Coil diagnostic method and coil diagnostic device

Abstract

This coil diagnosis method comprises: a voltage application step for applying an instantaneous voltage between wires of a coil (2); a response voltage measurement step for measuring a response voltage, which is the voltage across the wires of the coil (2); a determination index computation step for computing a determination index from the response voltage that changes with time; and a determination step for determining the presence or absence of a short circuit of the coil (2) using the determination index. The determination index is computed on the basis of the peak voltage and the zero-crossing time of the response voltage.

Description

[Technical Field]

[0001] The present disclosure relates to a coil diagnostic method and a coil diagnostic device. [Background technology]

[0002] In the coils that make up a rotating electric machine, an insulating coating is formed on the surface of the magnet wire to prevent adjacent magnet wires from shorting out. However, if the insulating coating has pinholes or is insufficiently thick, adjacent magnet wires will short out, degrading the performance of the rotating electric machine. A known method for detecting a short circuit is to apply an impulse voltage between the wires of the coil, measure the response voltage, and calculate a judgment index from the response voltage.

[0003] A method has been disclosed in which the point where the response voltage crosses a predetermined reference voltage is defined as a zero-cross time, and the time between two zero-cross times is used as a determination index to determine whether a short circuit has occurred (for example, Patent Document 1). [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 6849080 Summary of the Invention [Problem to be solved by the invention]

[0005] However, in the method of the patent document, when an offset component is superimposed on the response voltage, the time between zero crossing times varies, which causes a problem of degraded accuracy in the determination.

[0006] The present disclosure discloses a technique for solving the above-described problems, and aims to provide a coil diagnostic method and a coil diagnostic device that can reduce the influence of an offset component superimposed on a response voltage and improve determination accuracy. [Means for solving the problem]

