Method for detecting electrical insulation faults between a power source and electrical ground - Patent Application 20070122997

Abstract

The present invention relates to a method for detecting an electrical insulation fault between a power source (20) and an electrical ground (30) by means of a circuit (40) including a controllable voltage generator and an electrical measuring resistor connected in series between the terminals of the power source and electrical ground, and further including means for measuring the voltage at the terminals of the electrical measuring resistor, the method comprising the steps of: - controlling the voltage generator so that the voltage generator establishes a non-zero voltage value across its terminals; and - measuring the voltage at the terminals of the electrical measuring resistor in order to detect an electrical insulation fault between the power source and electrical ground. According to the present invention, the method further comprises a checking step, during which a computer: - determines at least two parameters specific to disturbances affecting the detection of the electrical insulation fault depending on at least the voltage measured at the terminals of the electrical measuring resistor; and - calculates a level of confidence in the detection of the electrical insulation fault depending on at least one of the two parameters.

Description

[Technical Field]

[0001] The present invention relates generally to the electrical safety of power sources such as batteries.

[0002] The present invention relates inter alia to a device for detecting electrical insulation faults between a power source and an electrical ground, the device comprising: - a controllable voltage generator and an electrical measuring resistor connected in series between the terminals of the power supply and electrical ground; - means for measuring the intensity of the current circulating through an electrical measuring resistor; Includes:

[0003] The present invention relates primarily to a method for detecting electrical insulation faults, the method comprising: - controlling the voltage generator to establish a non-zero voltage across the terminals of the voltage generator; - measuring the voltage at the terminals of the electrical measuring resistor to detect electrical insulation faults between the power source and the electrical ground.

[0004] The invention is particularly advantageously applicable to electric motor vehicles. [Background technology]

[0005] An electric motor vehicle typically includes an electric motor and a battery that is specially designed to provide electrical current to the electric motor to propel the vehicle forward.

[0006] Such batteries typically include multiple low-voltage cells connected in series so that the total voltage at the terminals of the battery is high enough to propel the vehicle forward, often around 400V.

[0007] Such batteries are usually controlled by a management system designed to monitor the correct operation of the battery.

[0008] Equipping such control systems with the ability to detect electrical insulation faults is known practice, for example from document EP 0 654 673. Indeed, in the event of a short circuit between the accumulator battery and the vehicle body, it would be dangerous to allow the situation to persist.

[0009] To that end, document EP 0 654 673 teaches measuring the insulation resistance between the accumulator battery and the vehicle body using a detection circuit comprising a controllable voltage generator and an electrical measuring resistor connected in series between the terminals of the accumulator battery and the body.

[0010] It is explained that by generating two voltage pulses, it is possible to measure two voltage values at the terminals of an electrical measuring resistor and from there deduce the value of the insulation resistance.

[0011] It is also explained that these voltage measurements must be performed in steady state, and to that end, the document proposes a method that makes it possible to determine when the transient state has ended so that the desired measurements can be performed as soon as possible.

[0012] This document further teaches that errors may affect the measurements made. To prevent this type of error, this document suggests calculating the arithmetic mean of successively measured voltages and not using the results of measurements if this mean changes excessively over time.

[0013] One drawback of this solution is that it may in fact falsely detect measurement errors, resulting in long periods during which the measurement results are unusable, which may be potentially dangerous in the event of a short circuit.

[0014] Another drawback of this solution is that when an electrical insulation fault is discovered, it is not possible to know exactly between which accumulator cells of the battery the fault is located. Summary of the Invention

[0015] In order to remedy the above-mentioned drawbacks of the prior art, the present invention proposes a detection method such as defined in the opening paragraph, the method further comprising a checking step, during which the computer: - determining at least two parameters specific to disturbances affecting the detection of electrical insulation faults as a function of the voltage measured at least at the terminals of the electrical measuring resistor; - Calculating the level of reliability of the detection of electrical insulation faults as a function of at least one of two parameters.

