JP7389641B2 - Winding diagnosis device and winding diagnosis method - Google Patents

Description

Application of Article 30, Paragraph 2 of the Patent Act (1) Published in “1D2-03” of the 62nd Automatic Control Conference Lecture Proceedings published by the Japan Society of Mechanical Engineers on November 8, 2020 ( 2) Presented at the 62nd Automatic Control Union Lecture on November 8, 2020

The present invention relates to a winding diagnostic device and a winding diagnostic method for diagnosing the winding condition of an electrical device, including the presence or absence of an abnormality in the winding.

Japanese Unexamined Patent Publication No. 2008-286611 (Patent Document 1) describes a winding diagnostic device that can diagnose whether or not there is an abnormality in the winding in an electrical device including a winding without disassembling the electrical device. There is.

The winding diagnostic device measures the voltage value observed at both ends of the winding of an electrical device when an impulse voltage is applied to the winding, and also determines whether the winding is connected to the impulse based on the measured voltage. Calculate the multiplication value LC and multiplication value RC of the inductance L and resistance R as equivalent circuit constants of the diagnostic circuit consisting of the voltage generation circuit and the capacitance C as the equivalent circuit constant of the diagnostic circuit, and It is diagnosed whether the winding is normal or not based on the value LC and the multiplication value RC.

By the way, the winding diagnostic device described in the above-mentioned publication uses a determinant to derive the multiplication value LC and the multiplication value RC. For this determinant, it is necessary to calculate the first-order differential and the second-order differential for the voltage value observed at both ends of the winding. Generally, in differential calculations, noise components included in the voltage signal during measurement have a large effect on the calculation results. In other words, if a differential operation is performed on a voltage signal containing many noise components, it may not be possible to calculate the correct multiplication value LC or multiplication value RC due to the influence of noise. It is possible to design a low-pass filter or band-pass filter to remove high-frequency noise components included in the measured voltage, but the frequency components necessary to derive the multiplication value LC and multiplication value RC differs for each winding to be diagnosed, so it takes a lot of man-hours to design the cutoff frequency, such as determining which frequency components to cut.

Japanese Patent Application Publication No. 2008-286611

The present invention has been made in view of the above, and its main purpose is to provide a winding diagnostic device capable of diagnosing the presence or absence of abnormalities in windings in electrical equipment while reducing design man-hours. do.

According to a preferred embodiment of the winding diagnostic device according to the present invention, the winding diagnostic device detects the winding based on the voltage observed at both ends of the winding when an impulse voltage with predetermined characteristics is applied from the impulse generating circuit to the winding of the electrical equipment. A winding diagnostic device is configured to diagnose the winding condition including the presence or absence of an abnormality. The winding diagnostic device includes a reading section, a calculation section, and a diagnosis section. The reading unit reads the voltage every first time over a predetermined period of time. The calculation section defines the first time as Ts, the voltage read by the reading section at each first time Ts as Vn , the equivalent circuit constant inductance of the measurement circuit including the winding and the impulse generation circuit as L, and the measurement circuit as The equivalent circuit constant capacitance is C, the equivalent circuit constant resistance of the measurement circuit is R, the product of the equivalent circuit constant inductance L and the equivalent circuit constant capacitance C is LC, and the product of the equivalent circuit constant resistance R and the equivalent circuit constant capacitance C is When the value is RC, a multiplication value LC and a multiplication value RC that minimize the error e shown in the following equation (1) are calculated using the following equations (2) and (3). Then, the diagnosis section diagnoses whether or not there is an abnormality in the winding based on the multiplication value LC and the multiplication value RC calculated by the calculation section.
(Number 1)

According to the present invention, since the multiplication value LC and the multiplication value RC that minimize the error e in equation (1) are only determined by the solution search method, There is no need to design a filter, and the multiplication value LC and multiplication value RC can be easily obtained. Thereby, it is possible to diagnose the presence or absence of an abnormality in the windings in electrical equipment while reducing the number of design steps.

