DE102018208618A1 - Method for determining the temperature of an electrical / electronic component, circuit - Google Patents

Abstract

The invention relates to a method for determining the temperature of at least one electrical / electronic heat-generating component (2,6) of an electrical circuit (1), in particular control unit or power electronics, which has at least one temperature sensor (3) and one or more between the temperature sensor (3). and the component (2,6) arranged, heat from the component (2,6) to the temperature sensor (3) conductive elements (5), comprising the following steps: a) detecting a temperature (Tmittels of the temperature sensor (3), b ) Determining the power loss (P) of the component (2,6), c) determining the heat capacity (C) of each element (5) arranged between the component (2,6) and the temperature sensor (3), d) determining heat transfer resistances ( R), each acting between the adjacent elements (5), the component (2,6) and the temperature sensor (3), e) determining the temperature of the component (2,6) in dependence on the determined power loss, heat transfer resistances (R), heat capacities (C) and the detected temperature (T).

Description

The invention relates to a method for determining the temperature of at least one electrical / electronic heat-generating component of an electrical circuit, in particular control unit or power electronics, which at least one temperature sensor and one or more arranged between the temperature sensor and the component, heat from the component to the temperature sensor conductive Has elements.

Furthermore, the invention relates to a circuit having at least one electrical / electronic component, whose temperature is to be monitored, with a remote from the component temperature sensor and one or more elements which are arranged between the component and the temperature sensor and heat from the component to to the temperature sensor, and to a control unit configured to operate the circuit and to monitor the temperature of the component.

State of the art

Methods and circuits of the type mentioned are known from the prior art. Heating an electrical / electronic component under load can be a problem for many applications. Thus, the heating on the one hand lead to damage of the component itself or to the damage of adjacent components or distort, for example, measurement results. It is therefore important to be able to predict the thermal load resulting from heating with sufficient accuracy. Due to the difficult feasibility to design or provide a measuring point on each component, as well as due to the desire to save additional temperature sensors, it is known to form on the component itself no temperature sensor / provide. However, it is known to estimate the thermal behavior of the component using the available temperature information by means of a separate temperature sensor, which is arranged in particular away from the component. For this purpose, a physically plausible model is usually constructed, which simulates the heat transfer from the component, which represents a heat source and the temperature sensor. On the basis of the detected temperature information in the vicinity of the component and, for example, the calculated power loss of the component, a sufficiently accurate temperature behavior of the component can be estimated using the model.

Disclosure of the invention

The method according to the invention has the advantage that the temperature of the component can advantageously be determined precisely, and that, in addition, it is possible to precisely determine the component temperature in the presence of a plurality of heat sources in the circuit. The inventive method provides for this purpose that in a first step a) a temperature is detected by means of the temperature sensor. In a second step b), the power loss of a heat-generating component is determined or calculated (eg by currents and resistances to be detected (I ² R)). In addition, the heat capacity of each element arranged between the component and the temperature sensor is determined in a step c). In addition, in a step d), the heat transfer resistances which each act between the adjacent elements, the component and the temperature sensor are determined. In this case, the heat transfer resistances are respectively determined between two adjacent units, that is to say between two adjacent elements, one element and the component adjacent thereto, or the element at the adjacent temperature sensor. In a step e), the temperature of the component is determined as a function of the determined heat transfer resistances, the determined heat capacities, the determined power loss and the detected temperature. This results in an accurate temperature detection with only a small measurement error, which allows an advantageous estimation and monitoring of the temperature of the component.

In particular, the heat capacities of the elements are each calculated and / or determined by experiments. Thus, heat capacities can already be taken into account in the design of the elements and / or by experiments, even before a shunt in the circuit, determined and stored as values in a memory, for example in the form of a map and / or the characteristic.

Accordingly, the heat transfer levels are preferably also calculated and / or determined by experiments. By storing the determined values in a characteristic curve and / or a characteristic map, a rapid estimation of the temperature of the component by means of the heat capacities and the heat transfer resistances is ensured, for example, by a control unit or a control unit.

Furthermore, it is preferably provided that the temperature of the component in step e) is determined by means of the explicit Euler method. This maps a temperature model that is performed by the control unit in a discrete manner. The heat transfer resistances between the elements, the component and the temperature sensor and the respective heat capacities are taken into account and the temperature of the component is determined in a simple manner.

If there are several components in the circuit that generate heat during operation, the associated method is carried out for each of the components, so that a temperature value is determined or estimated for each of the components. For example, in the above-described circuit, when an element interposed between the component and the temperature sensor also generates heat, it also applies as a component for which the method for determining its temperature is performed using the temperature value detected by the temperature sensor.

