring means for recording the shaft speed n of the motor shaft, wherein an evaluation device is also provided, which is designed to determine the respectively current motor temperature of the motor (M) from the input power (P), a reference temperature Tref, the determined shaft speed n, and the thermal model (MOD), characterized in that the speed is determinable by measuring the frequency of a phase voltage and a corresponding torque of the motor by means from a correspondingly stored characteristic map determinable, or by means of the q component of the current iq, or by means of a strain gauge, wherein a voltage proportional to the speed is generated from a phase voltage of the motor to be measured with the aid of an analog frequency-voltage conversion in order to determine the shaft power therefrom.

2. The system (1) according to claim 1, characterized in that the thermal model (MOD) comprises a differential equation system (DGL) which simulates the thermal behavior of the motor (M).

3. The system (1) according to claim 2, characterized in that the differential equation system (DGL), as a modular system, enables several alternative solution methods for determining the temperature, in particular by means of numerical integration, by means of an analog calculation method based on an analog computer, or by means of an analog calculation method by means of a model comprising passive components.

4. A method for monitoring the temperature of a motor (M), with a system according to any of claims 1 to 3, using a differential equation system (DGL) which describes the thermal behavior of the motor (M), with the following steps: a. recording the electrical input power of the motor; b. Determining a speed of the motor (M) by measuring the frequency of a phase voltage and determining a corresponding torque of the motor (M), wherein determining the corresponding torque from a correspondingly stored characteristic map, or by means of the q component of the current iq, or by means of a strain gauge c. calculating the mechanical shaft power from the in step b) determined speed of the motor (M), wherein a voltage proportional to the speed is generated from a phase voltage of the motor to be measured with the aid of an analog frequency-voltage conversion in order to determine the shaft power therefrom. d. calculating the thermal power loss that leads to the increase in motor temperature, in particular by subtracting the mechanical shaft power from the electrical input power; e. the determined thermal power loss is fed into the differential equation system (DGL) as an input variable for solving the differential equation system (DGL), and a temperature increase is calculated therefrom, in particular using an additional reference temperature Tref, which is added to the temperature increase in order to determine the motor temperature therefrom.

5. The method according to claim 4, characterized in that the differential equation system (DGL) for determining the motor temperature is solved by numerical integration.

6. The method according to claim 4, characterized in that the solution of the differential equation system (DGL) for determining the motor temperature is calculated analogously by using a corresponding circuit in the manner of an analog computer with corresponding computer modules or on the basis of freely programmable circuits or by means of suitable RC modules and passive components.

7. The method according to claim 4 or 5, characterized in that the electrical input power of the motor is recorded by measuring an intermediate circuit voltage and an intermediate circuit current and by multiplying the two measured values, preferably by means of a microcomputer or programmable circuits.