[0007] The coil diagnosis method of the present disclosure includes a voltage application step of applying an instantaneous voltage between the wires of the coil, a response voltage measurement step of measuring a response voltage that is a voltage applied between the wires of the coil, and a judgment index calculation step of calculating a judgment index from the response voltage that changes over time. a determination step of determining whether or not the coil is short-circuited using the determination index, the determination index being calculated based on a peak voltage and a zero-cross time of the response voltage; The judgment index is calculated based on a peak voltage of the response voltage and a slope which is an increase rate of the response voltage at successive zero crossing times. It is something. The coil diagnosis method of the present disclosure includes a voltage application step of applying an instantaneous voltage between the wires of the coil, a response voltage measurement step of measuring a response voltage that is a voltage applied between the wires of the coil, and a judgment index calculation step of calculating a judgment index from the response voltage that changes over time. The method includes a judgment step of judging whether or not the coil is short-circuited using the judgment index, which is calculated based on the peak voltage and zero-cross time of the response voltage, and the judgment index is calculated based on the difference between successive peak voltages of the response voltage and the slope, which is the rate of increase of successive zero-cross times of the response voltage. The coil diagnosis method of the present disclosure includes a voltage application step of applying an instantaneous voltage between the wires of the coil, a response voltage measurement step of measuring a response voltage that is a voltage applied between the wires of the coil, and a judgment index calculation step of calculating a judgment index from the response voltage that changes over time. The method includes a judgment step of judging whether or not the coil is short-circuited using the judgment index, wherein the judgment index is calculated based on the peak voltage and zero-crossing time of the response voltage, and wherein, when sequence a is a sequence that increases by one from 1, and sequence b is a sequence that is the natural logarithm of the absolute value of the difference sequence of the sequence of the peak voltage of the response voltage counted from the second peak voltage onwards, the slope of sequence b with respect to sequence a is defined as slope p, sequence c is a sequence of the zero-crossing times of the response voltage counted from the second zero-crossing time onwards, and slope q is defined as γ, the value obtained by dividing slope p by slope q and multiplying the result by −1 is defined as the judgment index. The coil diagnosis method of the present disclosure includes a voltage application step of applying an instantaneous voltage between the wires of the coil, a response voltage measurement step of measuring a response voltage that is a voltage applied between the wires of the coil, and a judgment index calculation step of calculating a judgment index from the response voltage that changes over time. a determination step of determining whether or not the coil is short-circuited using the determination index, wherein the determination index is calculated based on the peak voltage and zero-crossing time of the response voltage, and wherein a sequence that increases by one from 1 is defined as sequence a, a sequence in which the zero-crossing times of the response voltage are counted from a second zero-crossing time onward is defined as sequence c, and a slope of the sequence c with respect to the sequence a is defined as slope q, and ω is defined as a value obtained by dividing π by the slope q, the determination index includes ω. It is something. The coil diagnosis method of the present disclosure includes a voltage application step of applying an instantaneous voltage between the wires of the coil, a response voltage measurement step of measuring a response voltage that is a voltage applied between the wires of the coil, and a judgment index calculation step of calculating a judgment index from the response voltage that changes over time. a determination step of determining whether or not the coil is short-circuited using the determination index, wherein the determination index is calculated based on the peak voltage and zero-crossing time of the response voltage, and wherein a sequence a is a sequence that increases by one from 1, and a sequence b is a sequence that is the natural logarithm of the absolute value of a difference sequence of a sequence in which the peak voltage of the response voltage is counted from a second peak voltage onward, where p is a slope of the sequence b with respect to the sequence a, c is a sequence in which the zero-crossing times of the response voltage are counted from a second zero-crossing time onward, and q is a slope of the sequence c with respect to the sequence a, where γ is a value obtained by dividing the slope p by the slope q and multiplying the result by −1, and ω is a value obtained by dividing pi by the slope q, the determination index is It is something. The coil diagnosis method of the present disclosure includes a voltage application step of applying an instantaneous voltage between the wires of the coil, a response voltage measurement step of measuring a response voltage that is a voltage applied between the wires of the coil, and a judgment index calculation step of calculating a judgment index from the response voltage that changes over time. the determination index is calculated based on the peak voltage and zero-crossing times of the response voltage, and wherein a sequence a is a sequence that increases by one from 1, and a sequence b is a sequence that is the natural logarithm of the absolute value of a difference sequence of the peak voltage of the response voltage counted from a second peak voltage onward, where p is a slope of the sequence b with respect to the sequence a, c is a sequence of the zero-crossing times of the response voltage counted from a second zero-crossing time onward, and q is a slope of the sequence c with respect to the sequence a, and L is an inductance of an equivalent circuit of the coil, R is a resistance of the coil, and C is a capacitance of the coil, γ is a value obtained by dividing the slope p by the slope q and multiplying the result by −1, and ω is a value obtained by dividing pi by the slope q, and the determination index includes LC, which is the reciprocal of the sum of the squares of γ and ω. The coil diagnosis method of the present disclosure includes a voltage application step of applying an instantaneous voltage between the wires of the coil, a response voltage measurement step of measuring a response voltage that is a voltage applied between the wires of the coil, and a judgment index calculation step of calculating a judgment index from the response voltage that changes over time. a judgment step of judging whether or not the coil is short-circuited using the judgment index, wherein the judgment index is calculated based on the peak voltage and zero-crossing times of the response voltage, and wherein a sequence a is a sequence that increases by one from 1, and a sequence b is a sequence that is the natural logarithm of the absolute value of a difference sequence of the peak voltage of the response voltage counted from a second peak voltage onward, where p is a slope of the sequence b with respect to the sequence a, c is a sequence of zero-crossing times of the response voltage counted from a second zero-crossing time onward, and q is a slope of the sequence c with respect to the sequence a, and L is an inductance of an equivalent circuit of the coil, R is a resistance of the coil, and C is a capacitance of the coil, where γ is a value obtained by dividing the slope p by the slope q and multiplying the result by -1, and ω is a value obtained by dividing pi by the slope q, and the judgment index includes RC, which is a value obtained by dividing γ by LC and multiplying it by 2. The coil diagnosis method of the present disclosure includes a voltage application step of applying an instantaneous voltage between the wires of the coil, a response voltage measurement step of measuring a response voltage that is a voltage applied between the wires of the coil, and a judgment index calculation step of calculating a judgment index from the response voltage that changes over time. a determination step for determining whether or not the coil is short-circuited using the determination index, wherein the determination index is calculated based on the peak voltage and zero-crossing time of the response voltage, and the determination index is calculated by: The sequence counted from a is defined as sequence c, the slope of sequence c relative to sequence a is defined as slope q, the inductance of the equivalent circuit of the coil is defined as L, the resistance of the coil is defined as R, and the capacitance of the coil is defined as C. Then, γ is defined as the value obtained by dividing the slope p by the slope q and multiplying the result by -1, and ω is defined as the value obtained by dividing pi by the slope q. LC is the reciprocal of the sum of the squares of γ and ω, and RC is the value obtained by dividing γ by LC and multiplying it by 2, and these are the judgment indicators. The coil diagnosis method of the present disclosure includes a voltage application step of applying an instantaneous voltage between the wires of the coil, a response voltage measurement step of measuring a response voltage that is a voltage applied between the wires of the coil, and a judgment index calculation step of calculating a judgment index from the response voltage that changes over time. The method includes a judgment step of judging whether or not the coil is short-circuited using the judgment index, wherein the judgment index is calculated based on the peak voltage and zero-crossing time of the response voltage, and when the base of the natural logarithm is e and the reciprocal of the time constant at which the response voltage decays to a magnitude of 1 / e is defined as γ, the judgment index includes either γ, the reciprocal of γ, or a power of γ. The coil diagnostic device of the present disclosure includes a voltage application unit that applies an instantaneous voltage between the wires of a coil, a response voltage measurement unit that measures a response voltage that is a voltage applied between the wires of the coil, a judgment index calculation unit that calculates a judgment index from the response voltage that changes over time, and a judgment unit that uses the judgment index to determine whether or not the coil is short-circuited, and the judgment index calculation unit calculates the judgment index based on a peak voltage and a zero-cross time of the response voltage. The judgment index calculation unit calculates the judgment index based on a peak voltage of the response voltage and a slope that is an increase rate of successive zero crossing times of the response voltage. It is something. The coil diagnostic device of the present disclosure includes a voltage application unit that applies an instantaneous voltage between the wires of a coil, a response voltage measurement unit that measures a response voltage, which is a voltage applied between the wires of the coil, a judgment index calculation unit that calculates a judgment index from the response voltage that changes over time, and a judgment unit that uses the judgment index to determine whether or not the coil is short-circuited, and the judgment index calculation unit calculates the judgment index based on the peak voltage and zero-cross time of the response voltage, and defines a sequence a that increases by one from 1, and calculates the peak voltage of the response voltage as When sequence b is the natural logarithm of the absolute value of the difference sequence of the sequence counted from after the second peak voltage, the slope of sequence b with respect to sequence a is slope p, sequence c is the sequence counted from the zero-cross times of the response voltage from after the second zero-cross time, and the slope of sequence c with respect to sequence a is slope q, the judgment index calculation unit calculates, as the judgment index, γ defined by the value obtained by dividing the slope p by the slope q and multiplying the result by −1, and ω defined by the value obtained by dividing pi by the slope q. The coil diagnostic device of the present disclosure includes a voltage application unit that applies an instantaneous voltage between the wires of a coil, a response voltage measurement unit that measures a response voltage that is a voltage applied between the wires of the coil, a judgment index calculation unit that calculates a judgment index from the response voltage that changes over time, and a judgment unit that uses the judgment index to judge whether or not the coil is short-circuited, and the judgment index calculation unit calculates the judgment index based on the peak voltage and zero-crossing time of the response voltage, and calculates the judgment index by calculating the natural logarithm of the absolute value of the difference sequence of the peak voltage of the response voltage counted from a second peak voltage onwards. When the slope of the sequence b with respect to the sequence a is defined as slope p, the sequence obtained by counting the zero-cross times of the response voltage from the second zero-cross time onwards is defined as sequence c, the slope of the sequence c with respect to the sequence a is defined as slope q, the inductance of the equivalent circuit of the coil is defined as L, the resistance of the coil is defined as R, and the capacitance of the coil is defined as C, the judgment index calculation unit defines γ as the value obtained by dividing the slope p by the slope q and multiplying the result by −1, and defines ω as the value obtained by dividing pi by the slope q, and calculates LC which is the reciprocal of the sum of the squares of γ and ω, and RC which is the value obtained by dividing γ by LC and multiplying it by 2, as the judgment index. [Effects of the Invention]