In determining the temperature of a plurality of heat-generating components, at least one difference temperature between the respective component and the temperature sensor and between the respective component and an element lying between the temperature sensor and the component are preferably taken into account. By taking into account the differential temperatures, the thermal behavior of the individual elements is matched with the temperature model, so that an accurate estimate of the temperature of the component to be monitored is ensured even in the presence of multiple heat sources in the circuit.

The respective differential temperature is preferably calculated on the basis of the capacitance, the resistance and the power loss, in particular with the aid of equations 7, 8, 9 and 10 described below.

Particularly preferably, the determined differential temperatures for the components for the determination of the temperature of the respective component are correlated. In particular, the resulting differential temperatures are summed up for the determination of the thermal behavior.

The circuit according to the invention with the features of claim 9 is characterized in that the control unit is specially adapted to carry out the method according to the invention. This results in the already mentioned advantages.

In particular, the circuit is designed as power electronics for an electric machine, in particular drive machine, preferably media flow machine of a compressor of an internal combustion engine for a motor vehicle.

• 1 ein Ersatzschaltbild zur Erläuterung eines vorteilhaften Verfahrens zum Bestimmen der Bauteiltemperatur in einer Schaltung,
• 2 ein weiteres Schaltbild zur Erläuterung des Verfahrens und
• 3 ein weiteres Schaltbild zur Erläuterung des Verfahrens.

In the following, the invention will be explained in more detail with reference to the drawing. Show this

• 1 an equivalent circuit diagram for explaining an advantageous method for determining the component temperature in a circuit,
• 2 another diagram for explaining the method and
• 3 another circuit diagram for explaining the method.

1 shows a simplified circuit diagram of a multi-element electrical / electronic circuit 1 , At least one of the elements is used as a heat generating electrical / electronic component 2 educated. The circuit 1 also has a temperature sensor 3 , by means of which help a current temperature of the component 2 is to be recorded during operation. This is also a control unit 4 present with the temperature sensor 3 is technically connected. Because the temperature sensor 3 not directly on the component 2 is arranged, its temperature can not be detected, but only estimated or calculated. This is the in 1 shown model in the form of an equivalent circuit diagram, which is used to determine the temperature of the component 2 to determine. It is assumed that between the component 2 and the temperature sensor 3 one or more other elements 5 are arranged, which the heat transfer from the component 2 to the temperature sensor 3 can influence.

Creating the temperature model in the present case means the modeling of the thermal behavior through the mathematical description of the thermal conductivity and heat capacity between the component 2 at which a power loss pv is generated, and the temperature sensor 3 , 1 shows the thermal behavior for n-members, ie for a total number of n-elements 5 of which at least one is a heat-generating component 2 is.

The component 2 is connected by heat conduction in connection with the other elements 5 up to the temperature sensor 3 , The heat capacity of the respective element 5 is in the equivalent circuit diagram with C indicated, and the heat transfer resistance with R , The more links there are, the more accurately the thermal behavior is described. In the present case, the component 2 a heat capacity C1 and a heat transfer resistance R1 to the next adjacent element 5 on. This in turn has a heat capacity C2 and a heat transfer resistance R2 on.

$$T_2=T_3+\mathrm{\Delta }{T}_2$$

$$P_v=C_1\times \frac{d\mathrm{\Delta }{T}_2}{dt}+\frac{\left(\mathrm{\Delta }{T}_1-\mathrm{\Delta }{T}_2\right)}{R_1}$$

$$\frac{\left(\mathrm{\Delta }{T}_1-\mathrm{\Delta }{T}_2\right)}{R_1}=C_2\times \frac{d\mathrm{\Delta }{T}_2}{dt}+\frac{\mathrm{\Delta }{T}_2}{R_2}$$

The following is the mathematical description for the two-limb system (2-mass model). For a system with additional links, the description can be derived in a similar manner. For the 2-limb model is that of the temperature sensor 3 recorded temperature T ₃ known. Depending on the power loss pv on the component 2 is the thermal load of the component 2 (first element) and the element 5 to determine (second element). The mathematical description is as follows: $$T_1={\mathrm{<figure-callout\; id="3"\; label="T"\; filenames="DE102018208618A1\_0018.png"\; state="\{\{state\}\}">T</figure-callout>}}_3+\mathrm{\Delta }{T}_1$$

