JP2016099323A - Insulation state detector - Google Patents

Abstract

PROBLEM TO BE SOLVED: To correctly measure ground-fault resistance having a large resistance value by making a necessary measurement time longer even when a charging voltage of a flying capacitor is affected large by a Y capacitor of a high-voltage power source.SOLUTION: In a Vc1n measurement section A2 and a Vc1p measurement section A4, measured values (a), (b) of voltage are acquired at two different timing points shifting from each other in the same charging section. On the basis of the ratio of the acquired measured values (a), (b) at the two points, a ground-fault resistance value is calculated using a ground-fault resistance map. An influence of an applied voltage V0 can be excluded using the ratio of the measured value at the two points. At a timing of a short charging time, an influence of a difference in ground-fault resistance value is apt to appear since a gradient of rising of a charging voltage waveform is large. When a difference between a plus-side measured value and a minus-side measured value is large, only a smaller one is corrected and the calculation is performed.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an insulation state detection device that can be mounted on a vehicle.

  For example, in a vehicle that uses electric power as propulsion energy, such as an electric vehicle, a DC power supply device that outputs a high voltage of about 200 V may be mounted. In the case of a vehicle equipped with such a high-voltage DC power supply device, it is used in a state where the positive and negative power supply lines of the DC power supply device and the vehicle body are electrically insulated. That is, the vehicle body is not used as a negative power line for a power source that outputs a high voltage.

  In such a vehicle, it is necessary to detect and confirm that the high-voltage DC power output wiring and the vehicle body are sufficiently electrically insulated in order to ensure safety. For example, Patent Literature 1, Patent Literature 2, and Patent Literature 3 disclose conventional techniques of an insulation state detection device used when performing such detection.

  This type of insulation state detection device uses a flying capacitor. That is, a detection capacitor (referred to as a flying capacitor) is connected between a high-voltage positive and negative power supply line and a ground line (vehicle body) through a switching element for a predetermined time. The charging voltage of the flying capacitor is monitored, and a ground fault resistance, that is, an insulation resistance between the power supply line and the ground line is calculated from the charging voltage.

JP 2009-281986 A JP 2009-281987 A JP 2011-102788 A

  By the way, noise generated from various devices and noise entering from the outside are propagated to other devices via the power supply line and cause various malfunctions. Therefore, for example, in the case of a high-voltage power supply for vehicles, it is necessary to use a line bypass capacitor connected between each electrode of the power supply and the earth line in order to reduce common mode noise. Further, stray capacitance exists between the power supply line and the ground line. They are collectively referred to as Y capacitors.

  However, when using the insulation state detection device as described above, when the flying capacitor is charged, the flying capacitor and the Y capacitor are connected in parallel. For this reason, electric charge flows into the flying capacitor from the high-voltage power supply of the vehicle via the ground fault resistance, and at the same time, electric charge accumulated in the Y capacitor also flows into the flying capacitor due to the potential difference. Therefore, the influence of the Y capacitor appears on the charging voltage of the flying capacitor.

  In addition, there are actually a positive side Y capacitor and a negative side Y capacitor, and the potential at these connection points is normally stable in a balanced state, but the balance is caused by the outflow of charge from one of the Y capacitors. Collapse. Even if the balance of the potentials of these two Y capacitors is lost due to the outflow of electric charge, after the measurement of the positive side ground fault resistance measurement voltage VC1p and the negative side ground fault resistance measurement voltage VC1n, Since charge flows from the high-voltage power supply to the low-voltage side Y capacitor via the loop resistance, the potentials of the two Y capacitors naturally return to a balanced state after a certain amount of time has passed. Further, when the potentials of the two Y capacitors are stable in a balanced state, the influence of the Y capacitor on the charging voltage of the flying capacitor can be calculated in advance, so that the ground fault resistance can be obtained by correction.

  However, in order to shorten the time required for measuring the ground fault resistance, it is necessary to shorten the cycle from the charging of the flying capacitor to the next measurement. In this case, the next measurement is performed in a state where the potentials of the two Y capacitors are not balanced. However, since it is unclear how much the balance has been lost, an accurate value can be obtained by correcting when the measured value is balanced. There is a problem that it will not be possible. That is, if the measurement cycle is short, the Y capacitor has a large capacitance (for example, about 0.5 [μF]) as described above. The initial charge transfer from the Y capacitor to the flying capacitor, which occurs to return to the balanced state, is superimposed on the charge transfer from the high-voltage power supply to the flying capacitor. Become.

  The higher the ground fault resistance, the longer the time until the Y capacitor balances. Therefore, when the capacitance of the Y capacitor is large and the measurement cycle is short, the higher the ground fault resistance, the more the initial state. In this case, the degree of imbalance of the Y capacitor increases, the amount of initial charge transferred from the Y capacitor to the flying capacitor increases, and the charging voltage of the flying capacitor greatly increases.

By the way, the measurement of the ground fault resistance is performed by utilizing the fact that the charging voltage of the flying capacitor increases as the ground fault resistance decreases. On the other hand, when the capacitance of the Y capacitor is large and the measurement cycle is short, as described above, the higher the ground fault resistance, the greater the degree of unbalance of the Y capacitor, and the charging of the flying capacitor. The voltage rises more greatly. Therefore, when the capacitance of the Y capacitor is large, if the measurement cycle is shortened, whether the voltage is caused by the ground fault resistance or not even if the charging voltage of the flying capacitor is measured, It cannot be distinguished whether it is caused by balance, and it is difficult to correctly determine the value of ground fault resistance.
Therefore, the conventional insulation state detection device cannot accurately measure the value of the ground fault resistance if the measurement cycle is shortened when the capacitance of the Y capacitor is large, and it is difficult to accurately detect the decrease in the ground fault resistance. There was a problem that.

