A method and device for identifying motor slip ring fault based on characteristic curve classification

Abstract

The present invention relates to a motor slip ring fault identification method and device based on characteristic curve classification, the method comprising: S100. Pulse discontinuous cycle curve identification: judging the number of arc pulses generated in the cycle period, and issuing a fault alarm when the number of arc pulses exceeds the set value; S200. Pulse top continuous curve identification: according to the arc intensity collected at each sampling point in the cycle period, calculating the correlation coefficient and the average value of the arc intensity within the time difference of multiple sampling points before and after, so as to determine and issue a fault alarm. This method classifies and identifies curves reflecting different wear conditions by reflecting the changes in the collected arc intensity curve according to the wear state of the slip ring carbon brush, thereby improving the judgment accuracy and fault conditions, and the judgment accuracy is high.

Description

Technical Field

[0001] The invention belongs to the technical field of motor monitoring, and in particular relates to a method and a device for identifying motor slip ring faults based on characteristic curve classification. Background Art

[0002] At present, high-voltage motors have been widely used in the industrial field. When the carbon brushes of high-voltage motors are constantly worn during rotation, the gap increases, and an excessively large electrical gap easily causes gap spark discharge.

[0003] At the same time, the applicant found that: one existing monitoring method is to use a temperature and humidity sensor to monitor the slip ring. When the temperature around the slip ring rises, it means that the arc light is burning violently, and it will be displayed as a fault; however, when the temperature rises, it means that a fault has occurred, and the motor slip ring has been damaged, and advance warning cannot be achieved; when the fault occurs, it is unexpected, resulting in production stoppage, unexpected waste, defective products and personnel confusion. Another monitoring method is to use humidity monitoring. When the humidity is high, the slip ring box is dehumidified to reduce the impact of humid air on the electrical gap of the slip ring to reduce the probability of discharge. However, this monitoring method can only reduce the impact of humid air, and does not essentially monitor and alarm the electrical gap caused by slip ring wear.

[0004] Therefore, it is of great significance to propose a new method to monitor the electrical clearance problem caused by the wear of the motor slip ring, give early warning before the serious damage of the motor occurs, arrange maintenance in time, adjust production measures, and avoid shutdown. Summary of the invention

[0005] In order to solve the above problems existing in the prior art, the present invention provides a method and device for identifying motor slip ring faults based on characteristic curve classification.

[0006] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0007] In a first aspect, a method for identifying a motor slip ring fault based on characteristic curve classification according to the present invention comprises:

[0008] S100. Pulse discontinuous cycle curve identification: determine the cumulative number of arc pulses generated in each cycle, and compare it with the arc pulse number setting value, and mark the identification bit when it exceeds the arc pulse number setting value, and issue a fault alarm when the cumulative number of identification bits in multiple cycles exceeds the setting value;

[0009] S200. Pulse top continuous curve identification: Based on the arc intensity collected at each sampling point in each cycle, calculate the correlation coefficient and the average arc intensity within the time difference of multiple sampling points before and after, and determine whether to issue a fault alarm based on the correlation coefficient and the average arc intensity within the time difference of multiple sampling points before and after.

[0010] Furthermore, the pulse discontinuous cycle curve identification specifically includes:

[0011] S101. Determination of arc pulse generation: Set the arc intensity constants h1 and h2, and the time constant t1. When the arc intensity is greater than h1 and the duration is greater than t1, it is determined that an arc pulse is generated;

[0012] S102. Arc pulse end determination: After determining that an arc pulse is generated, if the currently collected arc intensity drops below h2, the arc pulse is determined to be ended, and the arc pulse generation and arc pulse end process is counted as one complete arc pulse;

[0013] S103. Calculation of the number of arc pulses: Taking the cycle period t2 as the marking time, determine and count the number of arc pulses n generated within the cycle period t2 according to the method of S101-S102;

[0014] If only the arc pulse described in S101 is generated within the cycle period t2, and the arc pulse described in S102 is not terminated, it is determined that the number n of arc pulses generated within the cycle period t2 is 1;

[0015] If the number n of arc pulses generated within the cycle period t2 exceeds the set value Nset, the arc pulse cycle is marked as a flag bit, and the flag bits are accumulated;

[0016] The arc intensity constant h1>h2, the time constant t1<cycle period t2, and the cycle period t2>1s, and the cycle period t2 is greater than or equal to the time for the motor shaft to rotate one circle;

[0017] S104. Alarm determination: When the accumulated flag is greater than the set value Lset within the cumulative time t3, an alarm is issued to notify the user to arrange maintenance.