According to a further embodiment of the winding diagnostic device according to the present invention, the calculation unit calculates the multiplication value LC and the multiplication value RC using a particle swarm optimization method. The inertia constant and acceleration constant in the particle swarm optimization method are set to a value of 1 or less.

In order to obtain the multiplication value LC and the multiplication value RC using the particle swarm optimization method, it is necessary to determine the inertia constant and the acceleration constant in advance. Generally, it is desirable that the inertia constant and acceleration constant have a value larger than 1 (for example, 2). However, as a result of extensive research, the inventor found that when determining the multiplication value LC and multiplication value RC, if the inertia constant and acceleration constant were set to a value greater than 1, the solution would diverge. . Based on such research results, in this embodiment, the multiplication value LC and the multiplication value RC can be determined by setting the inertia constant and the acceleration constant to a value of 1 or less.

According to a further embodiment of the winding diagnostic device according to the present invention, the predetermined time is set to one or more cycles of the damped oscillating voltage.

As a result of intensive research, the present inventor has found that, by using data in an interval less than one cycle of the measured voltage, a set of multiplication values LC and multiplication values RC that minimizes the error e shown in equation (1) can be obtained. We have found that there are cases where it is not found. Based on the results of this research, in this embodiment, it is possible to calculate the multiplier value LC and the multiplier value RC by using data for an interval of one cycle or more of the voltage after the voltage starts damped oscillation. It has become.

According to a further embodiment of the winding diagnostic device according to the present invention, the reading period f ₁ of the voltage by the reading section is determined to satisfy the following formula (4), where the period of the voltage is f ₀ . There is. Then, the first time is obtained by the following equation (5).
(Number 2)
10×f ₀ ≦f ₁ ≦1000×f ₀ (4)
Ts=1/f ₁ (5)

As a result of intensive research, the present inventors have found that depending on the time interval at which the voltage is measured, it may not be possible to find a set of multiplication values LC and multiplication values RC that minimizes the error e. Based on this research result, in this embodiment, by setting the time interval for measuring the voltage to the time shown by equation (5), it is easy to create a set of multiplication values LC and multiplication values RC that minimizes the error e. It is now possible to calculate

According to a further embodiment of the winding diagnostic device according to the present invention, the calculation unit calculates the multiplication value LC and the multiplication value RC using linear programming.

According to this embodiment, the multiplication value LC and the multiplication value RC can be easily obtained by using linear programming.

According to a further embodiment of the winding diagnostic device according to the present invention, the relationship between a plurality of coordinate groups consisting of the multiplication value LC and the multiplication value RC and the winding state corresponding to each of the plurality of coordinate groups is determined in advance. The apparatus further includes a storage section that stores the determined winding diagnosis map. The diagnosis section uses the winding diagnosis map to derive a winding state corresponding to the diagnosis target coordinates formed by the multiplication value LC and the multiplication value RC calculated by the calculation section. Then, the diagnosis section outputs the derived winding state as a diagnosis result.

According to this embodiment, since the winding condition can be diagnosed using a predetermined winding diagnosis map, it is possible to easily diagnose whether there is an abnormality in the winding.

According to a further embodiment of the winding diagnosis device according to the present invention, the diagnosis unit is configured to determine the distance between the diagnosis target coordinates and a normal coordinate in which the winding state is normal among the plurality of coordinate groups, and the diagnosis target coordinates. and an abnormal coordinate having an abnormal winding state among a plurality of coordinate groups are calculated, and the winding state corresponding to the coordinate having the shorter calculated distance is output as a diagnosis result.

According to this embodiment, the winding state can be diagnosed more easily using the winding diagnosis map.

According to a further aspect of the winding diagnosis device according to the present invention, the winding diagnosis map includes a plurality of coordinate groups, a normal coordinate group in which the winding condition is normal, and an abnormal coordinate group in which the winding condition is abnormal. This is a map classified into groups. Then, the diagnosis unit diagnoses that the winding condition is normal when the diagnosis target coordinates belong to the normal coordinate group, and diagnoses that the winding condition is abnormal when the diagnosis target coordinates belong to the abnormal coordinate group. do.