  The present invention has been made in view of the above situation, and even if the charging voltage of the flying capacitor is greatly influenced by the charge of the Y capacitor connected to the high voltage power source, the measurement time is lengthened. An object of the present invention is to provide an insulation state detection device capable of correctly measuring a ground fault resistance.

The above object of the present invention is achieved by the following configuration.
(1) A positive-side input terminal and a negative-side input terminal connected to a positive-side power supply line and a negative-side power supply line of a predetermined high-voltage DC power supply output, respectively, and the positive-electrode side based on the charging voltage of the flying capacitor An insulation state detection device for grasping an insulation state between a power line and a negative side power line and ground as a ground fault resistance,
Each of the flying capacitors at a first time and a second time having different timings in one charging section when charging the flying capacitor via a predetermined energization path affected by the ground fault resistance. A charging voltage measuring unit for measuring the charging voltage of
A ground fault resistance value calculation unit that calculates a value of the ground fault resistance based on the voltage measurement value of the first time measured by the charging voltage measurement unit and the voltage measurement value of the second time;
It is characterized by providing.

(2) The insulation state detection device according to (1),
The ground fault resistance value calculation unit obtains a plurality of constant data in a state in which the ratio of the voltage measurement value at the first time and the voltage measurement value at the second time is associated with the resistance value of the ground fault resistance. Hold ground fault resistance map,
It is characterized by providing.

(3) The insulation state detection device according to (1) or (2),
Two points where the ground fault resistance value calculation unit calculates the value of the ground fault resistance based on the voltage measurement value at the first time measured by the charging voltage measurement unit and the voltage measurement value at the second time. Measurement mode and
The charging voltage measuring unit measures the charging voltage of the flying capacitor at a third time in the one charging period, and the ground fault resistance calculating unit is measured by the charging voltage measuring unit. A normal measurement mode for calculating the value of the ground fault resistance based on the voltage measurement value of the time;
It is characterized by having.

(4) The insulation state detection device according to (1),
The charging voltage measurement unit measures a voltage of the flying capacitor as a negative measurement value when charging the flying capacitor through an energization path affected by a ground fault resistance between the negative power supply line and the ground electrode. And measuring the voltage of the flying capacitor as a positive measurement value when charging the flying capacitor in an energization path affected by a ground fault resistance between the positive power line and the ground electrode,
When the difference between the negative side measured value and the positive side measured value is equal to or greater than a predetermined value, the ground fault resistance value calculating unit determines only the smaller one of the negative side measured value and the positive side measured value. Applying correction using a constant determined in advance, and calculating the value of the ground fault resistance based on the negative measurement value and the positive measurement value after correction,
It is characterized by that.

(5) The insulation state detection device according to (1),
In a situation where the first time is earlier than the second time, the voltage measurement value at the first time is determined in advance immediately after the charging voltage measurement unit measures the voltage measurement value at the first time. A measurement control unit that suppresses the maximum value of the charging voltage by at least shortening the time length of the charging section when the voltage measurement value at the first time exceeds the upper limit value,
Is further provided.

  According to the insulation state detection device of the present invention, even if the charging voltage of the flying capacitor is greatly affected by the charge of the Y capacitor connected to the high voltage power source, the ground fault resistance is not increased without increasing the measurement time. The correct measurement becomes possible.

  The present invention has been briefly described above. Further, details of the present invention will be further clarified by reading through the modes for carrying out the invention described below with reference to the accompanying drawings.

FIG. 1 is a time chart showing the operation of a basic measurement cycle including features of the present invention. FIG. 2 is an electric circuit diagram illustrating a configuration example of the insulation state detection device and its peripheral circuits according to the embodiment. 3A and 3B are operations of a basic measurement cycle that does not include the features of the present invention. FIG. 3A is a state transition diagram showing an operation state transition, and FIG. It is a wave form diagram showing the change of the voltage between the terminals of a flying capacitor. FIG. 4 is an electric circuit diagram showing a state where a part of the circuit shown in FIG. 2 is extracted and showing a charging state during the V0 measurement operation. FIG. 5 is an electric circuit diagram showing a state in which a part of the circuit shown in FIG. 2 is extracted and showing a state when charging voltage is measured. FIG. 6 is an electric circuit diagram showing a state where a part of the circuit shown in FIG. 2 is extracted and showing a charging state during the VC1n measurement operation. FIG. 7 is an electric circuit diagram showing a state in which a part of the circuit shown in FIG. 2 is extracted and showing a charging state during the VC1p measurement operation. 8A and 8B show a state where a part of the circuit shown in FIG. 2 is extracted. FIG. 8A shows the charge of the Y capacitor Yp when charging for VC1n measurement. FIG. 8B is an equivalent circuit diagram showing the influence of the ground fault resistance RLn when charging for measuring VC1n. FIG. 9 is a waveform diagram showing a specific example of fluctuations in the inter-terminal voltage of C1 and the balance potential of Yp and Yn. FIG. 10 is a waveform diagram showing a specific example of the charge / discharge voltage of the detection capacitor C1 when measuring VC1n or VC1p under two types of conditions with different resistance values of ground fault resistance. FIG. 11 is a waveform diagram showing changes in the maximum charging voltage when the charging time is shortened. FIG. 12A and FIG. 12B are flowcharts showing a characteristic operation (2) and its modification in the embodiment of the present invention.

  Specific embodiments relating to the insulation state detection device of the present invention will be described below with reference to the drawings.

<Overview of overall configuration and operation>
FIG. 2 shows the configuration of the insulation state detection device 10 mounted on the vehicle and its peripheral circuits.