[0018] Furthermore, the pulse top continuous curve identification specifically includes:

[0019] S201. Calculate the correlation coefficient within the time difference of multiple sampling points before and after: Sample the arc intensity at a frequency of p points per second, collect p*t2 points within the cycle time t2, and divide the sampling points within the cycle time t2 into m equal parts, m ≥ 10, and m is an even number, and then solve the before and after according to the following formula The correlation coefficient pM within the time difference of the sampling points x M x+1 , a total of m / 2 correlation coefficients are solved;

[0020]

[0021] Where x = 0, 2, 4…m, i represents the time of the sampling point, Δh1(i) = h(i+1)-h(i),

[0022] h() represents the arc intensity collected at the sampling point, and M and p are both marking symbols;

[0023] S202. Calculate the average arc intensity within the time difference of multiple sampling points before and after: Solve the following formula before and after The average arc intensity within the time difference of sampling points A total of m / 2 arc intensity averages were solved;

[0024]

[0025] Where x = 0, 2, 4…m, i represents the time of the sampling point, and M and h are marking symbols;

[0026] S203. Obtain the values ​​of the minimum correlation coefficient and the second-to-last correlation coefficient: Compare the first two correlation coefficients among the obtained m / 2 correlation coefficients to obtain the minimum value p of the two correlation coefficients. MM2min , and compare the remaining (m / 2)-2 correlation coefficients to get the second-to-last largest value p MMmax ;

[0027] S204. Obtain the values ​​of the maximum arc intensity average value and the second-to-last arc intensity average value: compare the first two arc intensity average values ​​among the m / 2 arc intensity average values ​​obtained to obtain the maximum value h of the two arc intensity average values. M2max , and compare it with the remaining (m / 2)-2 arc intensity averages to get the second smallest value h Mmin ;

[0028] S205, judging whether to issue a fault alarm according to the value of the correlation coefficient obtained in S203 and the value of the average arc intensity obtained in S204;

[0029] If h Mmin >h M2max When the fault occurs, a fault alarm is issued;

[0030] If h Mmin >h M2max And p MMmax <pMM2min A serious fault alarm is issued.

[0031] On the one hand, the method of the present invention can monitor the changes in arc characteristics caused by changes in the electrical gap during the wear process of the slip ring, and perform monitoring and early warning judgments, that is, it fully reflects the wear state of the slip ring carbon brush on the collected arc intensity curve change process. Different curves reflect different wear conditions. Through the classification and identification of different curves, the judgment accuracy and fault conditions are improved, and the judgment accuracy is high; on the other hand, the fixed value of the cycle time t2 can be set according to different speeds to ensure at least one rotation time, which can match different motor speed working scenarios and the algorithm has strong adaptability.

[0032] In the second aspect, the present invention also provides a motor slip ring fault identification device based on characteristic curve classification, including a pulse discontinuous cycle curve identification module and a pulse top continuous curve identification module, the pulse discontinuous cycle curve identification module has the above-mentioned pulse discontinuous cycle curve identification function, and the pulse top continuous curve identification module has the above-mentioned pulse top continuous curve identification function.

[0033] Furthermore, the identification device includes a CPU, a multi-channel sampling circuit, an arc sensor, an optical fiber, and an arc receiver; wherein the CPU integrates the pulse discontinuous cycle curve identification module and the pulse top continuous curve identification module, the arc sensor is connected to the arc receiver via the optical fiber, and the arc receiver is connected to the CPU via the sampling circuit.

