DE102019214532A1 - Method for determining an insulation resistance - Google Patents

Abstract

The invention relates to a method for determining an insulation resistance between an active conductor and a ground of a high-voltage circuit.

Description

The invention relates to a method for determining an insulation resistance between an active conductor and a ground of a high-voltage circuit.

Such methods are used, for example, in the field of electric vehicles when it comes to testing high-voltage batteries, electrified drive axles or hybrid transmissions with regard to their insulation properties. The test procedures include insulation measurement, insulation monitoring and high-voltage testing. The focus here is on the insulation measurement, in which the insulation resistance is measured between an active conductor and the earth, with or without connected equipment. An insulation measurement is a measurement of the insulation resistance in order to be able to make a qualitative statement about the safety and freedom from defects of the insulation. Such a test of the insulation resistance can be carried out as a type or routine test and as an initial and repeat test. The measurement itself is only a snapshot and has a measurement time of up to one minute. For example, measurement times of approx. 10s are usual for simple cable tests and approx. 60s for testing a battery. The measurement process starts with the contacting of the test object and the build-up of a voltage between approx. 500V to 1000V, depending on the application. After the measurement has been carried out, a discharge takes place via, for example, an 8MOhm resistor or a discharge board.

When measuring the insulation resistance, it is crucial that a final statement about the level of the resistance value can only be made at a certain point in time after the start of the measurement, since current components become effective immediately after the start of the measurement, which decay over time and which are not used to determine the insulation resistance will. Consequently, it is always necessary to wait a certain period of time after the start of the measurement until the irrelevant current components have subsided and only a pure and constant insulation leakage current can be measured. This need to wait until the irrelevant current component has subsided after the start of the measurement appears to be tolerable when it comes to individual measurements, for example in the context of prototype production or small series. In the course of the emerging series production of larger quantities in the field of electric vehicle production or its components, these waiting times during the test must be reduced, especially since not only one test may be required for a manufactured component, for example an electrified axle, but usually several in the course of the Production. It would therefore be desirable and consequently an object if a method for determining an insulation resistance can be provided, for example in the case of the components mentioned, with which one does not have to wait such a long measurement period until the current components irrelevant for the pure insulation current have decayed.

The object is achieved by a method for determining an insulation resistance between an active conductor and a grounding of a high-voltage circuit, in which an approximation polynomial with _{measured values obtained over a measuring period t M} and representing a total current I _{tot is} fed and a constant insulation _leakage current I const over the approximation polynomial is determined in a time period t _{M1 that extends beyond the measurement period t M.}

The basic idea of the method according to the invention is that the total current I _{tot is measured over a measurement period t M that is acceptable from the point of view of series production, and the approximation polynomial} is used to calculate how high the insulation leakage current I const is. Since the constant _{current component I const is the insulation leakage current of} interest, which flows as a result of insulation that has not been carried out to 100%, one would be able to make a statement about its level _{immediately after the end of the measurement period t M and thus a classification} to be carried out in OK or not OK for the component. In practical application, this can mean, for example, that in the case of a battery to be tested for an electric vehicle, the irrelevant current component has subsided to such an extent that it is only possible to make a statement about the level of the insulation leakage current I _const or the insulation leakage current after a period of around 60s. Using the method according to the invention, it is now possible _{to shorten the measurement period t M} to 20 s, for example, and then to make a statement about the level of the insulation _{leakage current} I const immediately.

gilt und

• I_maxC = der maximale kapazitive Strom zu Beginn des Messzeitraums t_M,
• I_maxA = der maximale Absorptionsstrom zu Beginn des Messzeitraums t_M,
• I_konst= der konstante Isolationsleckagestrom, und
• τ_C , τ_A die jeweilige Abklingrate des kapazitiven Stroms und des Absorptionsstroms sind.