  The insulation state detection apparatus 10 shown in FIG. 2 can be used by being mounted on a vehicle such as an electric vehicle or a hybrid vehicle including an engine and an electric motor as drive sources. The in-vehicle DC high-voltage power supply 50 outputs high-voltage DC power of about 200V, for example. The electric motor M that generates the driving force of the vehicle can be driven by the electric power output from the in-vehicle DC high-voltage power supply 50.

  The positive power supply line 111 and the ground line 103 at the output of the in-vehicle DC high voltage power supply 50 are electrically insulated. Further, the negative power supply line 112 and the ground line 103 are also electrically insulated. The ground line 103 corresponds to a ground portion such as a vehicle body. Here, the insulation state between the positive power supply line 111 and the ground line 103 can be expressed as a ground fault resistance RLp. Further, the insulation state between the negative power supply line 112 and the ground line 103 can be expressed as a ground fault resistance RLn.

  Further, in order to reduce common mode noise, as shown in FIG. 2, a Y capacitor Yp is connected between the positive power supply line 111 and the ground line 103, and between the negative power supply line 112 and the ground line 103. A Y capacitor Yn is connected to this. The Y capacitors Yp and Yn also include stray capacitance.

  By mounting the insulation state detection device 10 shown in FIG. 2 on the vehicle, the insulation state of the vehicle can be monitored at any time as necessary. That is, the insulation state detection device 10 can be used to detect the ground fault resistances RLp and RLn at the output of the in-vehicle DC high voltage power supply 50 and grasp the insulation state.

  Therefore, as shown in FIG. 2, the positive input terminal 13 and the negative input terminal 14 of the insulation state detection device 10 are connected to the positive power supply line 111 and the negative power supply line 112, respectively. The grounding point 15 of the insulation state detection device 10 is connected to the grounding line 103.

  An output terminal 21 is provided as shown in FIG. 2 in order to output the measurement result of the insulation state detection device 10 and alarm information. The output terminal 21 can be connected to, for example, a vehicle-side electronic control unit (ECU).

<Configuration Example of Insulation State Detection Device 10>
As shown in FIG. 2, the circuit of the insulation state detection device 10 is provided with a detection capacitor C1 that operates as a flying capacitor. A ceramic capacitor is employed as the detection capacitor C1 in consideration of being used in a vehicle.

  Further, in order to control charging and discharging of the detection capacitor C1, four switching elements S1 to S4 are provided in the vicinity thereof. Each of these switching elements S1 to S4 is a switch that can switch the open / close (off / on) state of a contact by controlling an insulated signal, such as an optical MOSFET.

  The switching element S1 has one end connected to the positive input terminal 13 via the resistor R01 and the other end connected to the wiring 31. The switching element S2 has one end connected to the negative input terminal 14 via the resistor R02 and the other end connected to the wiring 32 via the resistor R2.

  The switching element S <b> 3 has one end connected to the wiring 33 and the other end connected to the wiring 35. The switching element S4 has one end connected to the wiring 32 and the other end connected to the grounding point 15 via the resistor R4.

  The detection capacitor C <b> 1 has a negative terminal connected to the wiring 32. The positive terminal of the detection capacitor C1 is connected to the wiring 31 through a series circuit including a diode D1 and a resistor R1. The positive terminal of the detection capacitor C1 is connected to the wiring 33 via a series circuit including a diode D3 and a resistor R3, and is further connected to the wiring 33 via a diode D2. The diode D2 is connected with a polarity that allows energization in the direction from the wiring 33 to the wiring 34, and the diode D3 is connected with a polarity that allows energization in the direction from the wiring 34 to the wiring 33.

  The microcomputer (CPU) 11 executes various controls required for the insulation state detection device 10 by executing a program incorporated in advance. Specifically, the microcomputer 11 controls the switching elements S1 to S4 individually to control charging and discharging of the detection capacitor C1. Further, the microcomputer 11 inputs an analog level corresponding to the charging voltage of the detection capacitor C1 from one analog input port AD1 via the wiring 36, performs calculation based on this input level, and performs ground fault resistance RLp and Know RLn.

  The input circuit 20 is connected between the wiring 35 and the wiring 36. The input circuit 20 performs signal processing for converting a signal appearing on the wiring 35 into a signal suitable for the processing of the microcomputer 11. Although various functions and configurations are conceivable for the input circuit 20, a sample hold circuit is assumed as a typical input circuit 20.

  For example, an analog switch is connected between the wiring 35 and the wiring 36, and a capacitor (capacitor) that holds a signal level is connected between the wiring 36 and the ground point 15. When the microcomputer 11 temporarily turns on (conducts) the analog switch at a specific timing, the signal level input at that timing can be sampled and held by the capacitor in the input circuit 20. . Of course, the function of such a sample and hold circuit can be omitted.

<Measurement of ground fault resistance>
<Description of basic measurement cycle>
FIG. 3A and FIG. 3B show the operating state transition and the change in the voltage between the terminals of the flying capacitor. In order to facilitate understanding of the basic operation, description of characteristic operations of the present invention is omitted in FIGS. 3 (A) and 3 (B).

  As shown in FIG. 3 (A), the basic measurement cycle of the insulation state detection apparatus 10 includes “V0 measurement section” A1, “VC1n measurement section” A2, “V0 measurement section” A3, and “VC1p measurement section” A4. It is composed of a series of measurement sections, and this basic measurement cycle is repeatedly executed.

  In the “V0 measurement section” A1 and A3, the charging voltage V0 that changes according to the power supply voltage is measured. The charging voltage V0 is independent of the resistance value of the ground fault resistance.

  In the “VC1n measurement section” A2, the charging voltage VC1n that changes under the influence of the resistance value of the ground-fault resistance RLn on the negative electrode side is measured. In the “VC1p measurement section” A4, the charging voltage VC1p that changes under the influence of the resistance value of the ground fault resistance RLp on the positive electrode side is measured.