According to this embodiment, the winding state can be diagnosed more easily using the winding diagnosis map.

According to a further embodiment of the winding diagnosis device according to the present invention, in the winding diagnosis map, the abnormal coordinate group is further classified into classes depending on the mode of the abnormality. Then, when the diagnosis target coordinates belong to the abnormal coordinate group, the diagnosis unit diagnoses the mode of the abnormality.

According to this embodiment, it is possible to identify not only that the winding is abnormal, but also the mode of the abnormality.

According to a preferred embodiment of the winding diagnostic method according to the present invention, a winding diagnostic method for diagnosing the presence or absence of an abnormality in the winding of an electrical device is configured. The winding diagnosis method involves (a) applying an impulse voltage with predetermined characteristics to the winding from an impulse generating circuit, and (b) measuring the voltage observed at both ends of the winding when the impulse voltage is applied to the winding. (c) The first time is Ts, the voltage measured every predetermined time is Vn , and the equivalent circuit constant inductance of the measurement circuit including the winding and the impulse generating circuit is is L, the equivalent circuit constant capacitance of the measurement circuit is C, the equivalent circuit constant resistance of the measurement circuit is R, the multiplication value LC of the equivalent circuit constant inductance and the equivalent circuit constant capacitance, and the equivalent circuit constant resistance and the equivalent circuit constant capacitance. When the multiplication value RC of The presence or absence of an abnormality in the winding is diagnosed based on LC and multiplication value RC.
(Number 3)

According to the present invention, since the multiplication value LC and the multiplication value RC that minimize the error e in equation (6) are only determined by the solution search method, There is no need to design a filter, and the multiplication value LC and multiplication value RC can be easily obtained. Thereby, it is possible to diagnose the presence or absence of an abnormality in the windings in electrical equipment while reducing the number of design steps.

According to the present invention, it is possible to diagnose the presence or absence of an abnormality in the windings of an electrical device while reducing the number of design steps.

1 is a configuration diagram schematically showing the configuration of a winding diagnostic device 1 according to an embodiment of the present invention. It is a flowchart which shows an example of winding diagnostic processing. FIG. 2 is an explanatory diagram showing the waveform of the voltage V n observed at both ends of the winding 20 when an impulse voltage is applied to the winding 20. FIG. FIG. 2 is an explanatory diagram showing an initial state of a calculation process of a multiplication value LC and a multiplication value RC according to the PSO method. FIG. 2 is an explanatory diagram showing an initial state of a calculation process of a multiplication value LC and a multiplication value RC according to the PSO method. FIG. 3 is an explanatory diagram showing a process of calculating a multiplication value LC and a multiplication value RC using the PSO method. FIG. 3 is an explanatory diagram showing a process of calculating a multiplication value LC and a multiplication value RC using the PSO method. FIG. 3 is an explanatory diagram showing a process of calculating a multiplication value LC and a multiplication value RC using the PSO method. FIG. 3 is an explanatory diagram showing how the state of a winding 20 to be diagnosed is diagnosed using a winding diagnosis map. It is an explanatory view showing an example of a map for winding diagnosis. It is an explanatory view showing an example of a map for winding diagnosis of a modification.

Next, the best mode for carrying out the present invention will be described using examples.

The winding diagnosis device 1 according to the embodiment of the present invention is configured as a device for diagnosing whether a winding 20 of a rotating machine such as a motor or a generator or an electric device such as a transformer is normal or abnormal. As shown in FIG. 1, the winding diagnostic device 1 includes terminals T1 and T2 to which the winding 20 is connected, and an impulse generation circuit electrically connected to the terminals T1 and T2 via power lines L1 and L2. 2, a voltage measurement section 4 electrically connected to the power lines L1 and L2, an A/D conversion circuit 6 electrically connected to the voltage measurement section 4, and a display section 8 that displays diagnostic results. , and an electronic control section 10 that controls the entire device.