[0016] The present invention therefore proposes to use two parameters as the basis for determining whether a situation exists which would prevent the detection of an insulation fault.

[0017] Both of these parameters are functions of the voltage measured at the terminals of the measuring resistor.

[0018] In fact, the applicant has discovered that in situations where electrical insulation faults cannot be reliably detected, characteristic signals appear at the terminals of the voltage source (battery), which can be modeled, for example, in the form of voltage pulses.

[0019] The appearance of such a characteristic signal can also be detected at the terminals of the measuring resistor. Using the two above parameters, it is then possible to reliably characterize the appearance of such a signal, effectively checking whether there is a risk of falsely detecting an insulation fault.

[0020] Other advantageous, non-limiting features of the detection method according to the invention, considered individually or in all technically possible combinations, are the following: - between the measuring step and the checking step, a step of detecting electrical insulation faults is provided; - the result of said detection step is used or not used depending on the calculated level of confidence, - after the checking step, the step of detecting an electrical insulation fault is carried out only if the level of confidence indicates that the detection of the electrical insulation would be reliable, - to detect electrical insulation faults, the computer calculates a value of the insulation resistance of the power source with respect to electrical ground and then compares said value with a threshold value; - if an electrical insulation fault is detected, the computer calculates a coefficient of location of said electrical insulation fault in said power supply; - in the checking step, the calculation of the level of confidence is carried out as a function of two parameters, - in the checking step, the calculation of the level of reliability is performed as a function of the variation over time of at least one of the two parameters, - if the voltage generator is controlled so that the voltage assumes two non-zero, mutually distinct values during two successive periods, two values of the voltage at the terminals of the electrical measuring resistor are measured during two periods, respectively, and each parameter is determined as a function of the two measurements; - one of the parameters is a function of the arithmetic mean between two measurements of the voltage at the terminals of the electrical measuring resistor, - The other parameter is a function of the deviation between two measurements of the voltage at the terminals of the electrical measuring resistor.

[0021] The invention also relates to a device for detecting electrical insulation faults between a power source and electrical ground, which device comprises, on the one hand, a circuit including a controllable voltage generator and an electrical measuring resistor connected in series between the terminals of the power source and electrical ground, and including means for measuring the intensity of the current circulating through the electrical measuring resistor, and, on the other hand, a computer suitable for implementing the detection method specified above.

[0022] The present invention also relates to a motor vehicle including the power supply, electrical ground, and detection devices specified above.

[0023] Obviously, the different features, variants and embodiments of the invention can be associated with one another according to various combinations, unless they are mutually exclusive or mutually exclusive.

[0024] The following description, taken in light of the accompanying drawings given as non-limiting examples, will give a better understanding of what the invention consists of and how it can be produced. [Brief explanation of the drawings]

[0025] [Figure 1] 1 is a schematic diagram of a motor vehicle according to the invention, including a DC voltage power supply and a circuit for detecting electrical insulation faults in this power supply; [Figure 2] FIG. 2 is a schematic diagram of the detection circuit of FIG. 1. [Figure 3] 3 is a graph showing, on the one hand, the state of the power relay connected to the DC voltage source of FIG. 1 and, on the other hand, the change in voltage measured by the detection circuit of FIG. 2; [Figure 4] FIG. 4 is a detailed view of zone IV of FIG. 3. [Figure 5] 2 is a graph showing, on the one hand, the state of a charger suitable for connection to the DC voltage source of FIG. 1 and, on the other hand, the change in voltage measured by the detection circuit of FIG. 2 when connecting the charger to the DC voltage source. [Figure 6] 3 is a graph showing, on the one hand, the state of the charger and, on the other hand, the change in voltage measured by the detection circuit of FIG. 2 when the charger is disconnected from the DC voltage source; [Figure 7] 3 is a graph illustrating an example of modeling the change in voltage measured by the detection circuit of FIG. 2. [Figure 8] 2 is a flow chart illustrating steps of a method according to the present invention; DETAILED DESCRIPTION OF THE INVENTION

[0026] In FIG. 1, an electric vehicle 1 is shown.