[0008] According to the coil diagnostic method of the present disclosure, even if an offset component is superimposed on the response voltage, the influence of the offset component can be reduced and the accuracy of determination can be improved. According to the coil diagnostic device of the present disclosure, even if an offset component is superimposed on the response voltage, the influence of the offset component can be reduced and the accuracy of determination can be improved. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a block diagram showing a schematic configuration of a coil diagnostic device according to a first embodiment. [Figure 2] 3 is an explanatory diagram of an equivalent circuit when the coil and the coil diagnostic device according to the first embodiment are connected. FIG. [Figure 3] FIG. 4 is an explanatory diagram of a response voltage during a coil test according to the first embodiment. [Figure 4] 4A is a diagram illustrating the peak voltage of the response voltage according to embodiment 1. FIG. 4B is a diagram illustrating the peak voltage of the response voltage (with offset) according to embodiment 1. [Figure 5] 5A is a diagram illustrating a difference between successive peak voltages of the response voltage according to embodiment 1. FIG. 5B is a diagram illustrating a difference (with offset) between successive peak voltages of the response voltage according to embodiment 1. [Figure 6] Fig. 6A is a diagram illustrating the zero cross times of the response voltage according to embodiment 1. Fig. 6B is a diagram illustrating the zero cross times (with offset) of the response voltage according to embodiment 1. [Figure 7] 4 is an explanatory diagram of how to determine the peak voltage and zero crossing time of the response voltage according to the first embodiment. FIG. [Figure 8] FIG. 2 is an explanatory diagram of a sample coil that was tested in actual measurements according to the first embodiment. [Figure 9] FIG. 10 is an explanatory diagram of actual measurement values ​​of a determination index (interval between zero cross times) in a comparative example according to the first embodiment. [Figure 10]FIG. 4 is an explanatory diagram of measured values ​​of decision indices (γ, ω) according to the first embodiment. [Figure 11] FIG. 4 is an explanatory diagram of actual measurement values ​​of decision indices (LC, RC) according to the first embodiment. [Figure 12] 3 is a flowchart of a coil diagnostic method according to the first embodiment. [Figure 13] 3 is a diagram showing an example of a hardware configuration of a diagnosis unit of the coil diagnosis device according to the first embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0010] Embodiment 1 The first embodiment relates to a coil diagnostic method that includes a voltage application step of applying an instantaneous voltage between the wires of the coil, a response voltage measurement step of measuring a response voltage, which is the voltage applied between the wires of the coil, a judgment index calculation step of calculating a judgment index from the response voltage that changes over time, and a judgment step of using the judgment index to determine whether or not the coil is short-circuited, and the judgment index is calculated based on the peak voltage and zero-cross time of the response voltage. Moreover, the first embodiment relates to a coil diagnostic device that includes a voltage application unit that applies an instantaneous voltage between the wires of the coil, a response voltage measurement unit that measures a response voltage, which is a voltage applied between the wires of the coil, a judgment index calculation unit that calculates a judgment index from the response voltage that changes over time, and a judgment unit that uses the judgment index to determine whether or not the coil is short-circuited, and the judgment index calculation unit calculates the judgment index based on the peak voltage and zero-cross time of the response voltage.

[0011] The coil diagnostic method and coil diagnostic device according to the first embodiment will be described below with reference to FIG. 1, a block diagram showing a schematic configuration of the coil diagnostic device; FIG. 2, an explanatory diagram of an equivalent circuit when a coil and the coil diagnostic device are connected; FIG. 3, an explanatory diagram of the response voltage during a coil test; FIGS. 4A and 4B, explanatory diagrams of the peak voltage of the response voltage; FIGS. 5A and 5B, explanatory diagrams of the difference between successive peak voltages of the response voltage; FIGS. 6A and 6B, explanatory diagrams of the zero-cross times of the response voltage; FIG. 7, an explanatory diagram of how the peak voltage of the response voltage and the zero-cross times are determined; FIG. 8, an explanatory diagram of a sample coil used as a test subject in actual measurements; FIG. 9, an explanatory diagram of the measured values ​​of the judgment index (the interval between the zero-cross times) of a comparative example; FIG. 10, an explanatory diagram of the measured values ​​of the judgment indexes (γ, ω); In each drawing, the same or corresponding parts are denoted by the same reference numerals.

[0012] First, the configuration of a coil diagnostic device 1 according to the first embodiment will be described with reference to Fig. 1. Note that the coil 2 is a diagnostic target and is not a component of the coil diagnostic device 1, but is closely related in function and will be described together. A coil diagnostic device 1 that diagnoses the soundness (whether a short circuit has occurred) of a coil 2 includes a voltage application unit 3, a measurement unit 4, and a diagnostic unit 5. The diagnostic unit 5 includes an index calculation unit 51 and a determination unit 52.

[0013] First, the functions of the components constituting the coil diagnosis device 1 will be described. The voltage application unit 3 has a function of applying an instantaneous voltage between the wires of the coil 2 to be diagnosed. The measurement unit 4 has a function of measuring a response voltage (V), which is the voltage applied between the wires of the coil 2. The index calculation unit 51 has a function of calculating a judgment index from the response voltage (V), which changes over time. The judgment unit 52 has a function of diagnosing the soundness of the coil 2 from the judgment index calculated by the index calculation unit 51. The details of the functions of each unit constituting the coil diagnostic device 1 will be explained in order. When the index calculation unit 51 and the determination unit 52 are collectively referred to, they will be referred to as the diagnostic unit 5.

[0014] Next, an equivalent circuit of the coil 2 connected to the coil diagnostic device 1 in the first embodiment and an internal circuit configuration of the voltage application unit 3 of the coil diagnostic device 1 will be described with reference to FIG. First, the equivalent circuit of the coil 2 will be described. The equivalent circuit of coil 2 is represented by a series circuit of coil 2a (inductance value L) and resistor 2b (resistance value R), which is the resistance of coil 2, connected in parallel with capacitor 2c (capacitance value C), which is the capacitance of coil 2. In the following description, for ease of understanding, the names and reference numerals of components (for example, coil 2a) will be omitted, and only symbols of physical quantities (for example, L) will be used as appropriate.

[0015] Next, the internal circuit configuration of the voltage application unit 3 will be described. The voltage application unit 3 is made up of a DC power supply 3a, a resistor 3b, an internal capacitor 3c, a switch 3d, and a commutator 3e. The following description will be given assuming that the capacitance value of the internal capacitor 3c is C1.

[0016] Next, the behavior when a voltage is momentarily applied from the voltage application unit 3 to the coil 2 will be described. In the state shown in FIG. 2, the switch 3d is connected in a direction that connects the internal capacitor 3c to the DC power supply 3a, and the internal capacitor 3c is charged via the resistor 3b. In response to a command to start the test, the switch 3d is switched so that the internal capacitor 3c is connected from the DC power supply 3a side to the coil 2 side. Immediately after the switch 3d is switched, the current flowing through the commutator 3e is in the forward direction, forming a resonant circuit with the combined capacitance of the coil 2a, resistor 2b, capacitor 2c and internal capacitor 3c. After that, from the time when the current flowing through commutator 3e becomes 0 for the first time, the current flowing through commutator 3e is in the opposite direction, so no current flows on the voltage application unit 3 side of coil diagnostic device 1. As a result, the equivalent circuit of coil 2 forms a resonant circuit of coil 2a, resistor 2b, and capacitor 2c. The response voltage in the resonant circuit after time 0 depends on the inductance value L, resistance value R, and capacitance value C of coil 2.