[0034] Furthermore, the identification device also includes a temperature sensor, a humidity sensor and an output relay, the temperature sensor and the humidity sensor are connected to the CPU through a sampling circuit, and the peripheral circuit is connected to the CPU to support the operation of the CPU; the temperature sensor and the humidity sensor respectively collect the temperature and humidity in the slip ring box, and transmit them to the CPU through the sampling circuit, and when the temperature is lower than the set value or the humidity is higher than the set value, the CPU controls the relay to operate the heater to heat and dehumidify the slip ring box.

[0035] For each aspect in the above-mentioned second aspect and the technical effects that may be achieved by each aspect, please refer to the above-mentioned description of the technical effects that can be achieved by various possible schemes in the first aspect, and no further details will be given here. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a flow chart of the motor slip ring fault identification method based on characteristic curve classification according to the present invention;

[0037] Figure 2It is a flow chart of pulse discontinuous cycle curve identification in the motor slip ring fault identification method based on characteristic curve classification according to the present invention;

[0038] Figure 3 It is a flow chart of pulse top continuous curve identification in the motor slip ring fault identification method based on characteristic curve classification according to the present invention;

[0039] Figure 4 It is a structural schematic diagram of the motor slip ring fault identification device based on characteristic curve classification described in the present invention. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0041] like Figure 1 As shown in , a motor slip ring fault identification method based on characteristic curve classification according to an embodiment of the present invention includes S100. pulse discontinuous cycle curve identification and S200. pulse top continuous curve identification.

[0042] Step S100. Identification of pulse discontinuous cycle curve: determine the cumulative number of arc pulses generated in each cycle, and compare it with the arc pulse number setting value, and mark the identification bit when it exceeds the arc pulse number setting value, and issue a fault alarm when the cumulative number of identification bits in multiple cycles exceeds the setting value; Figure 2 As shown, the specific steps include:

[0043] S101. Determination of arc pulse generation: Set the arc intensity constants h1 and h2 (h1>h2), the time constant t1, and when the arc intensity is greater than h1 and the maintenance time is greater than t1, it is determined that an arc pulse is generated;

[0044] S102. Arc pulse end determination: After determining that an arc pulse is generated, if the currently collected arc intensity drops below h2, the arc pulse is determined to be ended, and the arc pulse generation and arc pulse end process is counted as one complete arc pulse;

[0045] S103. Calculation of arc pulse number and alarm determination: With cycle t2 as the mark time, determine and count the number of arc pulses n generated within cycle t2 according to the method of S101-S102, where cycle t2> time constant t1, and cycle t2> 1s, and cycle t2 is greater than or equal to the time for the motor shaft to rotate one circle, because conventional motors rotate fast, up to several circles per second, but the motor speed in some high-voltage cable drum slip ring boxes is slow, and it takes several seconds to rotate one circle;

[0046] If only the arc pulse described in S101 is generated within the cycle period t2, and the arc pulse described in S102 is not terminated, it is determined that the number n of arc pulses generated within the cycle period t2 is 1;

[0047] If the number n of arc pulses generated within the cycle period t2 exceeds the fixed value Nset (the fixed value of the number of pulses in the cycle period), the arc pulse period is marked as a flag bit, and the flag bits are accumulated;

[0048] S104. Alarm determination: When the accumulated flag is greater than the set value Lset (accumulated period exceeds limit value) within the accumulated time t3, an alarm is issued to notify the user to arrange maintenance; the accumulated time t3 should be an integer multiple of the accumulated time t2.

[0049] For example: the cycle period t2 is 1s, the cumulative time t3 is 10s, Nset is 15, and Lset is 6. If the number of arc pulses detected within 1s exceeds 15, the cycle period will be marked as 1. If there are multiple 1s of identification bits within 10s and the accumulated number of these identification bits is greater than Lset, an alarm signal will be issued to tell the user to repair the slip ring box.