[0027] The motor vehicle 1 includes an electric powertrain 10, which in this example includes an inverter 11 and an electric motor 12. The inverter 11 is conventionally designed to convert a DC voltage into an AC voltage.

[0028] To supply current to the inverter 11, the motor vehicle 1 includes a DC voltage source.

[0029] This power source is here a rechargeable battery 20 .

[0030] Such a battery 20 typically includes a housing that houses a number of small battery cells 21, the number of which is calculated to enable the electric motor to generate sufficient torque and power to propel the vehicle for a given period of time.

[0031] In FIG. 1, for clarity, only three storage battery cells 21 are shown.

[0032] Each cell typically has a voltage of about 3-5 V across its terminals. These cells are then connected in series to achieve the voltage level required by the application. Thus, the voltage V measured at the terminals of a traction battery can reach over 400 V when the battery is charged.

[0033] The two terminals of the storage battery 20 to which all the storage battery cells 21 are connected are terminals V - and V + It is called.

[0034] Motor vehicle 1 further includes a body 30, here formed by the vehicle's chassis and bodywork, which are generally made of a conductive metallic material. Body 30 is not permanently electrically connected to ground and thus forms a floating electrical ground. Body 30 is indicated in Figures 1 and 2 by an electrical ground symbol.

[0035] As FIG. 1 further shows, the motor vehicle 1 now further includes a protection circuit 31 and a power coupling circuit 32 between the terminals of the storage battery 20 and the terminals of the inverter 11 .

[0036] The protection circuit 31 comprises, as known per se, a fuse arranged to break the connection in the event of a short circuit, where a fuse is provided for each line connecting one of the terminals of the accumulator battery 20 to the inverter 11. As a variant, this protection circuit can comprise only a single fuse.

[0037] The power coupling circuit 32 includes at least one switch that allows the terminals of the battery 20 to be alternately connected to and selectively disconnected from the inverter 11. In Figure 1, this circuit is shown as including a switch 33, 34 in each line connecting one of the terminals of the battery 20 to the inverter 11. In practice, this power coupling circuit 32 may include, for example, a switch 34 in each line connecting the terminals V of the battery 20 to the inverter 11. + and inverter 11.

[0038] The motor vehicle 1 further comprises a system for making it possible to measure the insulation of the accumulator 20. This system is in fact formed by a circuit 40 for detecting electrical insulation faults between the accumulator 20 and the body 30. This detection circuit 40, which will be explained in more detail hereinafter, is connected to the body 30 and to a single terminal of the accumulator 20, here the terminal V - and electrically connected to the

[0039] The motor vehicle 1 finally includes a battery management computer 50 ("battery management system" or "BMS"), which includes a processor, a memory, and a data input / output interface.

[0040] Through this interface and the data exchange bus, the computer is configured to receive information from the detection circuit 40 and to control the switches 33 , 34 of the power combining circuit 32 .

[0041] Thanks to the memory, the computer stores computer applications consisting of computer programs containing instructions, and the processor executes the computer applications, thereby enabling the computer to carry out the methods described below.

[0042] Here, this computer 50 is supplied by a DC voltage source 51 which is different from the accumulator 20. In this case, the DC voltage source 51 is the power supply battery of the vehicle's on-board network.

[0043] In practice, the computer 50 is programmed to drive the opening of the two switches 33, 34 of the power combining circuit 32 when it receives information from the detection circuit 40 indicating that an electrical insulation fault has been detected.

[0044] An insulation fault is understood here to mean the abnormal existence of an electrical contact of low electrical resistance between the body 30 and a potential point of the accumulator 20. This potential point is the terminal V of the battery. + , V - This potential point can be any zone located between these two terminals, but is not limited to this.