[0017] Next, the response voltage of the equivalent circuit of FIG. 2 in the first embodiment will be described with reference to FIG. 3 shows, as peak voltages, a first peak voltage 71, a second peak voltage 72, a third peak voltage 73, a fourth peak voltage 74, a fifth peak voltage 75, a sixth peak voltage 76, and a seventh peak voltage 77. Also, as zero cross times, a first zero cross time 81, a second zero cross time 82, a third zero cross time 83, a fourth zero cross time 84, a fifth zero cross time 85, and a sixth zero cross time 86 are shown. If the time when the current flowing through the commutator 3e becomes 0 is t=0 and the voltage at that time is V(0) (hereinafter, subscripts will be represented by "( )"), the response voltage is given by equation (1). In the formulas, the subscripts "( )" are omitted to simplify the formulas.

[0018]

number

[0019] In the formula (1), γ, ω, and ω1 are expressed by the formulas (2) to (4).

[0020]

number

[0021]

number

[0022]

number

[0023] The response voltage at times t ≥ 0 is a damped oscillation that depends on γ and ω, where γ is the reciprocal of the time constant, which is the time required for the response voltage to become 1 / e, and ω is the frequency at which the response voltage oscillates. Hereinafter, the response voltage, which is the voltage applied between the lines of coil 2, will be considered for times t≧0. In other words, the peak voltage will be considered to be the peak voltage at times after second peak voltage 72, and the zero-crossing times will be considered to be the zero-crossing times at times after second zero-crossing time 82.

[0024] Transforming equation (1), the relationship between γ, ω and L, R, and C of coil 2 to be diagnosed is expressed as equations (5) and (6). LC is the product of L and C of coil 2, and RC is the product of R and C of coil 2.

[0025]

number

[0026]

number

[0027] In the first embodiment, γ in equation (2), ω in equation (3), LC in equation (5), and RC in equation (6) are used as judgment indices for diagnosing the soundness of the coil 2. Then, first, γ and ω are calculated by analyzing the response waveform, and then LC and RC are calculated from equations (5) and (6) using these γ and ω.

[0028] First, the method for calculating γ will be described. To calculate γ, the voltage when the response voltage reaches its maximum value, that is, the peak voltage, is used. The response voltage reaches a peak voltage at each time t=π / ω when cosωt reaches its peak. The attenuation ratio between successive peak voltages is V(t=(n+1)(π / ω)) / V(t=n(π / ω)), which, calculated from equation (1), is -e^(-γπ / ω). Therefore, if consecutive peak voltages are considered as a sequence, it is a geometric progression with a common ratio of -e^(-γπ / ω), so γ can be calculated by finding the common ratio when consecutive peak voltages are considered as a sequence.

[0029] 4A and 4B show the peak voltage of the response voltage in embodiment 1. Fig. 4A shows the case where no offset component is superimposed, and Fig. 4B shows the case where an offset component is superimposed. In Figure 4A, when the magnitude (absolute value) of the second peak voltage 72 is A, the magnitude (absolute value) of the third peak voltage 73 is Ae^(-γπ / ω), the magnitude (absolute value) of the fourth peak voltage 74 is Ae^(-2γπ / ω), and the magnitude (absolute value) of the fifth peak voltage 75 is Ae^(-3γπ / ω). In FIG. 4B, α is added to the magnitude of each peak voltage due to the effect of the offset component.

[0030] 5A and 5B show the difference between successive peak voltages of the response voltage in embodiment 1. Fig. 5A shows the case where no offset component is superimposed, and Fig. 5B shows the case where the offset component is superimposed. In Figure 5A, when the magnitude (absolute value) of the difference between the second peak voltage 72 and the third peak voltage 73 is B, the magnitude (absolute value) of the difference between the third peak voltage 73 and the fourth peak voltage 74 is Be^(-γπ / ω), the magnitude (absolute value) of the difference between the fourth peak voltage 74 and the fifth peak voltage 75 is Be^(-2γπ / ω), and the magnitude (absolute value) of the difference between the fifth peak voltage 75 and the sixth peak voltage 76 is Be^(-3γπ / ω). In FIG. 5B, it can be seen that even if there is an effect of the offset component, the magnitude of the difference between the peak voltages does not change.

[0031] In actual measurements, offset components may be superimposed, and consecutive peak voltages may not follow a geometric progression. Even when offset components are superimposed, the effects of the offset components can be offset by taking the difference between consecutive peak voltages, so the differences between consecutive peak voltages follow a geometric progression. In other words, γ can be calculated by finding the difference between consecutive peak voltages.

[0032] Next, a method for calculating ω will be described. To calculate ω, the time when the response voltage becomes V=0, i.e., the zero-cross time, is used. From equation (1), the response voltage has a zero-cross time every time t=π / ω. In other words, if consecutive zero-cross times are considered as a sequence, it is an arithmetic progression with a common difference of π / ω, so ω can be calculated from the interval between consecutive zero-cross times.

[0033] 6A and 6B show the zero cross times of the response voltage in embodiment 1. Fig. 6A shows the case where no offset component is superimposed, and Fig. 6B shows the case where the offset component is superimposed. FIG. 6A shows that the time interval between the second zero cross time 82 and the third zero cross time 83, the time interval between the third zero cross time 83 and the fourth zero cross time 84, the time interval between the fourth zero cross time 84 and the fifth zero cross time 85, and the time interval between the fifth zero cross time 85 and the sixth zero cross time 86 are each π / ω. In FIG. 6B, β1 to β4 are added to each time interval due to the influence of the offset component.

[0034] In actual measurements, offset components may be superimposed, and consecutive zero-crossing times may not form an arithmetic progression. When offset components are superimposed, cycles larger and smaller than the tolerance π / ω appear alternately, so the effects of the offset components can be reduced by performing averaging processing. When offset components are superimposed, large and small periods alternate, so it is preferable to use periods of even minutes for averaging processing, that is, it is preferable to use zero cross times of odd minutes for calculation.