  In each of the sections A1, A2, A3, and A4, a “charging operation” for charging the detection capacitor (flying capacitor) C1 and a voltage between terminals of the detection capacitor C1 are measured. A “measurement operation” and a “discharge operation” for discharging the charge accumulated in the detection capacitor C1 are included.

  As shown in FIG. 3B, the voltage between the terminals of the detection capacitor C1 increases according to an exponential function (exp) curve determined by the time constant of the charging circuit when charging is started in each section, and is discharged when discharging is started. It descends according to an exponential curve determined by the time constant of the circuit. When the basic measurement cycle of FIG. 3A is performed, the detection capacitor C1 is repeatedly charged and discharged, so that the voltage of the detection capacitor C1 changes as shown in FIG.

  In “V0 measurement section” A1 and A3, the voltage when the charging of the detection capacitor C1 is completed is measured as V0.

  Further, in the “VC1n measurement section” A2, the voltage before the transition to the discharge after the charging of the detection capacitor C1 is completed is measured as VC1n. In the “VC1p measurement section” A4, the voltage before the detection capacitor C1 is completely discharged and shifted to discharge is measured as VC1p.

  Actually, the microcomputer 11 shown in FIG. 2 controls the on / off of the switching elements S1 to S4, whereby the operation of the basic measurement cycle shown in FIG. 3A can be executed. And ground fault resistance is computable based on the measured value V0, Vc1n, and Vc1p obtained as a result of this basic measurement cycle.

<Description of the status of each section>
<Charging state of “V0 measurement section” A1, A3>
A part is extracted from the circuit shown in FIG. 2, and locations related to the charging states of “V0 measurement sections” A1 and A3 are shown in FIG.

  In the charging state of “V0 measurement section” A1 and A3, the switching elements S1 and S2 are turned on (contacts are closed), and the other switching elements are turned off (contacts are opened). Therefore, an energization path as shown in FIG. 4 is formed, and a current flows as shown by an arrow in FIG.

  That is, since the contact point of the switching element S1 is closed, the positive side terminal 111, the switching element S1, the wiring 31, the diode D1, and the resistor R1 pass through the positive side power line 111 to the positive side terminal of the detection capacitor C1. Current flows. Further, since the contact of the switching element S2 is closed, a current flows from the negative terminal of the detection capacitor C1 to the wiring 32, the resistor R2, the switching element S2, the negative input terminal 14, and the negative power line 112. Therefore, the electric charge is charged in the detection capacitor C1 by this current. The charging voltage of the detection capacitor C1 at this time is determined according to the output voltage of the in-vehicle DC high voltage power supply 50, the time constant of the charging circuit, and the charging time, but is not affected by the ground fault resistors RLn and RLp.

<State at the time of charging voltage measurement>
A part extracted from the circuit shown in FIG. 2 is shown in FIG. 5 where the charging voltage is measured.

  When measuring the charging voltage in each of the sections A1 to A4, the contacts of the switching elements S3 and S4 are closed, and the other switching elements are turned off (contact open). Therefore, an energization path as shown in FIG. 5 is formed. That is, as shown in FIG. 5, the negative terminal of the detection capacitor C1 is connected to the ground point 15 via the resistor R4. The positive terminal of the detection capacitor C1 is connected to the analog input port of the microcomputer 11 via the diode D3, the resistor R3, the switching element S3, the wiring 35, the input circuit 20, and the wiring 36. Therefore, the microcomputer 11 can detect an analog level proportional to the charging voltage of the detection capacitor C1.

  Note that the same state as in FIG. 5 is obtained even when the detection capacitor C1 is discharged, except that the input circuit 20 is turned off. That is, the charge on the positive electrode side accumulated in the detection capacitor C1 passes through the diode D3, the resistors R3 and R5, and the charge on the negative electrode side naturally discharges with the passage of time through the resistor R4.

<Charging state of “VC1n measurement section” A2>
A part of the circuit shown in FIG. 2 is extracted, and a portion related to the state of charge of “VC1n measurement section” A2 is shown in FIG.

  In the charging state of “VC1n measurement section” A2, the contacts of the switching elements S1 and S4 are closed, and the other switching elements are turned off (contacts open). Therefore, an energization path as shown in FIG. 6 is formed.

  That is, as indicated by an arrow in FIG. 6, a current flows from the positive power supply line 111 to the positive terminal of the detection capacitor C1 through the positive input terminal 13, the switching element S1, the wiring 31, the diode D1, and the resistor R1. Flows. In addition, since the contact of the switching element S4 is closed, the negative power supply line from the negative terminal of the detection capacitor C1 passes through the switching element S4, the resistor R4, the ground point 15, the ground line 103, and the ground fault resistor RLn. A current flows through 112. This current charges the detection capacitor C1. The charging voltage at this time is a result reflecting the influence of the ground fault resistance RLn.

<Charging state of “VC1p measurement section” A4>
A part of the circuit shown in FIG. 2 is extracted, and a portion related to the state of charge of “VC1p measurement section” A4 is shown in FIG.

  In the charging state of the “VC1p measurement section” A4, the contacts of the switching elements S2 and S3 are closed, and the other switching elements are turned off (contacts open). Therefore, an energization path as shown in FIG. 7 is formed.

  That is, as indicated by an arrow in FIG. 7, the detection capacitor C1 passes through the grounding resistance RLp, the ground line 103, the ground point 15, the resistor R5, the switching element S3, and the diode D2 from the positive power supply line 111. Current flows through the positive terminal. In addition, a current flows from the negative terminal of the detection capacitor C1 to the wiring 32, the resistor R2, the switching element S2, the negative input terminal 14, and the negative power line 112. This current charges the detection capacitor C1. The charging voltage at this time is a result reflecting the influence of the ground fault resistance RLp.