The impulse generating circuit 2 is a circuit capable of applying an impulse voltage having predetermined characteristics to the winding 20. The voltage measurement unit 4 measures the voltage V observed at both ends of the winding 20 when an impulse voltage is applied to the winding 20 connected to the terminals T1 and T2. The A/D conversion circuit 6 converts the analog signal of the voltage V measured by the voltage measurement section 4 into a digital signal, and outputs the digital signal to the electronic control section 10.

As shown in FIG. 1, the electronic control unit 10 is configured as a microprocessor centered on a CPU 12, and in addition to the CPU 12, it includes a ROM 14 for storing processing programs, a RAM 16 for temporarily storing data, and a RAM 16 (not shown). It is equipped with an input/output port and a communication port. A voltage V n from the A/D conversion circuit is input to the electronic control unit 10 via an input port (not shown). Further, the electronic control unit 10 outputs an impulse voltage application signal to the impulse generation circuit 2, a diagnosis result display signal to the display unit 8, etc. via an output port (not shown).

Next, the operation of the winding diagnosing device 1 according to the embodiment of the present invention configured in this manner, particularly the operation when diagnosing whether or not an abnormality has occurred in the winding 20 will be described. FIG. 2 is a flowchart showing an example of a winding diagnosis process executed by the winding diagnosis apparatus 1. This process is executed when the winding 20 to be diagnosed is connected to the terminals T1 and T2 and an instruction to start diagnosis is given (for example, when a diagnosis start button is pressed).

When the winding diagnosis process is executed, the CPU 12 of the winding diagnosis device 1 first outputs an impulse voltage application signal to the impulse generation circuit 2 so as to apply an impulse voltage to the winding 20 (step S100). Subsequently, the process of reading the voltage n across the winding 20 measured by the voltage measurement unit 4 at every sampling time Ts is continuously executed for a predetermined time T (step S102), and the predetermined time T has elapsed. A process for determining whether or not the information has been performed is executed (step S104). Here, as shown in FIG. 3, the voltage V n across the winding 20 measured by the voltage measurement unit 4 is sampled when the amplitude value attenuates over time and its period is f ₀ . The time Ts is determined to satisfy the following equation (9). Further, the predetermined time T is set to a value equal to or longer than one cycle of the voltage V n whose amplitude value attenuates over time. The reason why the sampling time Ts and the predetermined time T are determined in this way is that if the sampling time Ts does not satisfy the following equation (9) or if the predetermined time T is set to a value less than one cycle of the voltage Vn, This is because, as a result of intensive research, the present inventors have found that there are cases in which a set of multiplication values LC and RC that minimizes the error e, which will be described later, cannot be found. The CPU 12 that continuously executes the process of reading the voltage V n across the winding 20 at every sampling time Ts for a predetermined time T is an example of an implementation configuration corresponding to the "reading section" in the present invention. Further, the sampling time Ts is an example of an implementation configuration corresponding to the "first time" in the present invention.

(Number 4)
10×f ₀ ≦1/Ts≦1000×f ₀ (9)
However, f ₁ =1/Ts, and f ₁ is the sampling period of the voltage Vn. Note that preferably 10×f ₀ ≦1/Ts≦100×f ₀ .

If the predetermined time T has not elapsed, the processes of steps S102 to S104 are repeatedly executed until the predetermined time T has elapsed. On the other hand, if the predetermined time T has elapsed, a process of calculating a multiplication value LC and a multiplication value RC is executed (step S106). Here, the multiplication value LC is the multiplication value of the equivalent circuit constant inductance L and the equivalent circuit constant capacitance C of the measurement circuit 30 (see FIG. 1) consisting of the winding 20 to be diagnosed and the impulse generation circuit 2. The multiplication value RC is the multiplication value of the equivalent circuit constant resistance R and the equivalent circuit constant capacitance C of the measurement circuit 30. The CPU 12 that executes the process of calculating the multiplication value LC and the multiplication value RC is an example of an implementation configuration corresponding to the "calculation unit" in the present invention.