[0035] If the zero crossing times are counted from the second zero crossing time onwards and are denoted as {c(n)}: c(1), c(2), ..., c(n), then the intervals between successive zero crossing times are c(2)-c(1), c(3)-c(2), ..., c(n)-c(n-1). The simple average of the intervals between zero crossing times is {c(2)-c(1)+c(3)-c(2)+...+c(n)-c(n-1)} / (n-1)=(c(n)-c(1)) / (n-1), which is a value that does not depend on the intermediate zero crossing times. Therefore, a different method is used for averaging.

[0036] When there is an arithmetic sequence {z(n)}, the slope of {z(n)} for a sequence (1, 2, ..., n) that increases by one is equal to the common difference of {z(n)}. From this, it can be seen that by calculating the slope of the target sequence for the sequence (1, 2, ..., n), it is possible to average the intervals of the target sequence. The least squares method is used to calculate the slope.

[0037] If the variance of the sequence {x(n)} is σ(x) and the covariance of the sequences {x(n)} and {y(n)} is σ(xy), then the slope of {y(n)} with respect to {x(n)} can be expressed as σ(xy) / σ(x) using the least squares method. Let {a(n)}: 1, 2, ..., n, and the slope q of {c(n)} with respect to {a(n)} is given by equation (7), where a(ave) and c(ave) are the simple averages of {a(n)} and {c(n)}, respectively.

[0038]

number

[0039] Since the slope q corresponds to the tolerance π / ω after averaging, ω is expressed by equation (8).

[0040]

number

[0041] The same can be said for the common ratio of the peak voltages, -e^(-γπ / ω). A geometric progression becomes an arithmetic progression by taking its logarithm, so if the peak voltages are counted from the second peak voltage onwards and are {p(n)}: p(1), p(2), ..., p(n+1), then the natural logarithm of the absolute values ​​of this difference progression, {b(n)}: ln|p(1)-p(2)|, ln|p(2)-p(3)|, ..., ln|p(n)-p(n+1)|, is an arithmetic progression, and the common difference is -γπ / ω. The slope p of {b(n)} with respect to {a(n)} is given by equation (9), where b(ave) is the simple average of {b(n)}.

[0042]

number

[0043] Since the slope p corresponds to the tolerance −γπ / ω after averaging, γ is expressed by equation (10).

[0044]

number

[0045] Note that b(i)-b(ave) is a dimensionless quantity, and c(i)-c(ave) is the dimension of time, so the dimension of γ is the same as the dimension of frequency (T- 1 )

[0046] Here, γ, ω, LC, and RC, which are the judgment indices for diagnosing the soundness of the coil 2, and their prerequisites will be described.

[0047]

number

[0048]

number

[0049]

number

[0050]

number

[0051]

number

[0052] The method for determining the peak voltage and the zero crossing time of the response voltage in the first embodiment will be described with reference to FIG. 7, p1 to p6 correspond to the previously described second peak voltage 72 to seventh peak voltage 77. Furthermore, c1 to c5 correspond to the previously described second zero cross time 82 to sixth zero cross time 86.

[0053] Let {c(n)} be the zero-crossing times counting from the second zero-crossing time onwards: c(1), c(2), ..., c(n), and let {p(n)} be the peak voltages counting from the second peak voltage onwards: p(1), p(2), ..., p(n+1). Then, {b(n)} is: ln|p(1)-p(2)|, ln|p(2)-p(3)|, ..., ln|p(n)-p(n+1)|. Here, to ensure that {b(n)} and {c(n)} have the same number of items, the number of peak voltages is one more than the number of zero-crossing times. Figure 7 shows the case where n = 5.

[0054] FIG. 8 is a diagram showing the configuration of a sample coil that was tested in the actual measurement according to the first embodiment. The three-phase coils, U-phase coil 21, V-phase coil 22, and W-phase coil 23, are Y-connected, with four coils per phase connected in parallel. Of the four V-phase coils, a sample coil was produced in which the insulating coating of one of the magnet wires constituting the coil was partially removed, allowing adjacent magnet wires to be short-circuited at one point.

[0055] The response voltage between the U and V phases, V(UV), the response voltage between the V and W phases, V(VW), and the response voltage between the W and U phases, V(WU), were measured for the sample coil. For V(UV) and V(VW), the response voltage was also measured when the adjacent magnet wire was short-circuited in one place. Note that to distinguish the state where there is a short circuit in one place in the V phase from the normal state, the state is referred to as V(UV*) and V(V*W). With the rotor assembled in the sample coil, the rotor phase was rotated in 30-degree electrical increments and the response voltage was measured for each. This measurement was repeated until the rotor was rotated 180 degrees, and measurements were taken with variations in response voltage due to the rotor phase included.

[0056] 9 shows actual measured values ​​when the interval between zero-crossing times is used as the judgment index as a comparative example for embodiment 1. In Fig. 9, the horizontal axis represents the sample number, and the vertical axis represents the normalized value of c(2)-c(1). In FIG. 9, a circle indicates no short circuit, a triangle indicates a short circuit (V(UV*)), and a square indicates a short circuit (V(V*W)). For each measured value in Figure 9, counting from the second zero-crossing time, we designated c(1), c(2), ... and calculated c(2)-c(1) to be the interval between zero-crossing times. We also normalized this by dividing it by the average value of all measured values ​​without short circuits and subtracting 1, so that the average value without short circuits would be 0%.

[0057] Actual measurements without short circuits include measurements of V(UV), V(VW), and V(WU) without short circuits. It can be seen that the intervals of V(UV*) with and without a short circuit are distributed in the region of -4±2%, and the intervals of V(V*W) with and without a short circuit are distributed in the region of -6±2%.As can be seen from equation (3), the smaller the inductance L of coil 2, the larger the frequency ω, and therefore the smaller the intervals of the zero-crossing times. Furthermore, the fact that the interval between zero crossings is smaller when there is a short circuit compared to when there is no short circuit is consistent with the fact that when adjacent magnet wires are shorted, the shorted magnet wires act as a differentially connected transformer, reducing the inductance of coil 2. In the comparative example, the vertical axis of FIG. 9 shows that the region without a short circuit and the region with a short circuit overlap, making it impossible to determine the soundness of the coil (whether or not there is a short circuit).

[0058] Fig. 10 shows the measured values ​​of γ and ω, which are indicators for determining the soundness of coil 2 in embodiment 1. In Fig. 10, ○ indicates no short circuit, △ indicates a short circuit (V(UV*)), and □ indicates a short circuit (V(V*W)). For each measured value in Figure 10, γ and ω were calculated from the peak voltage and zero-crossing time using the least squares method according to equations (10) and (8).The values ​​were also normalized by dividing by the average value of all measured values ​​without short circuits and subtracting 1 so that the average value without short circuits was 0%.