<Basic grounding resistance measurement operation>
The operation of the insulation state detection device 10 shown in FIG. 2 basically has the following relationship.

(RLp + RLn) / (RLp × RLn) = f [{(Vc1p) + (Vc1n)} / V0] (1)
However,
V0: Charge voltage Vc1n of the detection capacitor C1 corresponding to the output voltage of the in-vehicle DC high-voltage power supply 50: Charge voltage Vc1p of the detection capacitor C1 affected by the negative ground fault resistance RLn: Positive ground fault resistance RLp Voltage RLp, RLn of the detection capacitor C1 affected by the resistance: resistance value of the local resistance

  Therefore, the microcomputer 11 grasps each charging voltage “V0”, “Vc1n”, “Vc1p” from the signal level inputted to the analog input port (AD1) in each state, and the ground fault resistance based on the above relational expression. RLp and RLn can be calculated.

<Description of influence of Y capacitors Yp and Yn>
Y capacitors Yp and Yn are connected between the positive power supply line 111, the negative power supply line 112 and the ground line 103, which are the outputs of the in-vehicle DC high voltage power supply 50 shown in FIG. These Y capacitors Yp and Yn are for reducing common mode noise, but may have a significant effect when the insulation state detection device 10 detects the insulation state. Specifically, for example, when the resistance values of the ground fault resistances RLn and RLp are about 1 [MΩ] or more, the measurement values “Vc1n” and “Vc1p” that change due to the influence of the ground fault resistances RLn and RLp are Since the situation is difficult to distinguish from the effects of the capacitors Yp and Yn, the ground fault resistances RLn and RLp having large resistance values cannot be measured. In particular, when the electrostatic capacities of the Y capacitors Yp and Yn are relatively large (for example, 0.5 [μF]), the influence becomes significant.

<Explanation of charge transfer>
The influence of Y capacitors Yp and Yn in FIG. 2 will be described with reference to the equivalent circuit diagrams of FIGS. 8 (A) and 8 (B). FIG. 8A shows a path through which the charge of the Y capacitor Yp moves when the detection capacitor C1 for “VC1n” measurement is charged in the circuit shown in FIG. FIG. 8B shows a path through which a charging current that changes due to the influence of the ground fault resistance RLn flows when charging the detection capacitor C1 for measuring “VC1n”.

  That is, the charging state shown in FIG. 6 includes the path through which the charging current IRLn flows through the path including the ground fault resistance RLn as shown in FIG. 8B and the charge of the Y capacitor Yp as shown in FIG. 8A. It can be considered as a combined circuit with the path through which the current Iyp flows.

  In the state shown in FIG. 8A, since the voltage between the terminals of the Y capacitor Yp is large and the voltage between the terminals of the detection capacitor C1 is small at the beginning of charging, the charge accumulated in the Y capacitor Yp has a low voltage. The current Iyp flows to move to the detection capacitor C1 side. At the same time, a charging current IRLn affected by the ground fault resistance RLn flows as shown in FIG.

  Therefore, the current for charging the detection capacitor C1 is the result of adding Iyp in FIG. 8A and IRLn in FIG. 8B. That is, the measured value “Vc1n” of the charging voltage of the detection capacitor C1 is not only influenced by the ground fault resistance RLn (IRLn) but also influenced by the Y capacitor Yp (Iyp). When the influence of the Y capacitor Yp is larger than the influence of the ground fault resistance RLn, it is difficult to calculate the ground fault resistance RLn from the measured value.

  Although not shown, when “Vc1p” is measured, the electric charge moves from the Y capacitor Yn to the detection capacitor C1, and accordingly, the current Iyn flows. Then, the detection capacitor C1 is charged by a current obtained by adding the charging current IRLp affected by the ground fault resistance RLp and the current Iyn. Therefore, when the influence of the Y capacitor Yn is larger than the influence of the ground fault resistance RLp, the influence of the Y capacitor Yn included in the measured value “Vc1p” becomes large, and it is difficult to calculate the correct ground fault resistance RLp. become.

<Explanation of Balance of Yp and Yn Voltage>
In the circuit shown in FIG. 2, since the Y capacitors Yp and Yn have the same capacitance, the Y capacitors Yp and Yn are in a balanced state in a steady state before starting measurement. That is, the voltage output from the in-vehicle DC high-voltage power supply 50 is divided according to the ratio of the Yp and Yn capacitances, and the voltage between the terminals of the Y capacitor Yp and the voltage between the terminals of the Y capacitor Yn substantially match. State (for example, 100 [V]). The potential of the ground line 103 is stable at the median value of the potential of the positive power supply line 111 and the potential of the negative power supply line 112.

(1) On the other hand, when measuring “VC1n”, the balance is lost because the electric charge of the Y capacitor Yp flows as the current Iyp, as in the equivalent circuit shown in FIG. That is, the electric charge of the Y capacitor Yp flows out and the voltage between the terminals of the Y capacitor Yp decreases, so that the balance of the voltages of the Y capacitors Yp and Yn is lost, and the potential of the ground line 103 changes.

  When the voltage balance is lost in this way, in the “V0 measurement section” after the “VC1n” measurement, the charge flowing out from the Y capacitor Yp is charged again from the in-vehicle DC high voltage power supply 50 via the ground fault resistor RLn. Therefore, it will return to a balanced state. However, if the capacitor of the Y capacitor is large and the measurement cycle is short, the next measurement is started before returning to a balanced state.