The multiplication value LC and the multiplication value RC are calculated using the following equations (10) to (12). Specifically, in this embodiment, in equation (10), the multiplication value LC and the multiplication value RC that minimize the error e are determined using the particle swarm optimization method (hereinafter referred to as "PSO method"). The configuration is such that it is derived by

(Number 5)

The PSO method is based on the idea of updating the particle velocity v _i (t) at time t and the particle position p _i (t) at time t, as shown in the following equations (13) and (14) (however, i=1, 2, 3,...N). That is, in this embodiment, as shown in FIG. 4, all of a predetermined number of points p _i (LC _i , RC _i ) randomly arranged within the LC-RC coordinates are calculated using the following equation (13) and (14) while adjusting the moving speed (see FIGS. 4 to 8) to a position with a high degree of adaptation, in other words, a point p _i (LC _i , RC _i ) that minimizes the error e. We will explore while updating it from time to time.

(Number 6)

Here, w is a constant called an inertia constant, and indicates how much the moving speed at time t-1 before time t is maintained. Further, c ₁ and c ₂ are constants called acceleration constants. As a result of extensive research, the present inventors have determined that when determining the multiplication value LC and the multiplication value RC, the inertia constant w and the acceleration constants c ₁ and c ₂ are set to values larger than the generally desirable value 1 ( For example, it has been found that if the value is set to 2), the solution diverges, making it impossible to obtain the multiplication value LC and the multiplication value RC. Based on such research results, in this embodiment, the multiplier value LC and the multiplier value RC can be determined by setting the inertia constant w and the acceleration constants c ₁ and c ₂ to a value of 1 or less. It is something. Note that p _i ^best is a personal best indicating the position with the highest degree of fitness among the previous searches for a specific particle, and p _g ^best is a personal best for all particle groups (point p _i (LC _i , RC _i )) is a global best that indicates the position with the highest degree of fitness, including past searches, and is a variable that is updated from time to time during repeated searches.

After the multiplication value LC and the multiplication value RC are calculated in this way, a process of diagnosing the state of the winding 20 is executed based on the multiplication value LC and the multiplication value RC (step S108). Here, in the present embodiment, the determination as to whether or not there is an abnormality in the winding 20 is performed by mapping the calculated multiplication value LC and multiplication value RC onto a winding diagnosis map, as shown in FIG. This is performed based on the location of the mapped multiplication value LC and multiplication value RC. In addition, in this embodiment, as shown in FIG. 10, the average values LC _norav , RC _norav , and , the average value LC _abav1 , RC _abav1 , LC _{abav2 , RC abav2 ,} LC _abav3 , RC _abav3 and LC _abav4 for each abnormal state of the multiplication value LC of the plurality of abnormal windings 20 (of the plurality of abnormal states) and the multiplication value RC for each abnormal state. , RC _abav4 are determined in advance and are stored in the ROM 14 as a winding diagnosis map in association with the winding condition. Examples of abnormal states of the winding include, for example, a one-turn short circuit occurs in the winding 20, a two-turn short circuit occurs in the winding 20, and a phase-to-phase short circuit occurs in the winding 20. This may include a state where a disconnection has occurred in the winding 20. The ROM 14 in which the winding diagnosis map is stored is an example of an implementation configuration corresponding to the "storage section" in the present invention. In addition, the multiplication values LC and RC of the plurality of normal windings 20 (in which no abnormality such as short circuit has occurred) correspond to the "normal coordinate group" in the present invention, and The multiplication value LC and the multiplication value RC of the winding 20 correspond to the "abnormal coordinate group" in the present invention, and the multiplication value LC and the multiplication value of a plurality of normal windings 20 (no abnormalities such as short circuits have occurred) The value RC, and the multiplication value LC and multiplication value RC of the plurality of abnormal windings 20 (in the plurality of abnormality modes) are an example of an implementation configuration corresponding to "a plurality of coordinate groups" in the present invention.