[0059] Actual measurements without short circuits include measurements of V(UV), V(VW), and V(WU) without short circuits. It can be seen that the ω without short circuit is distributed in the range of 0±2%, the ω of V(UV*) with short circuit is 5±2%, and the ω of V(V*W) with short circuit is 11±2%. The reason why V(UV*) with a short circuit and V(V*W) with a short circuit are distributed further apart than the area without a short circuit is that the sample measured was a concentrated winding stator, and the shorted V-phase coil was adjacent to the W-phase coil but not to the U-phase, so the short circuit in the V-phase coil also affected the W-phase coil.

[0060] As can be seen from equation (3), the smaller the coil inductance L, the larger the frequency ω. Furthermore, the fact that ω, which is the frequency when there is a short circuit, is larger compared to when there is no short circuit is consistent with the fact that when adjacent magnet wires are shorted, the shorted magnet wires act as a differentially connected transformer, reducing the inductance of the coil. In the method of the first embodiment, there is no overlap between the region without a short circuit and the region with a short circuit, so it is clear that it is possible to determine the soundness of the coil (presence or absence of a short circuit).

[0061] Fig. 11 shows the measured values ​​of LC and RC, which are indicators for determining the soundness of coil 2 in embodiment 1. In Fig. 11, ○ indicates no short circuit, △ indicates a short circuit (V(UV*)), and □ indicates a short circuit (V(V*W)). For each measured value in Figure 11, γ and ω were calculated from the peak voltage and zero-crossing time using the least squares method according to equations (10) and (8), and LC and RC were calculated from equations (5) and (6) using these γ and ω. In addition, the average value of all measured values ​​without short circuits was divided by the average value of all measured values ​​without short circuits and then subtracted 1, thereby normalizing the average value of no short circuits to 0%. Actual measurements without short circuits include measurements of V(UV), V(VW), and V(WU) without short circuits. It can be seen that the LC without short circuit is distributed in the range of 0±4%, the LC of V(UV*) with short circuit is distributed in the range of -7±4%, and the LC of V(V*W) with short circuit is distributed in the range of -9±4%.

[0062] The distribution of V(UV*) with a short circuit and V(V*W) with a short circuit is greater than that of the area without a short circuit. The reason for this is that the sample coil used for measurement was a concentrated winding stator, and the shorted V-phase coil was adjacent to the W-phase coil but not to the U-phase, so the short circuit in the V-phase coil also affected the W-phase coil. Furthermore, the fact that LC is smaller in a short-circuited state compared to a state without a short circuit does not contradict the fact that the short-circuited magnet wires act as a differentially connected transformer due to the short circuit between adjacent magnet wires, thereby reducing the inductance of the coil. In the method of embodiment 1, the broken line (H) in FIG. 11 serves as the boundary, and there is no overlap between the area without a short circuit and the area with a short circuit, so it is clear that it is possible to determine the soundness of the coil (presence or absence of a short circuit).

[0063] Next, the coil diagnosis method will be summarized based on the flowchart in FIG. The coil diagnostic method according to the first embodiment comprises the following steps 1 (S01) to 4 (S04). In step 1 (S01), a voltage application step, a voltage is applied between the wires of the coil 2 instantaneously. In step 2 (S02), a response voltage measurement step, a response voltage, which is a voltage applied between the wires of coil 2, is measured. In step 3 (S03), an index calculation step, a judgment index is calculated from the response voltage that changes with time. In the determination step of step 4 (S04), it is determined whether or not the coil is short-circuited based on the determination index.

[0064] The coil diagnosis method of embodiment 1 includes a voltage application step of applying an instantaneous voltage between the wires of the coil, a response voltage measurement step of measuring a response voltage which is the voltage applied between the wires of the coil, a judgment index calculation step of calculating a judgment index from the response voltage which changes over time, and a judgment step of using the judgment index to determine whether or not the coil is short-circuited, and the judgment index is calculated based on the peak voltage and zero-cross time of the response voltage. More specifically, the judgment index is calculated based on the difference between the peak voltages of successive response voltages and the slope, which is the rate of increase of the zero crossing times of successive response voltages.

[0065] In the first embodiment, the use of γ, ω, LC, and RC as the judgment indices for diagnosing the soundness of the coil 2 and the calculation method of these judgment indices have been described. As the judgment indices for diagnosing the soundness of the coil 2, modifications of these judgment indices (for example, the reciprocal of γ or the power of γ) can be included as judgment indices.

[0066] As a judgment index for diagnosing the soundness of the coil 2, any one of γ, ω, LC, and RC can be used alone, or two or more of them can be used in combination. Furthermore, even when the judgment index includes either γ calculated by equation (10) or ω calculated by equation (8), or when either γ or ω is included in the process of calculating the judgment index, the influence of an offset component superimposed on the response voltage can be reduced.

[0067] When the calculation result obtained by replacing {b(n)} in equation (10) with {b(n)*}:ln|p(1)|, ln|p(2)|, ..., ln|p(n)| is defined as γ*, the calculation result obtained by replacing γ in equation (5) with γ* is defined as LC*, and the calculation result obtained by replacing γ in equation (6) with γ* is defined as RC*, LC* and RC* include ω calculated by equation (8) in their calculation process. Therefore, either LC* or RC* alone, or a combination of two or more of γ*, ω, LC*, and RC* can be used as a judgment index.

[0068] Furthermore, when the calculation result obtained by replacing ω in equation (10) with π(n-1) / (c(n)-c(1)) is defined as ω*1, the calculation result obtained by replacing ω in equation (5) with ω*1 is defined as LC*1, and the calculation result obtained by replacing ω in equation (6) with ω*1 is defined as RC*1, LC*1 and RC*1 include γ calculated by equation (10) in their calculation process. Therefore, either LC*1 or RC*1 alone, or a combination of two or more of γ, ω*1, LC*1, and RC*1 can be used as a judgment index.

[0069] For example, the judgment index is calculated based on the peak voltage of the response voltage and the slope, which is the rate of increase of the zero crossing times of successive response voltages.

[0070] An example of hardware of the diagnosis unit 5 of the coil diagnosis device 1 is shown in FIG. The diagnosis unit 5 of the coil diagnosis device 1 is composed of a processor 100 and a storage device 101. The storage device 101 includes a volatile storage device such as a random access memory and a non-volatile auxiliary storage device such as a flash memory, both of which are not shown. Also, a hard disk auxiliary storage device may be provided instead of flash memory. Processor 100 executes a program input from storage device 101. In this case, the program is input from the auxiliary storage device to processor 100 via a volatile storage device. Processor 100 may output data such as calculation results to the volatile storage device of storage device 101, or may store data in the auxiliary storage device via the volatile storage device.

