A method for evaluating insulation defects of a ring main unit based on on-line monitoring of partial discharge

Abstract

The application relates to a ring main unit insulation defect evaluation method based on local discharge online monitoring. The method constructs a new model on the basis of online monitoring data, obtains a score for insulation defects according to historical data, obtains an evaluation result for switch cabinet insulation defects, and helps staff to have a more intuitive understanding of the insulation defect state of the high-voltage switch cabinet and to carry out corresponding operation and maintenance work.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of insulation defect evaluation of medium-voltage distribution ring network cabinets, and particularly relates to a ring network cabinet insulation defect evaluation method based on partial discharge online monitoring. BACKGROUND

[0002] 1. Prior art:

[0003] a) partial discharge (PD)

[0004] An electrical discharge in which the insulation between conductors is only partially bridged. The discharge can or can not occur in the vicinity of the conductors.

[0005] Note 1: PD is generally caused by a particularly concentrated local electric field in the interior of an insulator or on the surface of an insulator. Usually, this discharge appears as a pulse with a duration of less than 1 μs. However, it can also occur in continuous form, such as the so-called non-pulse discharge in a gaseous medium, which is usually not detected by the measurement methods described in this standard.

[0006] Note 2: "Corona" is a form of PD that often occurs in a gaseous medium around a conductor that is remote from a solid or liquid insulator. "Corona" should not be used as a general term for all forms of PD.

[0007] Note 3: PD is usually accompanied by phenomena such as sound, light, heat, and chemical reactions.

[0008] A simple understanding is that it is a discharge phenomenon that has discharge but has not yet broken down.

[0009] b) probability PD intensity

[0010] The probability intensity refers to the maximum discharge pulse value when the cumulative probability value in a complete detection sample reaches a specified value. It is a calculated value obtained by interpolation calculation according to the measured data. In practice, only pulses higher than the specified amplitude or within the specified amplitude range are considered.

[0011] A simple understanding is that it is the maximum value of the PD pulse that can repeatedly occur. The device is generally set to have more than 20 pulses per second, which can reflect the maximum value of the PD and avoid the influence of extremely large values.

[0012] 2. Prior art:

[0013] At present, there is no evaluation method of ring network cabinet based on partial discharge online monitoring data, most of the schemes are how to deploy the online monitoring system. There is no evaluation system for defects. Such as "SF6 gas ring network cabinet partial discharge intelligent online monitoring system" authorized announcement number: CN 205301507 U. SUMMARY

[0014] The purpose of the present application is to provide a kind of ring network cabinet insulation defect evaluation method based on partial discharge online monitoring, which is based on the online monitoring data to build a new model, according to historical data to obtain the score for insulation defect, obtain the evaluation result for switchgear insulation defect, to help staff to have more intuitive understanding of the insulation defect state of high voltage switchgear, and carry out corresponding operation and maintenance work.

[0015] To achieve the above purpose, the technical scheme of the present application is: a kind of ring network cabinet insulation defect evaluation method based on partial discharge online monitoring, the partial discharge probability intensity data collected by the partial discharge online monitoring sensor based on pulse current method installed in ring network cabinet is used, an evaluation model is constructed from two dimensions of intensity and growth rate, finally the score is used as output data, and it is converted into the evaluation result in line with the operation and maintenance regulations of power grid company distribution equipment, to guide the operation and maintenance personnel to carry out differentiated operation and maintenance on the insulation defect state of ring network cabinet;The method is specifically implemented as follows:

[0016] S1, store 30-day probability intensity data: the partial discharge probability intensity data collected by the partial discharge online monitoring sensor based on pulse current method installed in ring network cabinet, the probability intensity reflects the possible discharge size of the insulation part of ring network cabinet at a certain time;The discharge probability intensity data collected every day is averaged to obtain the average discharge probability intensity of the day, and then the average discharge probability intensity data in the past 30 days is stored in cycles with day as the unit;

[0017] S2, calculate growth coefficient according to least square estimation: assuming that ring network cabinet has insulation defect, the growth rate of average discharge intensity is slow and uniform, that is, average discharge intensity is proportional to time, that is, M = a*t + b, wherein M is average discharge probability intensity, t is day 1~30, a is growth coefficient, and b is initial intensity parameter, and the growth coefficient a is obtained according to least square estimation of M, t and b;

[0018] S3, calculate the insulation state score of single-phase and four kinds of health states by bringing the fixed value into the evaluation model: according to the specific scene, three fixed values are designed: Md1: abnormal critical value, Md2: serious critical value, A1: attention growth value;

[0019] Normal state: When M30 < Md1 and a <= A1, the score F = 100 - 15 / (Md1 × A1) × (M30 × a); where, if a < 0, a is forced to be 0; M30 is the average probability intensity of the latest 1 day.

[0020] Attention state: When M30 < Md1 and a > A1, the score F = 85 - 10 / Md1 × M30 × (1 - A1 / a);

[0021] Abnormal state: When Md1 <= M30 < Md2, the score F = 75 - 15 / (Md2 - Md1) × (M30 - Md1);

[0022] Severe state: When M30 >= Md2, the score F = 60 × Md2 / M30.

[0023] Compared with the prior art, the present invention has the following beneficial effects: The present invention adopts a computer - recognizable practical model to process online monitoring parameters and obtains an evaluation result that can be directly recognized by operation and maintenance personnel, which is more intelligent than the existing deployed online monitoring system. Description of the Drawings

[0024] Figure 1 It is a flow chart of the method of the present invention. Detailed Embodiments

[0025] The technical solutions of the present invention will be specifically described below in conjunction with the drawings.