Then, when the multiplication value LC and multiplication value RC of the winding 20 to be diagnosed are calculated, the multiplication value LC and the multiplication value RC are calculated, and each value (average value LC _norav , RC _norav , average value LC _abav1 , RC _abav1 and LC _abav2 , RC _abav2 ), and calculate the average values LC _norav , RC _norav with which the distances are closest, the average values LC _abav1 , RC _abav1 , LC _abav2 , RC _abav2 , LC _abav3 , RC _abav3 and LC _abav4 , RC The winding state associated with _abav4 is diagnosed as the state of the winding 20 to be diagnosed. The CPU 12 that executes diagnosis of the state of the winding 20 to be diagnosed is an example of an implementation configuration corresponding to the "diagnosis section" in the present invention.

If the condition of the winding 20 to be diagnosed is normal, "normal" is output to the display section 8 (step S110), and this processing routine is ended. On the other hand, if the condition of the winding 20 to be diagnosed is abnormal, "abnormal" and "abnormal state" are output to the display section 8 (step S112), and the present processing routine is ended.

According to the winding diagnostic device 1 of the present invention according to the present embodiment described above, when calculating the multiplier value LC and the multiplier value RC suitable for diagnosing whether the winding 20 is normal or not, the formula ( Since the multiplication value LC and the multiplication value RC that minimize the error e in 10) are only found by the solution search method, when finding the multiplication value LC and the multiplication value RC, it is necessary to consider the noise component contained in the voltage to be measured. There is no need to design a filter for removal. Thereby, it is possible to easily diagnose whether there is an abnormality in the winding 20 in the electrical device while reducing the number of design steps.

Further, according to the winding diagnostic device 1 of the present invention according to the present embodiment, since the inertia constant w and the acceleration constants c ₁ and c ₂ in the PSO method are set to a value of 1 or less, the solution does not diverge. The multiplication value LC and the multiplication value RC can be reliably obtained.

Furthermore, according to the winding diagnostic device 1 of the present invention according to the present embodiment, the measurement time of the voltage V n observed at both ends of the winding 20 when an impulse voltage is applied is one period of the voltage V n . Since the above settings are made, it is possible to reliably obtain a set of multiplication values LC and multiplication values RC that minimizes the error e.

Further, according to the winding diagnostic device 1 of the present invention according to the present embodiment, the reading period f ₁ of the voltage V n is 10×f ₀ ≦ f ₁ when the period of the voltage V n is f ₀ . Since the setting is made to satisfy ≦1000×f ₀ , it is possible to reliably obtain a set of multiplication value LC and multiplication value RC that minimizes the error e.

In this embodiment, the average value LC _{norav , RC norav of} the multiplication value LC and the multiplication value RC of a plurality of normal windings 20 (no abnormalities such as short circuits have occurred), and a plurality of abnormal windings 20 (no abnormality such as short circuit) average values LC _abav1 , RC _abav1 and LC _{abav2 , RC abav2 of} the multiplication value LC and multiplication value RC of the winding 20 (of the aspect), the multiplication value LC and the multiplication value RC of the winding 20 to be diagnosed, The winding conditions associated with the average values LC _norav , RC _norav , average values LC _abav1 , RC _abav1 and LC _abav2 , RC _abav2 with the closest distances are calculated and the winding states of the winding 20 to be diagnosed are calculated. Although the configuration is such that diagnosis is performed as a condition, the present invention is not limited to this.