[0071] As described above, the coil diagnostic method of embodiment 1 can reduce the influence of an offset component superimposed on the response voltage, thereby improving the accuracy of determination. Also, the coil diagnostic device of embodiment 1 can reduce the influence of an offset component superimposed on the response voltage, thereby improving the accuracy of determination.

[0072] Although the present disclosure describes exemplary embodiments, the various features, aspects, and functions described in the embodiments are not limited to application to a particular embodiment, but may be applied to the embodiments alone or in various combinations. Therefore, countless variations not illustrated are contemplated within the scope of the technology disclosed in this specification, including, for example, the modification, addition, or omission of at least one component. [Explanation of symbols]

[0073] 1 coil diagnostic device, 2 coil, 2a coil, 2b resistor, 2c capacitor, 3 voltage application unit, 3a DC power supply, 3b resistor, 3c internal capacitor, 3d switch, 3e commutator, 4 measurement unit, 5 diagnosis unit, 51 index calculation unit, 52 judgment unit, 71 first peak voltage, 72 second peak voltage, 73 third peak voltage, 74 fourth peak voltage, 75 fifth peak voltage, 76 sixth peak voltage, 77 seventh peak voltage, 81 first zero cross time, 82 second zero cross time, 83 third zero cross time, 84 fourth zero cross time, 85 fifth zero cross time, 86 sixth zero cross time, 21 U-phase coil, 22 V-phase coil, 23 W-phase coil, 100 processor, 101 storage device.

Claims

1. a voltage application step of applying a momentary voltage between the wires of the coil; a response voltage measuring step of measuring a response voltage which is a voltage applied between the wires of the coil; a determination index calculation step of calculating a determination index from the response voltage that changes over time; a determination step of determining whether or not the coil is short-circuited using the determination index, the determination index is calculated based on a peak voltage and a zero-cross time of the response voltage; A coil diagnostic method in which the judgment index is calculated based on a peak voltage of the response voltage and a slope, which is an increase rate of successive zero crossing times of the response voltage.

2. A voltage application step of applying a momentary voltage between the wires of a coil; a response voltage measuring step of measuring a response voltage which is a voltage applied between the wires of the coil; a determination index calculation step of calculating a determination index from the response voltage that changes over time; a determination step of determining whether or not the coil is short-circuited using the determination index, the determination index is calculated based on a peak voltage and a zero-cross time of the response voltage; A coil diagnostic method in which the judgment index is calculated based on the difference between the peak voltages of successive response voltages and the slope, which is the rate of increase of the zero-cross times of successive response voltages.

3. A voltage application step of applying a momentary voltage between the wires of a coil; a response voltage measuring step of measuring a response voltage which is a voltage applied between the wires of the coil; a determination index calculation step of calculating a determination index from the response voltage that changes over time; a determination step of determining whether or not the coil is short-circuited using the determination index, the determination index is calculated based on a peak voltage and a zero-cross time of the response voltage; Let a be a sequence of numbers that starts from 1 and increases by 1, When a sequence b is a sequence that is the natural logarithm of the absolute value of the difference sequence of the sequence counting the peak voltage of the response voltage from the second peak voltage onward, the slope of the sequence b with respect to the sequence a is defined as a slope p, A sequence obtained by counting the zero cross times of the response voltage from the second zero cross time onward is defined as a sequence c, When the gradient of the sequence c with respect to the sequence a is defined as gradient q, If the value obtained by dividing the slope p by the slope q and multiplying the result by -1 is defined as γ, then A diagnostic method for a coil, wherein the judgment index includes the γ.

4. A voltage application step of applying a momentary voltage between the wires of a coil; a response voltage measuring step of measuring a response voltage which is a voltage applied between the wires of the coil; a determination index calculation step of calculating a determination index from the response voltage that changes over time; a determination step of determining whether or not the coil is short-circuited using the determination index, the determination index is calculated based on a peak voltage and a zero-cross time of the response voltage; Let a be a sequence of numbers that starts from 1 and increases by 1, A sequence obtained by counting the zero cross times of the response voltage from the second zero cross time onward is defined as a sequence c, When the gradient of the sequence c with respect to the sequence a is defined as gradient q, If we define ω as the value obtained by dividing pi by the slope q, then A diagnostic method for a coil in which the judgment index includes ω.

5. A voltage application step of applying a momentary voltage between the wires of a coil; a response voltage measuring step of measuring a response voltage which is a voltage applied between the wires of the coil; a determination index calculation step of calculating a determination index from the response voltage that changes over time; a determination step of determining whether or not the coil is short-circuited using the determination index, the determination index is calculated based on a peak voltage and a zero-cross time of the response voltage; Let a be a sequence of numbers that starts from 1 and increases by 1, When a sequence b is a sequence that is the natural logarithm of the absolute value of the difference sequence of the sequence counting the peak voltage of the response voltage from the second peak voltage onward, the slope of the sequence b with respect to the sequence a is defined as a slope p, A sequence obtained by counting the zero cross times of the response voltage from the second zero cross time onward is defined as a sequence c, When the gradient of the sequence c with respect to the sequence a is defined as gradient q, If the value obtained by dividing the slope p by the slope q and multiplying the result by -1 is defined as γ, and the value obtained by dividing the constant π by the slope q is defined as ω, then A coil diagnostic method in which the judgment indexes are the γ and ω.

6. A voltage application step of applying a momentary voltage between the wires of a coil; a response voltage measuring step of measuring a response voltage which is a voltage applied between the wires of the coil; a determination index calculation step of calculating a determination index from the response voltage that changes over time; a determination step of determining whether or not the coil is short-circuited using the determination index, the determination index is calculated based on a peak voltage and a zero-cross time of the response voltage; Let a be a sequence of numbers that starts from 1 and increases by 1, When a sequence b is a sequence that is the natural logarithm of the absolute value of the difference sequence of the sequence counting the peak voltage of the response voltage from the second peak voltage onward, the slope of the sequence b with respect to the sequence a is defined as a slope p, A sequence obtained by counting the zero cross times of the response voltage from the second zero cross time onward is defined as a sequence c, The gradient of the sequence c with respect to the sequence a is defined as gradient q, When the inductance of the equivalent circuit of the coil is L, the resistance of the coil is R, and the capacitance of the coil is C, The value obtained by dividing the slope p by the slope q and multiplying the result by −1 is defined as γ, and the value obtained by dividing the ratio π of the circumference of a circle to its circumference by the slope q is defined as ω, A coil diagnostic method in which the judgment index includes LC, which is the inverse of the sum of the squares of γ and ω.