$${\mathrm{<figure-callout\; id="2"\; label="T"\; filenames="DE102018208618A1\_0018.png,DE102018208618A1\_0019.png"\; state="\{\{state\}\}">T</figure-callout>}}_2={\mathrm{<figure-callout\; id="3"\; label="T"\; filenames="DE102018208618A1\_0018.png"\; state="\{\{state\}\}">T</figure-callout>}}_3+\mathrm{<figure-callout\; id="2"\; label="\Delta \; T"\; filenames="DE102018208618A1\_0018.png,DE102018208618A1\_0019.png"\; state="\{\{state\}\}">\Delta \; </figure-callout>}{T}_{}{}_2$$

$$P_v=C_1\times \frac{d\mathrm{\Delta }{T}_2}{dt}+\frac{\left(\mathrm{\Delta }{T}_1-\mathrm{\Delta }{T}_2\right)}{R_1}$$

$$\frac{\left(\mathrm{\Delta }{T}_1-\mathrm{\Delta }{T}_2\right)}{R_1}={\mathrm{<figure-callout\; id="2"\; label="C"\; filenames="DE102018208618A1\_0018.png,DE102018208618A1\_0019.png"\; state="\{\{state\}\}">C</figure-callout>}}_2\times \frac{d\mathrm{\Delta }{T}_2}{dt}+\frac{\mathrm{\Delta }{T}_2}{{\mathrm{<figure-callout\; id="2"\; label="R"\; filenames="DE102018208618A1\_0018.png,DE102018208618A1\_0019.png"\; state="\{\{state\}\}">R</figure-callout>}}_2}$$

• P_v - Verlustleistung am Bauteil
• $C_1\times \frac{d\mathrm{\Delta }{T}_2}{dt}$
  
  - Wärmekapazität am Bauteil
• $\frac{\left(\mathrm{\Delta }{T}_1-\mathrm{\Delta }{T}_2\right)}{R_1}$
  
  - Wärmeleitung zum Element 5
• ${\mathrm{<figure-callout\; id="2"\; label="C"\; filenames="DE102018208618A1\_0018.png,DE102018208618A1\_0019.png"\; state="\{\{state\}\}">C</figure-callout>}}_2\times \frac{d\mathrm{\Delta }{T}_2}{dt}$
  
  - Wärmekapazität am Element 5
• $\frac{\mathrm{\Delta }{T}_2}{{\mathrm{<figure-callout\; id="2"\; label="R"\; filenames="DE102018208618A1\_0018.png,DE102018208618A1\_0019.png"\; state="\{\{state\}\}">R</figure-callout>}}_2}$
  
  - Wärmeleitung zum Temperatursensor 3

Here are:

• P _v - Power loss on the component
• $C_1\times \frac{d\mathrm{\Delta }{T}_2}{dt}$
  
  - Heat capacity on the component
• $\frac{\left(\mathrm{\Delta }{T}_1-\mathrm{\Delta }{T}_2\right)}{R_1}$
  
  - Heat conduction to the element 5
• $C_2\times \frac{d\mathrm{\Delta }{T}_2}{dt}$
  
  - Heat capacity on element 5
• $\frac{\mathrm{\Delta }{T}_2}{R_2}$
  
  - Heat conduction to the temperature sensor 3

$$\mathrm{\Delta }{T}_2\left(n+1\right)=\frac{T_s}{\left(c_2{R}_1\right)}\mathrm{\Delta }{T}_1\left(n\right)+\left(1-\frac{T_s}{\left(c_2{R}_1\right)}-\frac{T_s}{\left(c_2{R}_2\right)}\right)\mathrm{\Delta }{T}_2\left(n\right)$$

Because the temperature model in the control unit 4 preferably in a discrete manner, the description is to be discredited by Euler's forward method (explicit Euler method) as follows: $$\mathrm{\Delta }{T}_1\left(n+1\right)=\left(1-\frac{T_s}{\left(c_1{R}_1\right)}\right)\mathrm{\Delta }{T}_1\left(n\right)+\frac{T_s}{\left(c_1{R}_1\right)}\mathrm{\Delta }{T}_2\left(n\right)+\frac{T_s}{C_1}{P}_v$$

$$\mathrm{\Delta }{T}_2\left(n+1\right)=\frac{T_s}{\left(c_2{R}_1\right)}\mathrm{\Delta }{T}_1\left(n\right)+\left(1-\frac{T_s}{\left(c_2{R}_1\right)}-\frac{T_s}{\left(c_2{R}_2\right)}\right)\mathrm{\Delta }{T}_2\left(n\right)$$