[0026] As Figure 1 shown, a method for evaluating the insulation defect of a ring main unit based on partial discharge online monitoring according to the present invention uses the partial discharge probability intensity data collected by a partial discharge online monitoring sensor based on the pulse current method installed in the ring main unit, constructs an evaluation model from two dimensions of intensity and growth rate, and finally uses the score as the output data and converts it into an evaluation result that conforms to the operation and maintenance regulations of distribution equipment of power grid companies to guide operation and maintenance personnel to perform differential operation and maintenance on the insulation defect state of the ring main unit.

[0027] a) Store the probability intensity data for 30 days. The data source of this patented method is a partial discharge online monitoring sensor based on the pulse current method installed in the ring main unit. The probability intensity reflects the possible discharge magnitude of the insulation part of the ring main unit at a certain moment. The method averages the probability intensity data collected every day to obtain the average probability intensity of that day, and then stores the data within the past 30 days cyclically in units of days.

[0028] b) Calculate the growth coefficient according to the least square estimation. This patent method assumes that if the ring main unit has insulation defects, the growth rate of its average discharge intensity is slow and uniform, that is, the average discharge intensity is proportional to time. That is, M = a*t + b. Where M is the average discharge probability intensity, t is the number of days 1~30, a is the growth coefficient, and b is the initial intensity parameter. According to the least square estimation of M, t and b, the growth coefficient a can be obtained.

[0029] c) Enter the fixed value into the model to calculate the insulation state score of single-phase and four health states. The method uses personnel to design three fixed values according to the specific scene: Md1: abnormal critical value (such as 100pc); Md2: serious critical value (such as 600pc); A1: attention growth value (such as 3).

[0030] Normal state: when M30<Md1, and a<=A1. Score F = 100 - 15 / (Md1xA1) x (M30xa); Note that if a<0, a=0 is forced.

[0031] Attention state: when M30<Md1, and a>A1. Score F = 85-10 / Md1 x M30 x (1-A1 / a);

[0032] Abnormal state: when Md1=<M30<Md2. Score F = 75 - 15 / (Md2-Md1) x (M30-Md1);

[0033] Serious state: when M30>=Md2. Score F = 60 x Md2 / M30.

[0034] Note: M30 is the average probability intensity of the latest 1 day.

[0035] Implementation example:

[0036] a) Extract data sources

[0037] The 30-day average probability intensity of a certain ring main unit is as shown in Table 1:

[0038] Table 1

[0039]

[0040] d) Calculate parameters

[0041] After obtaining the 30-day probability intensity data, construct the model M = a*t + b, and substitute the 30 sets of data into the least square estimation to calculate the growth coefficient a = 2.061 and the initial probability intensity b = 73.62.

[0042] e) Calculate the score

[0043] The method takes default values Md1 = 100, Md2 = 600, and A1 = 3. Since M30 = 133, Md1 =< M30 < Md2, the score F = 75 - 0.99 = 74.01. The interval corresponds to an abnormal state.

[0044] f) The state should be appropriately arranged for power-off maintenance by the operation and maintenance personnel since it has entered an abnormal state.

[0045] The above is the preferred embodiment of the present application. Any changes made according to the technical solutions of the present application, as long as the resulting functional effects do not exceed the scope of the technical solutions of the present application, are within the scope of protection of the present application.

Claims

1. A method for evaluating insulation defects of a ring main unit based on online monitoring of partial discharge, characterized in that, The probability intensity data of partial discharge collected by the partial discharge on-line monitoring sensor based on the pulse current method installed in the ring network cabinet is used to construct an evaluation model from two dimensions of intensity and growth rate, and finally the score is taken as the output data, which is converted into the evaluation results conforming to the operation and maintenance regulations of power grid company distribution equipment, to guide the differential operation and maintenance of the insulation defect state of the ring network cabinet. The method is implemented as follows: S1, store 30-day probability intensity data: the probability intensity data of partial discharge collected by the partial discharge on-line monitoring sensor based on the pulse current method installed in the ring network cabinet, which reflects the discharge size of the insulation part of the ring network cabinet at a certain time; the collected discharge probability intensity data is averaged to obtain the average discharge probability intensity of the day, and then the average discharge probability intensity data in the past 30 days is stored in cycles in units of days; S2, calculate the growth coefficient according to the least square estimation: assuming that the ring network cabinet has insulation defects, the growth rate of the average discharge intensity is slow and uniform, i.e. the average discharge intensity is proportional to time, i.e. M = a*t + b, wherein M is the average discharge probability intensity, t is the number of days 1~30, a is the growth coefficient, and b is the initial intensity parameter; the growth coefficient a is obtained by the least square estimation according to M, t and b; S3, calculate the insulation state score of single-phase and four health states by bringing the fixed value into the evaluation model: three fixed values are designed according to the specific scene: Md1: abnormal critical value, Md2: serious critical value, A1: attention growth value; Normal state: when M30 < Md1 and a <= A1, the score F = 100 - 15 / (Md1*A1)* (M30*a); if a < 0, a is forced to be 0; M30 is the average probability intensity of the latest 1 day; Attention state: when M30 < Md1 and a > A1, the score F = 85 - 10 / Md1*M30*(1-A1 / a); Abnormal state: when Md1 <= M30 < Md2, the score F = 75 - 15 / (Md2-Md1)*(M30-Md1); Serious state: when M30 >= Md2, the score F = 60*Md2 / M30.

Images