[0045] Here, a resistance is said to be low if it is below a certain safety threshold, for example 100 kΩ.

[0046] Since the battery 20 and the loads supplied with current by this battery are fixed to the body of the vehicle, a so-called insulation resistance necessarily exists between the battery 20 and the body 30. Usually, this insulation resistance is much greater than the above-mentioned safety threshold.

[0047] 2 shows a single electrical insulation fault between point 22 of battery source 20 and body 30. This insulation fault is reflected here by the appearance of insulation resistance 66, which connects this point 22 with body 30 and has an R below the safety threshold mentioned above. i Indicates the value expressed as

[0048] In this situation, potentially dangerous leakage currents will flow through this insulation resistor 66 into the body 30. Such leakage currents are undesirable and can pose a danger to users of the motor vehicle 1 who may come into direct contact with the body 30.

[0049] The detection circuit 40 is then specifically designed to detect such a situation, and to that end is based on a modelling of an electrical insulation fault which takes the form of the diagram of FIG.

[0050] In this figure, the battery 20 has terminals V on either side of point 22. + , V - Note that the voltage between the two DC voltage sources 23, 24 is shown in series with each other.

[0051] These two DC voltage sources 23, 24 can then be written as supplying voltages (1-α)·Vbat and α·Vbat across their terminals, respectively, where α is a real number belonging to the interval [0;1].

[0052] By knowing this coefficient α, it is possible to accurately determine the location of the fault in the storage battery 20.

[0053] 2 also shows one embodiment of a detection circuit 40 in more detail.

[0054] In that regard, the detection circuit 40 - Terminal V of storage battery 20 - Between the body 30 and d a power supply 41 suitable for applying a DC voltage represented by -R m The measuring resistor 43, whose value is represented by C m a parallel RC circuit including a capacitor 44 having a capacitance represented by -R d and a current limiting resistor 42 with a value represented by

[0055] The power supply 41, the parallel RC circuit, and the current limiting resistor 42 are connected to the terminal V of the storage battery. - and the body 30.

[0056] The power supply 41 is controlled by the computer 50 to generate a voltage V d , which causes a current I passing through the current limiting resistor 42. d This leads to the emergence of

[0057] At that time, the value R of this resistor 52 d is the current I d is chosen to be low enough to facilitate measurement of R but high enough not to degrade the electrical insulation of the battery 20. d is at least 5 times greater than a predetermined safety threshold (recall that this is 100 kΩ).

[0058] The value of the measuring resistor 43, R m is advantageously a current value I d To keep the ratio R within a narrow range, d / R m is chosen to be between 1 and 100.

[0059] The detection circuit 40 here measures the voltage V at the terminals of the parallel RC circuit by means of an analog-to-digital converter, for example based on a differential amplifier, electrically connected in parallel with the parallel RC circuit. m This voltage V m The measurement of the current I d will be measured.

[0060] To detect electrical insulation faults in the battery 20, the computer 50 performs a loop detection method at regular time intervals. In each loop, the computer proceeds through two phases that can be performed sequentially (in any direction) or simultaneously.

[0061] Preferentially, the computer implements a first phase of the method, which consists in detecting a possible insulation fault. If such a fault is detected, the computer implements a second phase, which is more particularly the subject of the present invention and which consists in checking that the conditions for detecting such a fault have indeed been met and that the detection thus made is reliable.

[0062] To perform the first phase, the computer 50 calculates V during the second time period T2. d2 Since a second voltage of a value represented by V is applied during the first period T1, d1 The power supply 41 is controlled to apply a first voltage having a value expressed as:

[0063] These two periods are consecutive. They may immediately follow each other or may be separated, but the time interval that elapses between them must be short.

[0064] Here, each period has a duration of about 1 second.

[0065] Applied voltage value V d1 , V d2 are here different but of the same polarity.

[0066] Applied voltage value V d1 , V d2 Each value of d1 , I d2 and voltage V m Value V m1 , V m2 It has.