For example, without taking an average value, the multiplication value LC and multiplication value RC of a plurality of normal windings 20 (no abnormalities such as short circuits have occurred), and the multiplication value RC of a plurality of abnormal windings (with a plurality of abnormalities) The distance between the multiplication value LC and multiplication value RC of 20 and the multiplication value LC and multiplication value RC of the winding 20 to be diagnosed is calculated, and the multiplication value LC and multiplication value on the winding diagnosis map with the closest distance are calculated. The winding state associated with the multiplication value RC may be diagnosed as the state of the winding 20 to be diagnosed.

For example, as shown in FIG. 11, the multiplication values LC and RC of a plurality of normal windings 20 (no abnormalities such as short circuits have occurred), and the multiplication values LC and multiplication values RC of a plurality of abnormal windings 20 (of a plurality of abnormalities) A winding diagnosis map in which the multiplication value LC and multiplication value RC of the winding 20 are classified into normal regions and abnormal regions using machine learning such as a support vector machine is stored in advance in the ROM 14. When the multiplication value LC and the multiplication value RC of the winding 20 to be diagnosed are calculated, the value of the winding 20 to be diagnosed is determined depending on which region the multiplication value LC and the multiplication value RC belong to. It may also be configured to diagnose the condition. Note that the abnormality region can be further classified into different types of abnormality, such as one-turn short-circuit region, two-turn short-circuit region, phase-to-phase short circuit, and disconnection, so that even the abnormality mode can be specified.

In this embodiment and the above-mentioned modification, even the aspect of the abnormality is specified, but a configuration may be adopted in which only the abnormality is diagnosed without specifying the aspect of the abnormality.

In this embodiment, the PSO method is used as a solution search method for finding the multiplication value LC and the multiplication value RC that minimize the error e, but the method is not limited to this. For example, linear programming may be used as a solution search method for finding the multiplication value LC and the multiplication value RC that minimize the error e.

This embodiment shows an example of a mode for carrying out the present invention. Therefore, the present invention is not limited to the configuration of this embodiment.

1 Winding diagnostic device (winding diagnostic device)
2 Impulse generation circuit (impulse generation circuit)
4 Voltage measurement section 6 A/D conversion circuit 8 Display section 10 Electronic control section 12 CPU (reading section, calculation section, diagnostic section)
14 ROM (memory section)
16 RAM
20 Winding wire (winding wire)
30 Measurement circuit (measurement circuit)
Vn voltage (voltage observed across the winding)
T1 terminal T2 terminal L1 power line L2 power line Ts sampling time (first time)
T Predetermined time (predetermined time)
e Error (error)
L Equivalent circuit constant inductance (equivalent circuit constant inductance)
C Equivalent circuit constant capacitance (Equivalent circuit constant capacitance)
R Equivalent circuit constant resistance (Equivalent circuit constant resistance)
LC Multiply value (Multiply value)
RC Multiply value (Multiply value)
c ₁ acceleration constant (acceleration constant)
c ₂ acceleration constant (acceleration constant)
w inertia constant (inertia constant)
f ₁ Voltage reading cycle (voltage reading cycle)
f ₀ Voltage period (voltage period)

Claims (10)

A winding that diagnoses the state of the winding, including the presence or absence of abnormalities in the winding, based on the voltage observed at both ends of the winding when an impulse voltage with predetermined characteristics is applied to the winding of an electrical device from an impulse generation circuit. A diagnostic device,
a reading unit that reads the voltage every first time over a predetermined period of time;
The first time is Ts, the voltage read by the reading section every first time Ts is Vn , the equivalent circuit constant inductance of the measuring circuit including the winding and the impulse generating circuit is L, the measuring circuit The equivalent circuit constant capacitance of is C, the equivalent circuit constant resistance of the measurement circuit is R, the multiplication value of the equivalent circuit constant inductance L and the equivalent circuit constant capacitance C is LC, and the equivalent circuit constant resistance R and the equivalent circuit constant capacitance are an arithmetic unit that calculates the multiplier value LC and the multiplier value RC using the following equations (2) and (3) that minimizes the error e shown in the following equation (1), when the multiplication value with C is RC; ,
a diagnostic unit that diagnoses the winding state based on the calculated multiplication value LC and the multiplication value RC;
A winding diagnostic device comprising:
(Number 1)