An advantageous embodiment of the method provides that the total current I _tot designated by a first as a capacitive current component current I _C, a second, referred to as absorption power current component I _A and the third designated as insulation leakage current current component I is mapped _const, wherein $${\mathrm{I.}}_{Ges}={\mathrm{I.}}_{max\mathrm{C.}}\ast e^{-\frac{t}{\mathrm{\tau }\mathrm{C.}}}+{\mathrm{I.}}_{max\mathrm{A.}}\ast e^{-\frac{t}{\mathrm{\tau {\rm A}}}}+{\mathrm{I.}}_{knOst}$$

applies and

• I _maxC = the maximum capacitive current at the beginning of the measurement period t _M ,
• I _maxA = the maximum absorption current at the beginning of the measurement period t _M ,
• I const = the constant insulation _{leakage current, and}
• τ _C , τ _{A are} the respective decay rate of the capacitive current and the absorption current.

This allows the decaying behavior of the decaying current component to be determined even better, since with detailed modeling using an equivalent circuit it appears to be useful to map the decaying current component using a faster decaying capacitive current and a slower decaying absorption current.

durch das Approximationspolynom: $$P=A\ast e^{{}^{-\frac{t}{\mathrm{\tau {\rm A}}}}}-B\ast e^{{}^{-\frac{t}{\mathrm{\tau {\rm B}}}}}-C\ast e^{-\frac{t}{\mathrm{\tau }C}}$$

ersetzt wird, wobei der Anteil: $$C\ast e^{-\frac{t}{\mathrm{\tau }C}}$$

dem Isolationsleckagestrom I_konst entspricht und τ_C >> ∞ (unendlich) gilt.An advantageous embodiment of the method provides that the formula: $${\mathrm{I.}}_{Ges}={\mathrm{I.}}_{max\mathrm{C.}}\ast e^{-\frac{t}{\mathrm{\tau }\mathrm{C.}}}+{\mathrm{I.}}_{max\mathrm{A.}}\ast e^{-\frac{t}{\mathrm{\tau {\rm A}}}}+{\mathrm{I.}}_{knOst}.$$

by the approximation polynomial: $$P=\mathrm{A.}\ast e^{{}^{-\frac{t}{\mathrm{\tau {\rm A}}}}}-\mathrm{B.}\ast e^{{}^{-\frac{t}{\mathrm{\tau {\rm B}}}}}-\mathrm{C.}\ast e^{-\frac{t}{\mathrm{\tau }\mathrm{C.}}}$$

is replaced, with the proportion: $$\mathrm{C.}\ast e^{-\frac{t}{\mathrm{\tau }\mathrm{C.}}}$$

corresponds to the insulation _{leakage current I const and τ C} >> ∞ (infinite) applies.

der Gesamtstrom I_ges für den Messzeitraum t_M approximiert wird.An advantageous embodiment of the method provides that the approximation polynomial $$P=\mathrm{A.}\ast e^{{}^{-\frac{t}{\mathrm{\tau {\rm A}}}}}-\mathrm{B.}\ast e^{{}^{-\frac{t}{\mathrm{\tau {\rm B}}}}}-\mathrm{C.}\ast e^{-\frac{t}{\mathrm{\tau }\mathrm{C.}}}$$

the total current I _{tot is} approximated for the measurement period t _M.

der Gesamtstrom I_ges für zumindest einen nach dem Messzeitraum t_M liegenden Zeitpunkt t₁ extrapoliert wird.An advantageous embodiment of the method provides that the approximation polynomial $$P=\mathrm{A.}\ast e^{{}^{-\frac{t}{\mathrm{\tau {\rm A}}}}}-\mathrm{B.}\ast e^{{}^{-\frac{t}{\mathrm{\tau {\rm B}}}}}-\mathrm{C.}\ast e^{-\frac{t}{\mathrm{\tau }\mathrm{C.}}}$$

the total current I _{tot is} extrapolated for at least one point in time t ₁ after the measurement period t _M.

An advantageous embodiment of the method provides that the coefficients A, B and C and the time constants τ _A , τ _B , τ _{C of} the approximation polynomial are determined using a deterministic model for approximating the regressive exponential function.

An advantageous embodiment of the method provides that a modified Prony analysis using the least squares method is used as the deterministic model for approximation.

• 1a) Darstellung der Verläufe der Stromanteile während der Isolationsmessung;
• 1b) physikalisches Ersatzschaltbild und
• 2 schematische Darstellung des erfindungsgemäßen Verfahrens.