Is next to the component 2 yet another element of the elements 5 designed to generate heat during operation so that it serves as a heat source, a temperature model with at least n-members is constructed for each heat source of the circuit as described above. Each heat source acts separately a differential temperature .DELTA.T between the existing temperature information of the temperature sensor 3 and the elements 5 out. That is, a first heat source, in this case the component 2 , causes a differential temperature ΔT ₁₁ on the component 2 , and a differential temperature ΔT ₁₂ on the one element 5 and so on. The second heat source, in the present case of the second of the elements 5 formed and as another component 6 designates, causes a differential temperature ΔT ₂₁ on the component 2 or the first element 5 and a differential temperature ΔT ₂₂ on the second element 5 or the component 6 , The differential temperatures resulting from different heat sources for each individual element must be correlated for the determination of the thermal behavior of the individual element. In the following, a system with two heat sources through the temperature models with two members will be described.

2 shows the appropriate equivalent circuit diagram for the first heat source or the component 2 , It is ΔT ₁₁ through the component 2 caused difference temperature between that of the temperature sensor 3 recorded temperature T ₃ and the first element or component 2 , ΔT ₁₂ put that through the component 2 caused difference temperature between the sensor 3 recorded temperature T ₃ and the second element 5 or the component 6 represents.

$$\mathrm{\Delta }{T}_{12}\left(n+1\right)=\frac{T_s}{\left(C_{12}{R}_{11}\right)}\mathrm{\Delta }{T}_{11}\left(n\right)+\left(1-\frac{T_s}{\left(C_{12}{R}_{11}\right)}-\frac{T_s}{\left(C_{12}{R}_{12}\right)}\right)\mathrm{\Delta }{T}_{12}\left(n\right)$$

The difference temperatures ΔT ₁₁ and ΔT ₁₂ are to be described with the following equations: $$\mathrm{\Delta }{T}_{11}\left(n+1\right)=\left(1-\frac{T_s}{C_{11}{R}_{11}}\right)\mathrm{\Delta }{T}_{11}\left(n\right)+\frac{T_s}{\left(C_{11}{R}_{11}\right)}\mathrm{\Delta }{T}_{12}\left(n\right)+\frac{T_s}{C_{11}}{P}_{v1}$$

$$\mathrm{\Delta }{T}_{12}\left(n+1\right)=\frac{T_s}{\left(C_{12}{R}_{11}\right)}\mathrm{\Delta }{T}_{11}\left(n\right)+\left(1-\frac{T_s}{\left(C_{12}{R}_{11}\right)}-\frac{T_s}{\left(C_{12}{R}_{12}\right)}\right)\mathrm{\Delta }{T}_{12}\left(n\right)$$

$$\mathrm{\Delta }{T}_{22}\left(n+1\right)=\frac{T_s}{\left(C_{22}{R}_{21}\right)}\mathrm{\Delta }{T}_{21}\left(n\right)+\left(1-\frac{T_s}{\left(C_{22}{R}_{21}\right)}-\frac{T_s}{\left(C_{22}{R}_{22}\right)}\right)\mathrm{\Delta }{T}_{22}\left(n\right)$$

For the second heat source or the component 6 is the equivalent circuit in 3 shown. It is ΔT ₂₁ the difference temperature between temperature information caused by the second heat source T ₃ and the second element 5 or the component 6 and ΔT ₂₂ the difference temperature between the temperature information caused by the second heat source T ₃ and the first element 5 or the component 2 , ΔT ₂₁ and ΔT ₂₂ are to be described with the following equations: $$\mathrm{\Delta }{T}_{12}\left(n+1\right)=\left(1-\frac{T_s}{C_{21}{R}_{21}}\right)\mathrm{\Delta }{T}_{21}\left(n\right)+\frac{T_s}{\left(C_{21}{R}_{21}\right)}\mathrm{\Delta }{T}_{22}\left(n\right)+\frac{T_s}{C_{21}}{\mathrm{<figure-callout\; id="2"\; label="P\; v"\; filenames="DE102018208618A1\_0018.png,DE102018208618A1\_0019.png"\; state="\{\{state\}\}">P\; </figure-callout>}}_v$$

$$\mathrm{\Delta }{T}_{22}\left(n+1\right)=\frac{T_s}{\left(C_{22}{R}_{21}\right)}\mathrm{\Delta }{T}_{21}\left(n\right)+\left(1-\frac{T_s}{\left(C_{22}{R}_{21}\right)}-\frac{T_s}{\left(C_{22}{R}_{22}\right)}\right)\mathrm{\Delta }{T}_{22}\left(n\right)$$

$$T_2=T_3+\mathrm{\Delta }{T}_{12}\left(n+1\right)+\mathrm{\Delta }{T}_{21}\left(n+1\right)$$