[0067] Due to the presence of a capacitor, when a voltage is applied, a current I d and the measured voltage V m has an initial phase called the transient region where these values change significantly, followed by a phase called the constant region where these values stabilize around a constant. m Value V m1 , Vm2 It is preferable to wait until the invariant region is established before proceeding to measure T. In practice, measurements of these values are made at the end of periods T1, T2.

[0068] The computer 50 then calculates the value R of the insulation resistance 66 using the following formula: i We proceed to calculate [Formula 1] TIFF0007723885000001.tif15170

[0069] The value of the coefficient α is calculated for that part by the computer 50 according to the following formula: [Formula 2] TIFF0007723885000002.tif16170

[0070] In these two equations, the voltage V at the terminals of the storage battery 20 bat is assumed to remain constant from one period to the other.

[0071] As a variant, V bat1 From a certain value expressed as V bat2 If it is desired to take into account possible changes in this voltage to another value, represented by , the following formula can be used: [Formula 3] TIFF0007723885000003.tif14170[Formula 4] TIFF0007723885000004.tif17170

[0072] In practice, the measured voltage V m The values that R takes are generally noisy. The results of the two equations used are then filtered. Preferentially, an RLS ("recursive least squares") filter is used to filter the value R of the insulation resistance 66. i and the value of the coefficient α.

[0073] The computer then calculates this value R i is compared with a predetermined safety threshold, here 100 kΩ.

[0074] As long as this value is above the threshold, no action is taken by the computer: in fact, in this situation the resistance is assumed to be such that there is no leakage current.

[0075] If this is not the case, a possible leakage current is detected in an accumulator cell whose position can be determined by the coefficient α.

[0076] In this case, the computer performs the second phase of the method to check that this detection is reliable.

[0077] In reality, there are situations in which the computer 50 will falsely detect a current leak, hereinafter referred to as "dangerous situations."

[0078] These dangerous situations occur when the signal measured by the detection circuit 40 is disturbed by electronic components connected to the battery.

[0079] Therefore, the second phase will make it possible to check whether one of these risk situations applies.

[0080] These dangerous situations occur when the voltage V d and the measured voltage V m This occurs when synchronization between

[0081] Before describing how this second phase is actually implemented, two non-limiting examples of hazardous situations that cause failure of insulation fault detection will first be described.

[0082] A first example of a hazardous situation relates to the switching of the switches 33 , 34 of the power combining circuit 32 .

[0083] As already explained, the storage battery 20 can be connected to the inverter 11 by closing one or more switches 33, 34. The connection is controlled, for example, every time the vehicle is started after being parked for a long period of time, and is therefore made periodically.

[0084] It can be seen that the switches 33, 34 are not closed immediately after the computer 50 sends the switching signal.

[0085] Therefore, the "open" state E T and "closed" state E T The transition between is not instantaneous but necessarily involves intermediate so-called "transient" states reflecting the phenomenon of charging and discharging of the control coils of the switches 33, 34.

[0086] 3 and FIG. 4, which is a detailed view of FIG. 3, the state E of these switches 33 and 34 T The variation of is shown using curve C1, which takes on the value 1 when the switch is closed, the value 2 when it is open, and the value 3 when it is in a transient state.

[0087] Curve C2 represents the value of the measured voltage V when the power supply 41 periodically establishes the same voltage at its terminals. m Shows the change in

[0088] Then, when the switches 33 and 34 are closed, during the transient state, the voltage V m It is observed that there is a risk that the leakage current may be falsely detected.

[0089] A second example of a hazardous situation relates to the charger for the traction battery 20.

[0090] If the motor vehicle 1 is of a rechargeable electric type, the motor vehicle 1 is equipped with an electric charger that allows the accumulator 20 to be recharged at an outlet of the local electric network (e.g. a household outlet), the terminals of which are then respectively connected to the terminals of the accumulator 20.