[0050] Step S200. Pulse top continuous curve identification: According to the arc intensity collected at each sampling point in each cycle, the correlation coefficient and the average arc intensity within the time difference of multiple sampling points are calculated, and the fault alarm is determined by the correlation coefficient and the average arc intensity within the time difference of multiple sampling points; Figure 3 As shown, the specific steps include:

[0051] S201. Calculate the correlation coefficient within the time difference of multiple sampling points before and after: Sample the arc intensity at a frequency of p points per second, collect p*t2 points within the cycle time t2, and divide the sampling points within the cycle time t2 into m equal parts, m ≥ 10, and m is an even number, and then solve the before and after according to the following formula The correlation coefficient pM within the time difference of the sampling points x M x+1 , a total of m / 2 correlation coefficients are solved;

[0052]

[0053] Where x = 0, 2, 4…m, i represents the time of the sampling point, Δh1(i) = h(i+1)-h(i),

[0054] h() represents the arc intensity collected at the sampling point, M and p are both marking symbols, when m is 10, there are ten equal divisions, namely PM0M1, PM2M3, PM4M5, PM6M7, PM8M9, and there are m / 2=5 correlation coefficients for ten equal divisions, M0M1 represents the correlation coefficient between the first two equal division data groups under m equal divisions;

[0055] The calculation of the correlation coefficient in this method uses the similarity of the trend of arc light changes to make a judgment. When the trends are similar, the correlation coefficient is large; when the trends are not similar, the correlation coefficient is small. Therefore, in the process of arc light state change, the trends are similar, which proves that the slip ring has not deteriorated significantly. If the trends are not similar, it means that the state of the slip ring has changed. The change in state trend (i.e., the correlation coefficient) is used to monitor the characteristics of the slip ring in the continuous time domain, taking into account the process of the slip ring state changing over time. S202. Calculate the average arc light intensity within the time difference of multiple sampling points before and after: Solve the before and after according to the following formula The average arc intensity within the time difference of sampling points A total of m / 2 arc intensity averages were solved;

[0056]

[0057] Where x = 0, 2, 4…m, i represents the time of the sampling point, and M and h are marking symbols;

[0058] The calculation of the average arc intensity in this method utilizes the fact that the slip ring has an obvious fault at one or some points, and the arc intensity generated is large, which has a better effect on the fault state identification with obvious fault characteristics.

[0059] S203. Obtain the values ​​of the minimum correlation coefficient and the second-to-last correlation coefficient: Compare the first two correlation coefficients among the obtained m / 2 correlation coefficients to obtain the minimum value p of the two correlation coefficients. MM2min , and compare the remaining (m / 2)-2 correlation coefficients to get the second-to-last largest value p MMmax ;

[0060] S204. Obtain the values ​​of the maximum arc intensity average value and the second-to-last arc intensity average value: compare the first two arc intensity average values ​​among the m / 2 arc intensity average values ​​obtained to obtain the maximum value h of the two arc intensity average values. M2max , and compare it with the remaining (m / 2)-2 arc intensity averages to get the second smallest value h Mmin ;

[0061] S205, based on the value of the correlation coefficient obtained in S203 and the value of the average arc intensity obtained in S204 (i.e., p MM2min 、pMMmax 、h M2max and h Mmin ) Determine whether to issue a fault alarm;

[0062] If h Mmin >h M2max When the fault occurs, a fault alarm is issued;

[0063] If h Mmin >h M2max And p MMmax <p MM2min A serious fault alarm is issued.

[0064] The motor slip ring fault identification method based on characteristic curve classification of the present invention can, on the one hand, monitor the change of arc characteristics caused by the change of electrical gap during the slip ring wear process, and perform monitoring and early warning judgment, that is, it fully reflects the change process of the collected arc intensity curve based on the wear state of the slip ring carbon brush. Different curves reflect different wear conditions. Through the classification and identification of different curves, the judgment accuracy and fault conditions are improved, and the judgment accuracy is high (because the gap spark is an arc light combustion process, and the main component of the arc light is ultraviolet light. The fault condition of the high-voltage motor slip ring can be identified by identifying the ultraviolet light band, and monitoring and Early warning, specifically, through predicting the typical characteristics of arc light caused by electrical gap caused by wear of slip ring carbon brush and contact parts, the arc light changes from weak to strong during the change of gap caused by wear, so it can be predicted during the change of arc light intensity; in addition, the arc light changes from nothing to something, from weak to strong, which will cause the arc light monitored in the time domain to be a discontinuous pulse. After the arc light intensity becomes stronger, the top of the pulse forms a continuous demand, and it is fully feasible to analyze and predict different characteristics); secondly, the constant value of the cycle time t2 can be set according to different speeds to ensure at least one rotation time, which can match different motor speed working scenarios, and the algorithm has strong adaptability.