The resulting temperatures for the individual elements must be correlated. Since the action of each heat source is independent of each other, a simple summation of the differential temperatures resulting from different heat sources is provided for the determination of the thermal behavior of the single element as follows: $$T_1={\mathrm{<figure-callout\; id="3"\; label="T"\; filenames="DE102018208618A1\_0018.png"\; state="\{\{state\}\}">T</figure-callout>}}_3+\mathrm{\Delta }{T}_{11}\left(n+1\right)+\mathrm{\Delta }{T}_{22}\left(n+1\right)$$

$${\mathrm{<figure-callout\; id="2"\; label="T"\; filenames="DE102018208618A1\_0018.png,DE102018208618A1\_0019.png"\; state="\{\{state\}\}">T</figure-callout>}}_2={\mathrm{<figure-callout\; id="3"\; label="T"\; filenames="DE102018208618A1\_0018.png"\; state="\{\{state\}\}">T</figure-callout>}}_3+\mathrm{\Delta }{T}_{12}\left(n+1\right)+\mathrm{\Delta }{T}_{21}\left(n+1\right)$$

The initial conditions are defined as follows: $$\mathrm{\Delta }{T}_{11}\left(0\right)=\mathrm{\Delta }{T}_{12}\left(0\right)=\mathrm{\Delta }{T}_{21}\left(0\right)=\mathrm{\Delta }{T}_{22}\left(0\right)=0$$

This results in an advantageous determination of the temperature of the respective component 2 and 6 , where the heat transfer resistances R and capacities C advantageously previously determined and / or calculated, and in the control unit 4 be stored in a non-volatile memory, so that a fast implementation of the method is guaranteed. Advantageously, the method finds application in a control unit and / or power electronics for an electrical machine, in particular a compressor of a motor vehicle.

Claims (10)

Method for determining the temperature of at least one electrical / electronic heat-generating component (2,6) of an electrical circuit (1), in particular control unit or power electronics, which has at least one temperature sensor (3) and one or more between the temperature sensor (3) and the component ( 2.6), heat from the component (2,6) to the temperature sensor (3) conductive elements (5), comprising the following steps: a) detecting a temperature (T ₃ ) by means of the temperature sensor (3), b) Determining the power loss (P _v ) of the component (2,6), c) determining the thermal capacity (C) of each between the component (2,6) and the temperature sensor (3) arranged element (5), d) determining heat transfer resistances ( R), each acting between the adjacent elements (5), the component (2,6) and the temperature sensor (3), e) determining the temperature of the component (2,6) in dependence on the determined power loss, heat transfer resistances (R), heat capacities (C) and the detected temperature (T ₃ ). Method according to Claim 1 , characterized in that the heat capacities (C) are respectively calculated and / or determined by experiments. Method according to one of the preceding claims, characterized in that the heat transfer resistances (R) are respectively calculated and / or determined by tests. Method according to one of the preceding claims, characterized in that the temperature (T ₁ ) of the component (2,6) in step d) is determined by means of the explicit Euler method. Method according to one of the preceding claims, characterized in that in the presence of a plurality of components (2,6) generating heat, the method is carried out for each of the components (2,6). Method according to one of the preceding claims, characterized in that in determining the temperature of a plurality of heat-generating components (2,6) at least one differential temperature (.DELTA.T) between the respective component (2,6) and the temperature sensor (3) and between the respective component (2,6) and a respectively between the component (2,6) and the temperature sensor (3) lying element (5) of the circuit (1) is taken into account. Method according to Claim 6 , characterized in that the respective differential temperature (.DELTA.T) is calculated. Method according to one of Claims 6 or 7 , characterized in that the differential temperatures (ΔT) for determining the temperature of the respective component (2,6) are correlated. Circuit (1) with at least one electrical / electronic component (2,6) whose temperature is to be monitored, with a temperature sensor (3) arranged away from the component (2,6) and with one or more elements (5) are arranged between the component (2,6) and the temperature sensor (3) and conduct heat from the component (2,6) to the temperature sensor (3), and with a control unit (4) which is adapted to the circuit ( 1) and to monitor the temperature of the component (2, 6), characterized in that the control unit (4) is specially adapted to perform the method according to any one of Claims 1 to 8th perform. Switching to Claim 9 , characterized in that the circuit (1) is designed as power electronics or control unit (4) for an electrical machine, in particular drive machine, preferably media flow machine, a compressor of an internal combustion engine for a motor vehicle.

Images