The calculation unit calculates the multiplication value LC and the multiplication value RC using a particle swarm optimization method,
The winding diagnostic device according to claim 1, wherein the inertia constant and acceleration constant in the particle swarm optimization method are set to a value of 1 or less. The winding diagnostic device according to claim 1 or 2, wherein the predetermined time is set to a value equal to or longer than one cycle of the voltage that undergoes damped oscillation. The reading period f ₁ of the voltage by the reading section is determined to satisfy the following formula (4), where the period of the voltage is f ₀ ,
The winding diagnostic device according to any one of claims 1 to 3, wherein the first time is obtained by the following equation (5).
(Number 2)
10×f ₀ ≦f ₁ ≦1000×f ₀ (4)
Ts=1/f ₁ (5) The winding diagnostic device according to claim 1, wherein the calculation unit calculates the multiplication value LC and the multiplication value RC using linear programming. a storage unit storing a winding diagnosis map that predetermines the relationship between a plurality of coordinate groups consisting of the multiplication value LC and the multiplication value RC and the winding state corresponding to each of the plurality of coordinate groups; More prepared,
The diagnosis section uses the winding diagnosis map to derive the winding state corresponding to the diagnosis target coordinates consisting of the multiplication value LC and the multiplication value RC calculated by the calculation section, and The winding diagnosis device according to any one of claims 1 to 5, wherein the winding condition is output as a diagnosis result. The diagnosis unit determines the distance between the diagnosis target coordinates and a normal coordinate in which the winding state is normal among the plurality of coordinate groups, and the distance between the diagnosis target coordinates and the winding state among the plurality of coordinate groups. The winding diagnostic device according to claim 6, wherein the distance between the abnormal coordinate and the abnormal coordinate having an abnormality in the wire state is calculated, and the winding state corresponding to the coordinate having the shorter calculated distance is output as the diagnosis result. The winding diagnosis map is a map in which the plurality of coordinate groups are classified into a normal coordinate group in which the winding state is normal and an abnormal coordinate group in which the winding state is abnormal;
The diagnosis unit diagnoses that the winding state is normal when the diagnosis target coordinates belong to the normal coordinate group, and diagnoses that the winding state is normal when the diagnosis target coordinates belong to the abnormal coordinate group. The winding diagnostic device according to claim 6, which diagnoses that the winding is abnormal. In the winding diagnostic map, the abnormal coordinate group is further classified into classes according to the abnormality mode,
The winding diagnostic device according to claim 8, wherein the diagnosis unit diagnoses the abnormality mode when the diagnosis target coordinates belong to the abnormal coordinate group. A winding diagnosis method for diagnosing the presence or absence of an abnormality in the windings of electrical equipment, the method comprising:
(a) applying an impulse voltage with predetermined characteristics to the winding from an impulse generation circuit;
(b) measuring the voltage observed at both ends of the winding at first time intervals over a predetermined period of time when the impulse voltage is applied to the winding;
(c) The first time is Ts, the voltage measured every first time Ts is Vn , the equivalent circuit constant inductance of the measuring circuit including the winding and the impulse generating circuit is L, the measuring circuit C is the equivalent circuit constant capacitance of the measurement circuit, R is the equivalent circuit constant resistance of the measurement circuit, is the product value LC of the equivalent circuit constant inductance L and the equivalent circuit constant capacitance C, and is the equivalent circuit constant resistance R and the equivalent circuit constant capacitance. When the multiplication value RC with C is calculated, the multiplication value LC and the multiplication value RC that minimize the error e shown in the following equation (6) are calculated using the following equations (7) and (8),
(d) A winding diagnosis method, which diagnoses whether or not there is an abnormality in the winding based on the calculated multiplication value LC and multiplication value RC.
(Number 3)

Images