(2) Further, when measuring “VC1p”, the electric charge of the Y capacitor Yn flows to the detection capacitor C1 side as the current Iyn, and the balance is lost. That is, since the charge of the Y capacitor Yn flows out and the voltage between the terminals of the Y capacitor Yn decreases, the balance of the voltages of the Y capacitors Yp and Yn is lost, and the potential of the ground line 103 changes. Furthermore, when the “VC1p” is to be measured in the state where the influence of the balance disruption that occurs when measuring “VC1n” remains, the voltage between the terminals of the Y capacitor Yn is larger than usual. The amount of charge flowing out to the detection capacitor C1 side further increases.

(3) Also, when executing the second and subsequent basic measurement cycles, an attempt is made to measure “VC1n” in a state where the influence of the balance loss that occurred when measuring “VC1p” of the previous cycle remains. Then, since the voltage across the terminals of the Y capacitor Yp is larger than usual, the amount of charge flowing out to the detection capacitor C1 side further increases.

  That is, since the states (1), (2), and (3) are repeated, the charging current of the detection capacitor C1 is increased more than usual due to the influence of the balance of the Y capacitors Yp and Yn being lost. Then, in the measured values of “VC1n” and “VC1p”, it becomes impossible to distinguish the influence of the ground fault resistances RLn and RLp from the influence of the Y capacitors Yp and Yn. A specific example of the fluctuation of the balance potential is shown in FIG.

  In addition, regarding the correspondence between the detected value of the ground fault resistance and the true value when measurement is performed in a normal procedure using the insulation state detecting device 10 shown in FIG. 2, the Y capacitors Yp and Yn (Ycon) The simulation was performed while changing the capacitance and ground fault situation. As a result, when the electrostatic capacitances of the Y capacitors Yp and Yn are about 0.3 [μF], a high resistance value of 1000 [kΩ] or more cannot be determined.

<Description of characteristic operation>
<Measurement operation of ground fault resistance using two-point measurement>
The operation of the basic measurement cycle including the features of the present invention is shown in FIG. The basic measurement cycle shown in FIG. 1 is similar to the basic measurement cycle of FIGS. 3A and 3B, “V0 measurement section” A1, “VC1n measurement section” A2, “V0 measurement section” A3, “VC1p measurement section” A4 is composed of a series of measurement sections, and this basic measurement cycle is repeatedly executed.

  In the “VC1n measurement section” A2 in FIG. 1, the first VC1n measurement value (a) is measured when a certain time TCna (for example, 0.1 second) has elapsed from the charging start time t3, and the certain time from the charging start time t3. When TCnb (for example, 0.4 seconds) elapses, the second VC1n measurement value (b) is measured.

  Further, in the “VC1p measurement section” A4 in FIG. 1, the first VC1p measurement value (a) is measured when a certain time TCpa (for example, 0.1 second) has elapsed from the charge start time t8, and from the charge start time t3. The second VC1p measurement value (b) is measured when a certain time TCpb (for example, 0.4 seconds) has elapsed.

  Further, in the ground fault resistance calculation processing unit 41 for calculating the ground fault resistance, the ratio of the two VC1n measurement values (a) and (b) and the ratio of the two VC1p measurement values (a) and (b) are obtained. Based on this, the resistance values of the ground fault resistances RLn and RLp are calculated. The V0 measurement value may not be used. The ground fault resistance calculation processing unit 41 uses the ground fault resistance map holding unit 42 when calculating the ground fault resistance.

  The ground fault resistance map held by the ground fault resistance map holding unit 42 is configured by a large number of predetermined constant data, and is different from the ratio of the measured values of the two points (a, b) of VC1n and VC1p. It is held in a state in which any of a number of ground fault resistance values is associated. That is, the resistance value of the ground fault resistance RLn can be acquired from the ratio of the measured values of the two points (a, b) of VC1n by the ground fault resistance map, and the ratio of the measured values of the two points (a, b) of VC1p. From the ground fault resistance map, the resistance value of the ground fault resistance RLp can be acquired.

  For regions where the resistance values of the ground fault resistances RLn and RLp are relatively small (for example, 400 [kΩ] or less), the ground fault resistance is calculated from the measured values of V0, VC1n, and VC1p using the above equation (1). A value may be calculated.

<Advantages of 2-point measurement>
FIG. 10 shows a specific example of the waveform of the charge / discharge voltage of the detection capacitor C1 when measuring VC1n or VC1p under two types of conditions where the resistance values of the ground fault resistors RLn and RLp are different.

  In the example shown in FIG. 10, it is assumed that the voltage at the point a1 or a2 is measured when the time ta has elapsed from the start of charging, and the voltage at the point b is measured when the time tb has elapsed. As shown in FIG. 10, even when there is no significant change in the charging voltage at point b, the rising speed of the charging voltage waveform varies due to the influence of the time constant including the ground fault resistance. That is, at the points a1 and a2 where the time from the start of charging is short, the influence of the difference in resistance value of the ground fault resistance appears more greatly. Therefore, by using the voltage measured at two points (a1 + b or a2 + b) having different timings, the ground fault resistance can be detected even in a region having a large resistance value.

  Further, when the ratio of the two measured values (b / a1, b / a2) is used, the ground fault resistance value can be calculated without being affected by the applied voltage V0 as described below.

The change in the voltage Vc between the terminals of the detection capacitor C1 from the start of charging during the charging operation is expressed by the following equation.
Vc = V0 (1-exp ^{(-t / C. (R + RL))} ) (2)
V0: Applied voltage C: Capacitance of C1 R: Resistance value of charging circuit RL: Resistance value of ground fault resistance t: Elapsed time from start of charging

Further, the charging voltage Vc (ta) at the timing ta and the charging voltage Vc (tb) at the timing tb are respectively expressed by the following equations.
Vc (ta) = V0 (1-exp ^{(-ta / C. (R + RL))} ) (3)
Vc (tb) = V0 (1-exp ^{(-tb / C. (R + RL))} ) (4)

Therefore, the ratio of the two points is expressed by the following equation.
Vc (tb) / Vc (ta) = exp ^{(−tb / C · (R + RL))} / exp ^{(−ta / C · (R + RL))} (5)

  That is, if Vc (tb) / Vc (ta) is used, the influence of the applied voltage V0 can be eliminated. Further, since the capacitance C1 of C1 and the resistance value R of the charging circuit are known constant values, the ratio of Vc (tb) / Vc (ta) depends on the charging time t and the ground fault resistance value RL. Change. Therefore, when the charging times ta and tb are determined in advance, the ratio Vc (tb) / Vc (ta) can be associated with the ground fault resistance value RL.