[0065] The method of the present invention is further described below by means of a simple example.

[0066] Example

[0067] The motor slip ring fault identification method based on characteristic curve classification of the present invention comprises the following steps:

[0068] 1. If the cycle period t2 is 1s and Nset is 13, when more than 13 arc pulses are detected within 1s, the flag is set; if the cumulative time t3 is 10s and Lset is 6, when the flag is set 6 times within 10s, an alarm signal is output.

[0069] 2. The acquisition frequency of 1s is 4000Hz, and m is 10. Then, 5 correlation coefficients of PM0M1, PM2M3, PM4M5, PM6M7, and PM8M9, and 5 arc average values ​​of hM0M1, hM2M3, hM4M5, hM6M7, and hM8M9 can be calculated. When the minimum value of hM4M5, hM6M7, and hM8M9 is larger than the maximum value of hM0M1 and hM2M3, an alarm signal is issued. If the middle value of PM4M5, PM6M7, and PM8M9 is smaller than the minimum value of PM0M1 and PM2M3 at the same time, a serious alarm is issued.

[0070] 3. There is no curve to reflect the effect.

[0071] like Figure 4 As shown, the present invention also provides a motor slip ring fault identification device based on characteristic curve classification, including a pulse discontinuous cycle curve identification module 1 and a pulse top continuous curve identification module 2, wherein the pulse discontinuous cycle curve identification module 1 has the pulse discontinuous cycle curve identification function in the method described in the present invention, and the pulse top continuous curve identification module 2 has the pulse top continuous curve identification function in the method described in the present invention.

[0072] In a possible implementation scheme, the specific structure of the motor slip ring fault identification device based on characteristic curve classification can be: the motor slip ring fault identification device based on characteristic curve classification described in the present invention includes a CPU, a multi-channel sampling circuit, an arc sensor, an optical fiber, and an arc receiver; wherein the CPU integrates the pulse discontinuous cycle curve identification module 1 and the pulse top continuous curve identification module 2, the arc sensor is connected to the arc receiver via an optical fiber, and the arc receiver is connected to the CPU via a sampling circuit.

[0073] The identification device may also include a temperature sensor, a humidity sensor and an output relay (not shown in the figure), the temperature sensor and the humidity sensor are connected to the CPU through a sampling circuit, and the peripheral circuit is connected to the CPU to support the operation of the CPU; the temperature sensor and the humidity sensor respectively collect the temperature and humidity in the slip ring box, and transmit them to the CPU through the sampling circuit, and when the temperature is lower than the set value or the humidity is higher than the set value, the CPU controls the relay to operate the heater to heat and dehumidify the slip ring box.

[0074] Of course, the identification device also includes peripheral circuits, which include crystal oscillators, power supplies, etc., and are mainly circuits that allow the CPU to work normally.