[0091] When the charger is connected to the local electrical network, the detection of electrical insulation faults is no longer handled by the battery management computer 50, but by a third party computer located in the charger.

[0092] At the time of switching, it is observed that there is a waiting period between the moment the charger is physically connected to the overall electrical network and the moment the computer 50 receives a request asking it to stop trying to detect electrical insulation faults.

[0093] During this time interval, the measured voltage V m This same kind of disturbance is observed after the charger is physically disconnected from the local electrical network.

[0094] In Figures 5 and 6, curve C3 is used to determine whether the charger is connected or disconnected. S This curve takes on the value 1 when the charger is connected to the local electrical network and the value 0 when the charger is disconnected. Curve C4 shows the relationship between the measured voltage V m Shows the change in

[0095] It should be noted in these figures that when the charger is in an open state, the power supply 41 periodically establishes a voltage at its terminals in order to perform detection of electrical insulation faults, whereas when the charger is connected, this type of voltage is not established by the power supply 41.

[0096] 5, just before computer 50 receives a request asking computer 50 to stop monitoring, voltage V m (Insulation resistance value R i It is observed that the current density (used to estimate the current density) is disturbed, which may lead to false detection of electrical insulation faults.

[0097] Further, in FIG. 6, immediately after computer 50 receives a request asking computer 50 to resume monitoring, voltage V m It is observed that the current is disturbed, which may lead to a false detection of an electrical insulation fault.

[0098] The present invention proposes to use a simple but effective criterion to identify dangerous situations (where there is a risk of a false detection of an electrical insulation fault). Thus, the present invention proposes that the measured voltage V m We propose to identify these dangerous situations based solely on the shape of the signal changes.

[0099] In FIG. 7, when one or other of the above dangerous situations occurs, the voltage signal V m In these dangerous situations, this signal actually increases rapidly and then drops back down rapidly.

[0100] To characterize the disturbances that are the cause of the false detection of electrical insulation faults, the invention proposes to base it on at least two parameters.

[0101] Here, preferentially, the check of the effectiveness of the detection of insulation faults is based on the use of exactly two parameters.

[0102] where these two parameters are: - ΔV m The measured voltage V between the two periods T1 and T2 is m deviations, and - The measured voltage V between the two periods T1 and T2, represented by V0 m is the average.

[0103] These two parameters are calculated by the following formula: [Formula 5] ΔVm = V m2 -V m1 [Formula 6] TIFF0007723885000005.tif17170

[0104] Each disturbance generates a rapid change in these two parameters, so during the second phase of the method it is possible to check the reliability of the insulation fault detection performed during the first phase by monitoring the trends of these parameters over time.

[0105] At this stage, these two parameters determine the value of the insulation resistance 66, R, according to the following formula: i and is directly related to the value of the coefficient α. [Formula 7] TIFF0007723885000006.tif16170[Formula 8] α=βV0+γ

[0106] In these equations, the variables β and γ are defined as follows: [Formula 9] TIFF0007723885000007.tif18170[Formula 10] TIFF0007723885000008.tif17170

[0107] Now, with reference to FIG. 8, it can be explained how the second phase is carried out after the initialization in step E0.

[0108] During a first step E2, the computer 50 calculates two previously measured voltage values V m1 , V m2 The computer 50 also obtains the two parameters ΔV calculated in the previous time interval. m ,V0 value (ΔV m,t-1 , V 0,t-1 (represented by

[0109] Then, during a second step E4, the computer calculates the two parameters ΔV m , the instantaneous value of V0 (ΔV m,t , V 0,t Calculate the

[0110] During a third step E6, the computer 50 monitors the evolution over time of the two parameters in order to detect whether a risk situation exists for a false detection of an electrical insulation fault.