The invention is explained below with further features, details and advantages on the basis of the accompanying figures. The figures merely illustrate exemplary embodiments of the invention. Show in it

• 1a) Representation of the progression of the current components during the insulation measurement;
• 1b) physical equivalent circuit diagram and
• 2 schematic representation of the method according to the invention.

The 1a) shows the current components considered, which flow according to a model on which the method according to the invention is based, and the 1b) a corresponding equivalent circuit diagram. The 1a) plots the current over time without the units current A and time seconds being related to one another in terms of scale. The three current components, capacitive current component I _C , absorption _{current component I A} , insulation _{leakage current I const} and the total insulation _{current I tot} and their respective course are shown. The insulation _{leakage current} I const corresponds to the proportion of the measured insulation current that is constant over time due to the finite ohmic resistance of the insulation material. The insulation _{leakage current} I const is not constant after the test voltage has been applied, because it only flows when the test voltage has been completely built up, which takes approx. 10 ms. The capacitive current component I _C corresponds to the current generated by charging the capacitance of the insulation material. It is essentially influenced by the mobility of the charge carriers - generally electrons. The dielectric absorption _{current component I A} corresponds to the current that is generated by the formation, alignment and migration of polar molecules in the insulator. Due to the low mobility of the molecules, the absorption current component generally decays more slowly than the capacitive current component I _C.

Die 2 zeigt das erfindungsgemäße Verfahren in einem schematisierten Ablauf. In Verfahrensschritt 10 werden über einen Messzeitraum t_M, der typischerweise im Bereich von wenigen bis mehrere Sekunden liegt, einen Isolationsstromverlauf I_ges repräsentierende Messwerte gewonnen. Anschließend wird in Verfahrensschritt 12 ein Approximationspolynom gebildet, dass in seinem strukturellen Aufbau der physikalischen Gleichung des gesamten Isolationsstrom I_ges entspricht. Hierbei ist es zweckmäßig, wenn das Approximationspolynom anhand einer regressiven Exponentialfunktion abgebildet wird, die folgende allgemeine Form aufweisen kann: $$P=A\ast e^{{}^{-\frac{t}{\mathrm{\tau {\rm A}}}}}-B\ast e^{{}^{-\frac{t}{\mathrm{\tau {\rm B}}}}}-C\ast e^{-\frac{t}{\mathrm{\tau }C}}$$

In dem Verfahrensschritt 14 werden über die regressive Exponentialfunktion die Stromanteile kapazitiver Stromanteil I_C, Absorptionsstromanteil I_A und Isolationsleckagestrom I_konst berechnet. Hierüber ist es möglich für jeden Zeitpunkt innerhalb des Messzeitraum t_M, aber auch für einen Zeitpunkt t₁ nach dem Messzeitraum t_M die beiden abklingenden Stromanteile I_C und I_A zu berechnen, so dass in einfacher Weise für jeden gewünschten Zeitpunkt die Höhe des Isolationsleckagestrom I_konst ermittelt werden kann und damit eine Einordnung in i.O. oder n.i.O. für das Bauteil vorgenommen werden kann.The 1b) shows an equivalent circuit diagram of the conductor to be tested under the assumption that the three current components, capacitive current component I _C , absorption _{current component I A} , and insulation _{leakage current I const become} effective. It would also be conceivable to model or describe the current flow during the test of the insulation current over more than three individual current components. After the test voltage has reached the intended value U _soll , a current component flows into each branch, which results _{in the total insulation current I tot.} The total insulation current I _tot can be represented using the following physical equation: $${\mathrm{I.}}_{Ges}={\mathrm{I.}}_{max\mathrm{C.}}\ast e^{-\frac{t}{\mathrm{\tau }\mathrm{C.}}}+{\mathrm{I.}}_{max\mathrm{A.}}\ast e^{-\frac{t}{\mathrm{\tau {\rm A}}}}+{\mathrm{I.}}_{knOst}$$