[0075] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A motor slip ring fault identification method based on characteristic curve classification, characterized in that: include: S100. Pulse discontinuous cycle curve identification: determine the cumulative number of arc pulses generated in each cycle, and compare it with the arc pulse number setting value, and mark the identification bit when it exceeds the arc pulse number setting value, and issue a fault alarm when the cumulative number of identification bits in multiple cycles exceeds the setting value; S200. Pulse top continuous curve identification: According to the arc intensity collected at each sampling point in each cycle, the correlation coefficient and the average arc intensity within the time difference of multiple sampling points are calculated, and the fault alarm is determined based on the correlation coefficient and the average arc intensity within the time difference of multiple sampling points; specifically including: S201. Calculate the correlation coefficient within the time difference of multiple sampling points before and after: Sample the arc intensity at a frequency of p points per second, collect p*t2 points within the cycle time t2, and divide the sampling points within the cycle time t2 into m equal parts, m ≥ 10, and m is an even number, and then solve the before and after according to the following formula The correlation coefficient within the time difference of sampling points , a total of m / 2 correlation coefficients are solved; Among them, =0,2,4…m, i represents the time of the sampling point, , , Indicates the arc intensity collected at the sampling point. and They are all marking symbols; S202. Calculate the average arc intensity within the time difference of multiple sampling points before and after: Solve the following formula before and after The average arc intensity within the time difference of sampling points , a total of m / 2 arc intensity average values ​​are solved; In the formula =0,2,4…m, i represents the time of the sampling point, and They are all marking symbols; S203. Obtain the values ​​of the minimum correlation coefficient and the second-to-last correlation coefficient: compare the first two correlation coefficients among the obtained m / 2 correlation coefficients to obtain the minimum value of the two correlation coefficients , and compare the remaining (m / 2)-2 correlation coefficients to get the second-to-last largest value ; S204. Obtain the values ​​of the maximum arc intensity average and the second-to-last arc intensity average: compare the first two arc intensity averages among the m / 2 arc intensity averages obtained to obtain the maximum value of the two arc intensity averages. , and compare it with the average value of the remaining (m / 2)-2 arc intensities to get the second smallest value ; S205, judging whether to issue a fault alarm according to the value of the correlation coefficient obtained in S203 and the value of the average arc intensity obtained in S204; like > When the fault occurs, a fault alarm is issued; Ruodang > and < A serious fault alarm is issued.

2. The method according to claim 1, characterized in that The pulse discontinuous cycle curve identification specifically includes: S101. Determination of arc pulse generation: Set the arc intensity constants h1 and h2, the time constant t1, and when the arc intensity is greater than h1 and the maintenance time is greater than t1, it is determined that an arc pulse is generated; S102. Arc pulse end determination: After determining that an arc pulse is generated, if the currently collected arc intensity drops below h2, the arc pulse is determined to be ended, and the arc pulse generation and arc pulse end process is counted as 1 complete arc pulse; S103. Calculation of the number of arc pulses: Using cycle t2 as the marking time, determine and count the number of arc pulses n generated within the cycle t2 according to the method of S101-S102; If only the arc pulse described in S101 is generated within the cycle period t2, and the arc pulse described in S102 is not terminated, it is determined that the number n of arc pulses generated within the cycle period t2 is 1; If the number n of arc pulses generated within the cycle period t2 exceeds the set value Nset, the arc pulse cycle is marked as a flag bit, and the flag bits are accumulated; The arc intensity constant h1>h2, the time constant t1<cycle period t2, and the cycle period t2>1s, and the cycle period t2 is greater than or equal to the time for the motor shaft to rotate one circle; S104. Alarm determination: When the accumulated flag is greater than the set value Lset within the cumulative time t3, an alarm is issued to notify the user to arrange maintenance.

3. A motor slip ring fault identification device based on characteristic curve classification, characterized in that: The invention comprises a pulse discontinuous cycle curve recognition module (1) and a pulse top continuous curve recognition module (2), wherein the pulse discontinuous cycle curve recognition module (1) has the function of pulse discontinuous cycle curve recognition as claimed in any one of claims 1 to 2, and the pulse top continuous curve recognition module (2) has the function of pulse top continuous curve recognition as claimed in any one of claims 1 to 2.

4. The device according to claim 3, characterized in that The invention comprises a CPU, a multi-channel sampling circuit, an arc light sensor, an optical fiber, and an arc light receiver; wherein the CPU is integrated with the pulse discontinuous cycle curve identification module (1) and the pulse top continuous curve identification module (2); the arc light sensor is connected to the arc light receiver via the optical fiber; and the arc light receiver is connected to the CPU via the sampling circuit.

5. The device according to claim 4, characterized in that It also includes a temperature sensor, a humidity sensor and an output relay. The temperature sensor and the humidity sensor are connected to the CPU through a sampling circuit. The peripheral circuit of the motor slip ring fault identification device is connected to the CPU to support the operation of the CPU. The temperature sensor and the humidity sensor respectively collect the temperature and humidity in the slip ring box, and transmit them to the CPU through the sampling circuit. When the temperature is lower than the set value or the humidity is higher than the set value, the CPU controls the relay to operate the heater to heat and dehumidify the slip ring box.

Images