High-voltage switchgear gas leakage prediction method and computer device

Abstract

The invention relates to a high-voltage switchgear gas leakage prediction method and a computer device, and belongs to the technical field of time sequence data intelligent processing algorithms. Under the condition that a high-voltage switchgear has a gas leakage risk, an average value of gas density data under different time scales is calculated; the leakage rate is calculated according to the average value of the gas density data under different time scales, and if the leakage rate exceeds a set leakage rate threshold value, it is judged that gas leakage affecting operation of the high-voltage switchgear occurs. According to the method, the leakage rate is calculated according to different time scales only by using the gas density data in the prediction time period without collecting a large amount of data, the influence of error data in the to-be-detected data on a calculation result can be avoided through different time scales, and whether gas leakage occurs or not is judged; the problems that in the prior art, a deep learning algorithm is used for predicting the air leakage problem, a large amount of training data is needed, consequently, long training time needs to be consumed, and prediction precision is low are solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent processing algorithms for time series data, and in particular relates to a method and a computer device for predicting gas leakage of high-voltage switchgear. Background Art

[0002] With the widespread use of gas-filled high-voltage switchgear in substations, gas leakage is one of the most common problems in the on-site operation, maintenance and inspection of high-voltage switchgear. High-voltage switchgear mainly uses density meters as monitoring devices to monitor the gas pressure in its gas chamber, and many density meters already have remote transmission functions. However, at present, the acquisition and judgment of gas leakage on the power station site mostly rely on manual inspection or simple threshold judgment.

[0003] In the research on leakage prediction algorithms for high-voltage switchgear, deep learning methods such as LSTM have been used to determine whether a high-voltage switch has leakage. However, such algorithms require the collection of a large amount of data in advance to train the prediction model. In actual engineering applications, the process of collecting a large amount of data is relatively complicated and time-consuming, which makes it difficult to implement in actual engineering applications. Moreover, the accuracy of the deep learning model is greatly affected by the training data. When there is less training data, it is difficult to ensure that the deep learning model has a high prediction accuracy. Summary of the invention

[0004] The purpose of the present invention is to provide a method and computer device for predicting gas leakage of high-voltage switchgear, so as to solve the problem in the prior art that a large amount of training data is required to predict gas leakage using a deep learning algorithm, which requires a long training time and has low prediction accuracy.

[0005] In order to solve the above technical problems, the present invention provides a method for predicting gas leakage of high-voltage switchgear, which comprises the following steps:

[0006] If it is determined that there is a gas leakage risk in the high-voltage switchgear within the predicted time period, the gas density data of the high-voltage switchgear in the predicted time period is segmented according to at least two different time scales, and the gas density average value of the gas density data of each period at different time scales is calculated, and then the leakage rate corresponding to each time scale is calculated based on the gas density average value. If there is a leakage rate exceeding the set leakage rate threshold, it is determined that a gas leakage that affects the operation of the high-voltage switchgear has occurred.

[0007] Furthermore, the specific process of determining whether there is a gas leakage risk in the high-voltage switchgear within the predicted time period is as follows: based on the gas density average of the gas density data for each period of time at the minimum time scale among at least two different time scales, the range of the gas density average value at the minimum time scale is calculated; if the range is greater than a set range threshold, it is determined that there is a gas leakage risk.

[0008] Furthermore, the specific process of calculating the leakage rate based on the average value of the gas density is: fitting the average value of the gas density at a certain time scale into a straight line. When the slope of the straight line is negative, the absolute value of the slope of the straight line is the corresponding leakage rate at the time scale.

[0009] Furthermore, the different time scales include three time scales, which are respectively a first time scale, a second time scale and a third time scale, the third time scale is at least ten times greater than the second time scale, and the second time scale is at least ten times greater than the first time scale.

[0010] Furthermore, the method also includes: when it is determined that a gas leakage that affects the operation of the high-voltage switchgear occurs, the time from gas leakage to gas density reaching a set density alarm value is calculated according to the maximum leakage rate among the leakage rates corresponding to each time scale.

[0011] Furthermore, the setting of the extreme difference threshold is determined according to the type of high-voltage switchgear and its density table error.