[0111] To that end, the computer 50 now compares the change in each of the two parameters between the two time intervals with two thresholds S1, S2, respectively. In other words, the computer 50 checks whether the following two inequalities hold: [Formula 11] |ΔV m,t -ΔV m,t-1 |≦S1 [Formula 12] |V 0,t -V 0,t-1 |≦S2

[0112] The computer 50 then calculates the insulation resistance value R obtained during the first phase of the method from the result of these two inequalities. i The level of trust of the object is estimated. This level of trust is here Boolean and can take two states: "trusted" and "untrusted."

[0113] In practice, the computer 50 can be programmed in a variety of ways to use the above two inequalities.

[0114] In a first embodiment, the computer 50 is programmed to check whether one or less of these two inequalities is false.

[0115] If so, the computer 50 considers the results of the first phase to be unreliable. As a result, the value R of the insulation resistance 66 i No calculations based on

[0116] If not, the results of the first phase are deemed reliable and are therefore used during step E8. To this end, the computer determines the value R of the insulation resistor 66.i and the values of the coefficient α, and then use these data to control, for example, the opening of one or more switches 33, 34.

[0117] The method is then completed during step E10.

[0118] In a second embodiment, the computer 50 is programmed to check whether these two inequalities are simultaneously false. If so, the computer 50 considers the results of the first phase to be unreliable. If not, the results of the first phase are deemed to be reliable.

[0119] The present invention is in no way limited to the embodiments described and depicted, and a person skilled in the art will be able to make any modifications that are compatible with the present invention.

[0120] For example, a second phase may be performed before a first phase, and the implementation of the first phase may be adjusted based on the results of the second phase.

[0121] In other words, the value of insulation resistance 66 R i Before calculating the values of the coefficient α and the coefficient α, it is possible to determine whether there is a risk of a false detection. To do so, the computer checks whether one or both of the two above inequalities hold. If there is a risk, no other action is taken. If not, the computer 50 calculates the value R of the insulation resistance 66. i and the values of the coefficient α, and then use these data directly.

[0122] According to another variant of the invention, the two parameters used may be different from those used in the described embodiment. In the described embodiment, the two parameters ΔV m , V0 is the measured voltage V m are characteristics of the signal. In a variant, these two parameters are related to the measured voltage V mAs an example, these two parameters may be a simple function of the value of the insulation resistance 66, R i and the value of the coefficient α.

[0123] According to yet another variant of the invention, the two inequalities above may be based on the variation of the two parameters not between two consecutive time intervals, but over a larger number of time intervals.

[0124] According to yet another variant of the invention, the two inequalities above can be based on the values of the two parameters rather than on their variations over time.

Claims

1. A circuit (40) comprising a controllable voltage generator (41), an electrical measurement resistor (43) in a parallel RC circuit and a current limiting resistor (42) connected in series between the terminals of a power source (20) and electrical ground (30), wherein the voltage (V) at the terminals of the electrical measurement resistor is ｍ 1. A method for detecting an electrical insulation fault between the power source (20) and the electrical ground (30) by a circuit (40) further comprising means (45) for measuring - the voltage generator (41) generates a voltage (V ｄ controlling said voltage generator (41) to establish a the voltage (V) at the terminals of the electrical measuring resistor (43) to detect electrical insulation faults between the power source (20) and the electrical ground (30); ｍ ) measuring the - checking the reliability of said detection of said electrical insulation fault; In a method comprising: During the checking step, the computer: - the voltage (V) measured at least at the terminals of the electrical measuring resistor (43) ｍ ) as a function of the two parameters (ΔV ｍ , V ０ ) is determined, - the two parameters (ΔV ｍ , V ０ ) between two times and comparing the change in value of at least one of the above-mentioned parameters with a threshold value to calculate a level of confidence in the detection of the electrical insulation fault. It is characterized by the fact that - said voltage generator (41) is controlled so that said voltage (V d ) assumes two non-zero and different applied voltage values (V d1 , V d2 ) during two successive time periods (T 1 , T 2 ); two measurements (V m1 , V m2 ) of the voltage (V m ) at the terminals of the electrical measuring resistor (43) are taken during the two periods (T 1 , T 2 ), respectively; the parameter (V 0 ) is the arithmetic mean between the two measurements (V m1 , V m2 ) of the voltage (V m ) at the terminals of the electrical measuring resistor (43); the parameter (ΔV m ) is the deviation between the two measurements (V m1 , V m2 ) of the voltage (V m ) at the terminals of the electrical measuring resistor (43); a step of detecting the electrical insulation fault is provided between the measuring step and the checking step; In the detection step, to detect the electrical insulation fault, the computer calculates a value (R i ) of the insulation resistance (66) of the power source (20) with respect to the electrical ground (30) based on the value of the electrical measuring resistor (43), the value of the current limiting resistor (42), the two applied voltage values (V d1 , V d2 ) and the two measured values (V m1 , V m2 ), and then compares the value (R i ) with a safety threshold; A detection method, wherein the result of said detection step is used or not used depending on said calculated level of confidence.