  A ground fault resistance map representing such a correspondence relationship is held in the ground fault resistance map holding unit 42 shown in FIG. Therefore, in the operation shown in FIG. 1, the ground fault resistance calculation processing unit 41 is based on the ratio between the Vc1n measurement value (a) acquired at time t4 and the Vc1n measurement value (b) acquired at time t5. The resistance value of the ground fault resistance RLn can be acquired from the ground fault resistance map. Further, the resistance value of the ground fault resistance RLp can be acquired from the ground fault resistance map based on the ratio between the Vc1p measurement value (a) acquired at time t9 and the Vc1p measurement value (b) acquired at time t10. If the resistance values of the ground fault resistors RLn and RLp are known, the applied voltage on the secondary side of the circuit can also be estimated.

<Charging time reduction process>
A characteristic operation (2) and its modification in the embodiment of the present invention are shown in FIG. 12 (A) and FIG. 12 (B).

  The charging voltage (Vc1n, Vc1p) at each time point of the detection capacitor C1 varies greatly depending on the resistance values of the ground fault resistors RLn, RLp and the elapsed time. For example, when the ground fault resistance value is very small, the voltage rapidly increases even if the elapsed time from the start of charging is short. Therefore, when selecting components for each part of the circuit of the insulation state detection device 10 shown in FIG. 4, it is assumed that the ground fault resistance is very small (for example, 0 [Ω]) and the worst case voltage rise is assumed. Therefore, it is necessary to fully consider the pressure resistance characteristics of each component so as not to cause destruction or deterioration. As a result, an increase in apparatus cost is inevitable.

  On the other hand, in the operation shown in FIG. 1, the measurement values are acquired at two points whose timings are different from each other, so that it is possible to detect an abnormality at an early time point before the charging operation is completed. For example, when the voltage of the measurement value detected at time t4 in the Vc1n measurement section A2 in FIG. 1 is abnormally large, it can be recognized that the ground fault condition is apparent at that time. However, if the charging operation is continued as it is after the abnormality is detected at time t4, a very high voltage may be reached at time t5 when the charging ends.

  Each of the operation shown in FIG. 12A and the operation shown in FIG. 12B is for suppressing an abnormally high voltage from appearing in the secondary circuit of the insulation state detection device 10 in the worst case. Useful. For example, in the charging voltage Vc of the detection capacitor C1 shown in FIG. 11, the maximum value of the voltage under the worst condition can be lowered from Vmax to Vm2 or Vm3. Thereby, it becomes possible to reduce the pressure | voltage resistance required for each component, and to suppress the cost of components.

<Operation of FIG. 12A>
If it becomes Vc1n measurement area A2 shown in FIG. 1, it will progress to S22 from step S21 of FIG. 12 (A). For example, when Vc1n (a) is acquired at time t4 in FIG. 1, the process proceeds from S22 to S23 immediately thereafter.

  In step S23, Vc1n (a) acquired in S22 is compared with a predetermined upper limit value Vcmax. The upper limit value Vcmax may be determined based on the estimated voltage when the elapsed time from the start of charging under the condition that the ground fault resistance value is lower than an allowable value (for example, 300 [kΩ]) is TCna. When the condition of “Vc1n (a)> Vcmax” is satisfied, the ground fault is surely abnormal, so that the process proceeds to the next S24, and when the condition is not satisfied, the process proceeds to S25.

  In step S24, the charging operation in the section is immediately forcibly terminated. Therefore, in this case, the measurement value Vc1n (b) is not acquired. Further, since a correct ground fault resistance cannot be calculated with only one measured value Vc1n (a), an abnormal value is output as the ground fault resistance value.

  Moreover, when it becomes Vc1p measurement area A4 shown in FIG. 1, it will progress to S26 from step S25 of FIG. 12 (A). For example, when Vc1p (a) is acquired at time t9 in FIG. 1, the process proceeds from S26 to S27 immediately thereafter.

  In step S27, Vc1p (a) acquired in S26 is compared with a predetermined upper limit value Vcmax. The upper limit value Vcmax may be determined based on the estimated voltage when the elapsed time from the start of charging under the condition that the ground fault resistance value is lower than an allowable value (for example, 300 [kΩ]) is TCpa. When the condition of “Vc1p (a)> Vcmax” is satisfied, the ground fault is surely abnormal, so that the process proceeds to the next S28, and when the condition is not satisfied, the process returns to S21.

  In step S28, the charging operation in the section is immediately forcibly terminated. Therefore, in this case, the measurement value Vc1p (b) is not acquired. Further, since a correct ground fault resistance cannot be calculated with only one measured value Vc1p (a), an abnormal value is output as the ground fault resistance value.

<Operation of FIG. 12B>
The operation shown in FIG. 12B is different from FIG. 12A only in steps S24B and S28B. An operation different from that in FIG. 12A in FIG. 12B is described below.

  In step S24B, the time length TCnb and the charge time length of the section are shortened from the normal state. For example, when TCna and TCnb at the normal time are 0.1 seconds and 0.4 seconds, respectively, TCnb is set to 0. 0. The time is shortened to 2 seconds, and the charging time length is also shortened in accordance with the shortened TCnb. Further, when the timing of the measured value Vc1n (b) changes, it is necessary to switch the condition (constant) of the calculation formula for calculating the ground fault resistance, so a flag indicating that TCnb has been shortened is set.