The 2 shows the method according to the invention in a schematic sequence. In method step 10, measured values representing _{an insulation current profile} I tot are obtained over a measuring period t _M , which is typically in the range from a few to several seconds. Subsequently, in method step 12, an approximation polynomial is formed that, in its structural design, corresponds to _{the physical equation of the total insulation current I tot.} It is useful here if the approximation polynomial is mapped using a regressive exponential function, which can have the following general form: $$P=\mathrm{A.}\ast e^{{}^{-\frac{t}{\mathrm{\tau {\rm A}}}}}-\mathrm{B.}\ast e^{{}^{-\frac{t}{\mathrm{\tau {\rm B}}}}}-\mathrm{C.}\ast e^{-\frac{t}{\mathrm{\tau }\mathrm{C.}}}$$

In method step 14, the current components of capacitive current component I _C , absorption _{current component I A} and insulation _{leakage current I const are} calculated using the regressive exponential function. This makes it possible to calculate the two decaying current components I _C and I _A for each point in time within the measurement period t _M , but also for a point in time t ₁ after the measurement period t _M , so that the level of the insulation leakage current can easily be calculated for each desired point in time I _const can be determined and thus a classification in OK or not OK for the component can be made.

Claims (6)

Method for determining an insulation resistance between an active conductor and a grounding of a high-voltage circuit, in which an approximation polynomial with _{measured values obtained over a measurement period t M} and representing a total current I _{tot is} fed and the constant leakage current I _const via the approximation polynomial in a via the Measurement period t _M beyond period t _{1 is} determined. Procedure according to Claim 1 Wherein the total current I _tot is mapped _const via a first capacitive current as designated current component I _C, a second, referred to as absorption power current component I _A and a third, referred to as leakage current component I, wherein $${\mathrm{I.}}_{Ges}={\mathrm{I.}}_{max\mathrm{C.}}\ast e^{-\frac{t}{\mathrm{\tau }\mathrm{C.}}}+{\mathrm{I.}}_{max\mathrm{A.}}\ast e^{-\frac{t}{\mathrm{\tau {\rm A}}}}+{\mathrm{I.}}_{knOst}$$

applies and I _maxC = the maximum capacitive current at the beginning of the measurement period t _M , I _maxA = the maximum absorption _{current at the beginning of the measurement period t M} , I _const = the constant leakage current, and τ _C , τ _A the respective decay rate of the capacitive current and the Absorption stream are. Procedure according to Claim 2 where the formula: $${\mathrm{I.}}_{Ges}={\mathrm{I.}}_{max\mathrm{C.}}\ast e^{-\frac{t}{\mathrm{\tau }\mathrm{C.}}}+{\mathrm{I.}}_{max\mathrm{A.}}\ast e^{-\frac{t}{\mathrm{\tau {\rm A}}}}+{\mathrm{I.}}_{knOst}.$$

by the approximation polynomial $$P=\mathrm{A.}\ast e^{{}^{-\frac{t}{\mathrm{\tau {\rm A}}}}}-\mathrm{B.}\ast e^{{}^{-\frac{t}{\mathrm{\tau {\rm B}}}}}-\mathrm{C.}\ast e^{-\frac{t}{\mathrm{\tau }\mathrm{C.}}},$$

is replaced where the portion: $$\mathrm{C.}\ast e^{-\frac{t}{\mathrm{\tau }\mathrm{C.}}}$$

the leakage current I _const corresponds to (τ _C -> ∞), the total current I t _ges for the measuring period _M is approximated. Procedure according to Claim 3 , for which the approximation polynomial $$P=\mathrm{A.}\ast e^{{}^{-\frac{t}{\mathrm{\tau {\rm A}}}}}-\mathrm{B.}\ast e^{{}^{-\frac{t}{\mathrm{\tau {\rm B}}}}}-\mathrm{C.}\ast e^{-\frac{t}{\mathrm{\tau }\mathrm{C.}}}$$

the total current I _ges for at least one measurement period t _M after the earliest time t is extrapolated. ₁ Method according to one of the Claims 3 or 4th , the coefficients A, B and C as well as the time constants τ _A, τ _B , τ _{C of} the approximation polynomial are determined using a deterministic model for approximating the regressive exponential function. Procedure according to Claim 6 , in which a modified Prony analysis based on the least squares method is used as a deterministic model for approximation.

Images