[0012] In order to solve the above technical problems, the present invention also provides a method for predicting gas leakage of high-voltage switchgear, which method includes the following steps: segmenting the gas density data of the high-voltage switchgear in the prediction time period according to at least two different time scales, and calculating the gas density average value of the gas density data of each time period at different time scales, and then calculating the leakage rate corresponding to each time scale according to the gas density average value of the gas density data of each time period at different time scales. If there is a leakage rate exceeding the set leakage rate threshold, it is determined that a gas leakage that affects the operation of the high-voltage switchgear has occurred.

[0013] Furthermore, the specific process of calculating the leakage rate based on the average value of the gas density is: fitting the average value of the gas density at a certain time scale into a straight line. When the slope of the straight line is negative, the absolute value of the slope of the straight line is the corresponding leakage rate at the time scale.

[0014] In order to solve the above technical problem, the present invention also provides a computer device, including a processor, and the processor is used to execute a computer program to implement the steps of a high-voltage switchgear gas leakage prediction method.

[0015] In order to solve the above technical problem, the present invention also provides a computer device, including a processor, and the processor is used to execute a computer program to implement the steps of a high-voltage switchgear gas leakage prediction method.

[0016] The beneficial effects of the above technical solution are as follows: the present invention is an improved invention. When there is a risk of gas leakage in the high-voltage switchgear, the present invention calculates the average value of the gas density data at different time scales, and calculates the leakage rate based on the average value of the gas density data at different time scales. If there is a leakage rate exceeding the set leakage rate threshold, it is determined that a gas leakage that affects the operation of the high-voltage switchgear occurs. The present invention does not need to collect a large amount of data and only uses the gas density data of the predicted time period. The gas density data of the predicted time period is used to calculate the leakage rate of the insulating gas in the gas chamber of the high-voltage switchgear according to different time scales. The influence of error data in the data to be tested on the calculation results can be avoided by using different time scales, and it is determined whether a gas leakage occurs, which solves the problem in the prior art that a large amount of training data is required to predict the leakage problem using a deep learning algorithm, which requires a long training time and has low prediction accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a flow chart of an embodiment of a method for predicting gas leakage of high-voltage switchgear according to the present invention. DETAILED DESCRIPTION

[0018] The present invention focuses on providing a method for predicting gas leakage of high-voltage switchgear, which comprises the following steps:

[0019] If it is determined that the high-voltage switchgear has a gas leakage risk in the prediction time period, the gas density data of the high-voltage switchgear in the prediction time period is segmented according to at least two different time scales, and the gas density average value of the gas density data of each time period under different time scales is calculated, and then the leakage rate corresponding to each time scale is calculated according to the gas density average value. If there is a leakage rate exceeding the set leakage rate threshold, it is determined that a gas leakage that affects the operation of the high-voltage switchgear occurs. In the case that the high-voltage switchgear has a gas leakage risk, the present invention calculates the average value of the gas density data under different time scales, and calculates the leakage rate according to the average value of the gas density data under different time scales. If there is a leakage rate exceeding the set leakage rate threshold, it is determined that a gas leakage that affects the operation of the high-voltage switchgear occurs. The present invention does not need to collect a large amount of data and only uses the gas density data of the prediction time period. The gas density data of the prediction time period is used to calculate the leakage rate of the insulating gas in the gas chamber of the high-voltage switchgear according to different time scales. The influence of the error data in the test data on the calculation result can be avoided by using different time scales, and it is determined whether a gas leakage occurs, which solves the problem that a large amount of training data is required to predict the leakage problem using a deep learning algorithm in the prior art, which requires a long training time and has low prediction accuracy.

[0020] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention is further described in detail below with reference to the accompanying drawings and embodiments.

[0021] High-voltage switchgear gas leakage prediction method embodiment 1:

[0022] A method for predicting gas leakage of a high-voltage switchgear according to this embodiment is as follows: Figure 1 The specific steps are as follows:

[0023] Step 1: Obtain gas density data of the high-voltage switchgear to be tested for a certain period of time.

[0024] Step 2: Determine the time scale according to the sampling period of the gas density data. The time scale is determined according to actual needs, and there are at least two time scales. In this embodiment, three time scales are included, namely, a short time period (i.e., the first time scale), a medium time period (i.e., the second time scale) and a long time period (i.e., the third time scale). The time span of the gas density data is much larger than the maximum time scale. If the time period of the gas density data is short, it may interfere with the accuracy of the prediction results.