  In step S28B, the time length TCpb and the charge time length of the section are shortened from the normal state. For example, when TCpa and TCpb in the normal time are 0.1 seconds and 0.4 seconds, respectively, TCpb is set to 0. 0. The time is shortened to 2 seconds, and the charging time is shortened in accordance with the shortened TCpb. Further, when the timing of the measured value Vc1p (b) changes, it is necessary to switch the condition (constant) of the calculation formula for calculating the ground fault resistance, so a flag indicating that TCpb has been shortened is set.

  In addition to the above two-point measurement mode, a normal measurement mode that calculates the value of the ground fault resistance based on the voltage measurement value obtained by measuring only one point (third time) in a situation where the measurement cycle can be long. It is also possible to switch between the two-point measurement mode and the normal measurement mode.

  As described above, according to the insulation state detection device according to the present embodiment, the ground fault resistance value calculation unit is based on the voltage measurement value at the first time and the voltage measurement value at the second time obtained by the measurement. Calculate the value of ground fault resistance. When charging the flying capacitor, the voltage between the terminals of the flying capacitor rises according to an exponential curve determined by the time constant of the charging circuit. This exponential function curve includes the influence of the electric charge flowing out from the Y capacitor. In the exponential function curve, the voltage rises with a large slope immediately after the start of charging, and changes to a gentle slope when a certain amount of time has passed. Then, at a relatively early timing after the start of charging, a change in the charging voltage appears remarkably according to a change in the resistance value of the ground fault resistance. In addition, the charging voltage when a certain amount of time has elapsed from the start of charging is greatly influenced by the power supply voltage, and the influence of the resistance value of the ground fault resistance and the influence of the electric charge flowing out from the Y capacitor is relatively small. Accordingly, the difference in the voltage measurement value at the first time and the voltage measurement value at the second time and the ratio thereof have a relatively large influence on the resistance value of the ground fault resistance. In other words, the correct ground fault resistance can be grasped by using the voltage measurement value at the first time and the voltage measurement value at the second time, even in a region where the resistance value of the ground fault resistance having a large influence of the Y capacitor is large. It becomes possible.

  Further, according to this insulation state detection device, by using constant data of a ground fault resistance map prepared in advance, it is possible to simply calculate from the ratio between the voltage measurement value at the first time and the voltage measurement value at the second time. The resistance value of the ground fault resistance can be calculated by simple conversion.

  Moreover, according to this insulation state detection apparatus, a relatively correct ground fault resistance value can be calculated even in a region where the resistance value of the ground fault resistance is large. For example, when the difference between the ground fault resistance (RLp) between the positive power line and the ground electrode and the ground fault resistance (RLn) between the negative power line and the ground electrode is large, the positive measurement is performed. The difference between the value and the negative measurement value increases. Then, the smaller of the positive measurement value and the negative measurement value, the greater the error factor due to the large ground fault resistance. Therefore, by correcting the smaller one of the positive measurement value and the negative measurement value and calculating using the corrected value, the detection error of the ground fault resistance can be reduced.

  Moreover, according to this insulation state detection apparatus, it is possible to prevent an abnormal increase in the charging voltage applied to the flying capacitor due to some cause. Therefore, it is not necessary to prepare a flying capacitor component having a particularly high breakdown voltage, and the cost of the component can be reduced.

DESCRIPTION OF SYMBOLS 10 Insulation state detection apparatus 11 Microcomputer 13 Positive side input terminal 14 Negative side input terminal 15 Grounding point 20 Input circuit 21 Output terminal 31-37 Wiring 41 Ground fault resistance calculation process part 42 Ground fault resistance map holding part 50 Car-mounted DC high voltage power supply 103 Grounding line 111 Positive power supply line 112 Negative power supply line C1 Detection capacitor (flying capacitor)
D1, D2, D3 Diode R01, R02, R1, R2, R3, R4, R5 Resistor RLp, RLn Ground fault resistor Yp, Yn Y capacitor S1, S2, S3, S4 Switching element

Claims (3)

A positive-side input terminal and a negative-side input terminal connected to a positive-side power line and a negative-side power line of a predetermined high-voltage DC power output, respectively, and the positive-side power line and the positive-side power line based on the charging voltage of the flying capacitor An insulation state detection device for grasping an insulation state between the negative power supply line and the ground as a ground fault resistance,
Each of the flying capacitors at a first time and a second time having different timings in one charging section when charging the flying capacitor via a predetermined energization path affected by the ground fault resistance. A charging voltage measuring unit for measuring the charging voltage of
A ground fault resistance value calculation unit that calculates a value of the ground fault resistance based on the voltage measurement value of the first time measured by the charging voltage measurement unit and the voltage measurement value of the second time;
An insulation state detection device comprising: The insulation state detection device according to claim 1,
The ground fault resistance value calculation unit obtains a plurality of constant data in a state in which the ratio of the voltage measurement value at the first time and the voltage measurement value at the second time is associated with the resistance value of the ground fault resistance. Hold ground fault resistance map,
An insulation state detection device comprising: An insulation state detection device according to claim 1 or 2,
Two points where the ground fault resistance value calculation unit calculates the value of the ground fault resistance based on the voltage measurement value at the first time measured by the charging voltage measurement unit and the voltage measurement value at the second time. Measurement mode and
The charging voltage measuring unit measures the charging voltage of the flying capacitor at a third time in the one charging period, and the ground fault resistance calculating unit is measured by the charging voltage measuring unit. A normal measurement mode for calculating the value of the ground fault resistance based on the voltage measurement value of the time;
An insulation state detection device comprising:

Images