2. A method for detecting electrical insulation faults between a power source (20) and electrical ground (30) by means of a circuit (40) including a controllable voltage generator (41), an electrical measuring resistor (43) and a current limiting resistor (42) in a parallel RC circuit, connected in series between the terminals of the power source (20) and electrical ground (30), the circuit (40) further including means (45) for measuring the voltage (V m ) at the terminals of the electrical measuring resistor, - controlling said voltage generator (41) so that it establishes a voltage (V d ) of non-zero value across its terminals; measuring the voltage (V m ) at the terminals of the electrical measuring resistor (43) to detect electrical insulation faults between the power source (20) and the electrical ground (30); - checking the reliability of said detection of said electrical insulation fault; In a method comprising: During the checking step, the computer: - determining two parameters (ΔV m , V 0 ) for checking the reliability of the detection of the electrical insulation fault as a function of the voltage (V m ) measured at least at the terminals of the electrical measuring resistor (43); calculating the level of confidence of said detection of said electrical insulation fault from the change in value between two times for at least one of said two parameters (ΔV m , V 0 ) and comparing it with a threshold value; It is characterized by the fact that - said voltage generator (41) is controlled so that said voltage (V d ) assumes two non-zero and different applied voltage values (V d1 , V d2 ) during two successive time periods (T 1 , T 2 ); two measurements (V m1 , V m2 ) of the voltage (V m ) at the terminals of the electrical measuring resistor (43) are taken during the two periods (T 1 , T 2 ), respectively; the parameter (V 0 ) is the arithmetic mean between the two measurements (V m1 , V m2 ) of the voltage (V m ) at the terminals of the electrical measuring resistor (43); the parameter (ΔV m ) is the deviation between the two measurements (V m1 , V m2 ) of the voltage (V m ) at the terminals of the electrical measuring resistor (43); after said checking step, said step of detecting an electrical insulation fault is performed only if said level of confidence indicates that the detection of said electrical insulation fault will be reliable, In the detection step, to detect the electrical insulation fault, the computer calculates a value (R i ) of the insulation resistance (66) of the power source (20) relative to the electrical ground (30) based on the value of the electrical measuring resistor (43), the value of the current limiting resistor (42), the two applied voltage values (V d1 , V d2 ) and the two measured values (V m1 , V m2 ), and then compares the value (R i ) with a safety threshold.

3. A detection method as described in claim 1 or 2, wherein if an electrical insulation fault is detected, the computer calculates a coefficient (α) for locating the electrical insulation fault in the power supply (20) based on the voltage at the terminals of the power supply (20), the two applied voltage values (V d1 , V d2 ) and the two measured values (V m1 , V m2 ).

4. A device for detecting electrical insulation faults between a power source (20) and an electrical ground (30), comprising: said computer implementing the detection method according to any one of claims 1 to 3; said circuit (40), Including, the device.

5. A motor vehicle including a power source (20) and an electrical ground (30), characterized in that the motor vehicle includes a device according to claim 4.

Images