7. A voltage application step of applying a momentary voltage between the wires of a coil; a response voltage measuring step of measuring a response voltage which is a voltage applied between the wires of the coil; a determination index calculation step of calculating a determination index from the response voltage that changes over time; a determination step of determining whether or not the coil is short-circuited using the determination index, the determination index is calculated based on a peak voltage and a zero-cross time of the response voltage; Let a be a sequence of numbers that starts from 1 and increases by 1, When a sequence b is a sequence that is the natural logarithm of the absolute value of the difference sequence of the sequence counting the peak voltage of the response voltage from the second peak voltage onward, the slope of the sequence b with respect to the sequence a is defined as a slope p, A sequence obtained by counting the zero cross times of the response voltage from the second zero cross time onward is defined as a sequence c, The gradient of the sequence c with respect to the sequence a is defined as gradient q, When the inductance of the equivalent circuit of the coil is L, the resistance of the coil is R, and the capacitance of the coil is C, The value obtained by dividing the slope p by the slope q and multiplying the result by −1 is defined as γ, and the value obtained by dividing the ratio π of the circumference of a circle to its circumference by the slope q is defined as ω, A coil diagnostic method in which the judgment index includes RC, which is a value obtained by dividing γ by LC and multiplying the result by 2.

8. A voltage applying step of applying a momentary voltage between the wires of the coil; a response voltage measuring step of measuring a response voltage which is a voltage applied between the wires of the coil; a determination index calculation step of calculating a determination index from the response voltage that changes over time; a determination step of determining whether or not the coil is short-circuited using the determination index, the determination index is calculated based on a peak voltage and a zero-cross time of the response voltage; Let a be a sequence of numbers that starts from 1 and increases by 1, When a sequence b is a sequence that is the natural logarithm of the absolute value of the difference sequence of the sequence counting the peak voltage of the response voltage from the second peak voltage onward, the slope of the sequence b with respect to the sequence a is defined as a slope p, A sequence obtained by counting the zero cross times of the response voltage from the second zero cross time onward is defined as a sequence c, The gradient of the sequence c with respect to the sequence a is defined as gradient q, When the inductance of the equivalent circuit of the coil is L, the resistance of the coil is R, and the capacitance of the coil is C, The value obtained by dividing the slope p by the slope q and multiplying the result by −1 is defined as γ, and the value obtained by dividing the ratio π of the circumference of a circle to its circumference by the slope q is defined as ω, A coil diagnostic method in which LC, which is the inverse of the sum of the squares of γ and ω, and RC, which is the value obtained by dividing γ by LC and multiplying the result by 2, are used as the judgment indicators.

9. A voltage applying step of applying a momentary voltage between the wires of a coil; a response voltage measuring step of measuring a response voltage which is a voltage applied between the wires of the coil; a determination index calculation step of calculating a determination index from the response voltage that changes over time; a determination step of determining whether or not the coil is short-circuited using the determination index, the determination index is calculated based on a peak voltage and a zero-cross time of the response voltage; When the base of the natural logarithm is e, and the reciprocal of the time constant at which the response voltage decays to 1 / e is defined as γ, then: A coil diagnostic method, wherein the judgment index includes any one of γ, the inverse of γ, or the power of γ.

10. 9. The coil diagnosis method according to claim 3, wherein the sequence b and the sequence c have the same number of terms.

11. The coil diagnosis method according to claim 10, wherein the number of terms in the sequence b and the sequence c is odd.

12. a voltage application unit that applies an instantaneous voltage between the wires of the coil; a response voltage measurement unit that measures a response voltage that is a voltage applied between the wires of the coil; a judgment index calculation unit that calculates a judgment index from the response voltage that changes over time; and a judgment unit that uses the judgment index to determine whether or not the coil is short-circuited, the determination index calculation unit calculates the determination index based on a peak voltage and a zero-cross time of the response voltage; The judgment index calculation unit calculates the judgment index based on a peak voltage of the response voltage and a slope that is an increase rate at successive zero crossing times of the response voltage.

13. A device comprising: a voltage application unit that applies an instantaneous voltage between the wires of a coil; a response voltage measurement unit that measures a response voltage that is a voltage applied between the wires of said coil; a judgment index calculation unit that calculates a judgment index from said response voltage that changes over time; and a judgment unit that uses said judgment index to judge whether or not said coil is short-circuited; the determination index calculation unit calculates the determination index based on a peak voltage and a zero-cross time of the response voltage; Let a be a sequence of numbers that starts from 1 and increases by 1, When a sequence b is a sequence that is the natural logarithm of the absolute value of the difference sequence of the sequence counting the peak voltage of the response voltage from the second peak voltage onward, the slope of the sequence b with respect to the sequence a is defined as a slope p, A sequence obtained by counting the zero cross times of the response voltage from the second zero cross time onward is defined as a sequence c, When the gradient of the sequence c with respect to the sequence a is defined as gradient q, The judgment index calculation unit calculates, as the judgment index, γ defined by the value obtained by dividing the slope p by the slope q and multiplying the result by −1, and ω defined by the value obtained by dividing the constant π by the slope q.

14. A device comprising: a voltage application unit that applies an instantaneous voltage between the wires of a coil; a response voltage measurement unit that measures a response voltage that is a voltage applied between the wires of said coil; a judgment index calculation unit that calculates a judgment index from said response voltage that changes over time; and a judgment unit that uses said judgment index to judge whether or not said coil is short-circuited; the determination index calculation unit calculates the determination index based on a peak voltage and a zero-cross time of the response voltage; Let a be a sequence of numbers that starts from 1 and increases by 1, When a sequence b is a sequence that is the natural logarithm of the absolute value of the difference sequence of the sequence counting the peak voltage of the response voltage from the second peak voltage onward, the slope of the sequence b with respect to the sequence a is defined as a slope p, A sequence obtained by counting the zero cross times of the response voltage from the second zero cross time onward is defined as a sequence c, The gradient of the sequence c with respect to the sequence a is defined as gradient q, When the inductance of the equivalent circuit of the coil is L, the resistance of the coil is R, and the capacitance of the coil is C, the judgment index calculation unit defines a value obtained by dividing the slope p by the slope q and multiplying the result by −1 as γ and defines a value obtained by dividing the ratio of the circumference of a circle to its circumference π by the slope q as ω; A coil diagnostic device that calculates, as the judgment indexes, LC, which is the reciprocal of the sum of the squares of γ and ω, and RC, which is the value obtained by dividing γ by LC and multiplying the result by 2.

Images