[0025] The time scales must be greater than the gas density data sampling period. There is an order of magnitude increase between different time scales. For example, the medium time period is at least ten times greater than the short time period, and the long time period is at least ten times greater than the long time period. For example, when the gas density data collection period is one minute, the three time scales are five minutes, one hour, and twenty-four hours.

[0026] Step 3: Calculate the average gas density of each time period after dividing the gas density data according to three time scales. For example, when the gas density data is 30 days of gas density data, if the time scale is 5 minutes, the average gas density data of 8640 can be calculated, if the time scale is one hour, the average gas density data of 720 can be calculated, and if the time scale is 24 hours, the average gas density data of 30 can be calculated.

[0027] Step 4: Calculate the range of the average value of the gas density data at the minimum time scale. The range is the difference between the maximum value of the gas density data and the minimum value of the gas density data. When the range of the average value of the gas density data at the minimum time scale is greater than the set range threshold, it is determined that there is a risk of gas leakage; then the average value of the gas density data at each time scale is subjected to regression analysis, and the average value of the gas density data at each time scale is fitted through regression analysis to obtain the corresponding straight line at each time scale. When the slope of the straight line is positive, it means that there is no gas leakage in the high-voltage switchgear, and the leakage rate is zero. When the slope of the straight line is negative, the absolute value of the slope of the straight line is the leakage rate at each time scale; for example, when there are three time scales, the average values ​​of the gas density data at the three time scales are fitted to obtain three straight lines, and then three leakage rates are calculated. If the range of the average value of the gas density data at the minimum time scale is less than or equal to the set range threshold, it is determined that there is no gas leakage in the high-voltage switchgear.

[0028] The extreme value threshold is set here based on experience and is selected according to the type of high-voltage switchgear and the error of the density table. Generally, 0.06 can be used for GIS equipment and 0.03 for bushing equipment.

[0029] Step 5: If the calculated leakage rate is greater than or equal to the set leakage rate threshold of the remote density meter of the acquisition device, it is determined that a gas leakage that affects the operation of the high-voltage switchgear occurs. When the leakage rate is less than the set leakage rate threshold, it is determined that the leakage rate is within the allowable range and can be ignored. The leakage rate threshold is set to prevent errors caused by temperature fluctuations during the day. In this embodiment, the leakage rate threshold is set to 1e-3.

[0030] When determining that a gas leakage has occurred that affects the operation of the high-voltage switchgear, the time for the gas density to leak to the set density alarm value is calculated based on the maximum leakage rate among the leakage rates corresponding to each time scale. If all the calculated leakage rates do not exceed the error range of the remote density meter of the acquisition device, it is determined that there is no gas leakage in the high-voltage switchgear.

[0031] The present invention calculates the average value and the range of the average value of the gas density data at different time scales. When the range of the average value of the gas density data at the minimum time scale is greater than the set range threshold, the leakage rate is calculated based on the average value of each gas density data. If there is a leakage rate exceeding the set leakage rate threshold, it is determined that a gas leak occurs, and the time for the gas density to leak to the set density alarm value is calculated based on the maximum leakage rate. The present invention can automatically calculate the leakage rate of the insulating gas in the gas chamber of the high-voltage switchgear using only the gas density data of the time period to be tested, and then predict the time for the insulating gas in the gas chamber to leak to the gas replenishment pressure and the minimum functional pressure, solving the problem in the prior art that the deep learning algorithm is used to predict the leakage problem, which requires the additional collection of a large amount of data, resulting in a complex data collection process and a lot of time consumption in actual engineering.

[0032] High-voltage switchgear gas leakage prediction method embodiment 2:

[0033] A method for predicting gas leakage of a high-voltage switchgear in this embodiment has the following specific steps:

[0034] Step 1: Obtain gas density data of the high-voltage switchgear to be tested for a certain period of time.

[0035] Step 2: Determine the time scale according to the sampling period of the gas density data. The time scale is determined according to actual needs, and there are at least two time scales. In this embodiment, three time scales are included, namely, a short time period (i.e., the first time scale), a medium time period (i.e., the second time scale) and a long time period (i.e., the third time scale). The time span of the gas density data is much larger than the maximum time scale. If the time period of the gas density data is short, it may interfere with the accuracy of the prediction results.

[0036] The time scales must be greater than the gas density data sampling period. There is an order of magnitude increase between different time scales. For example, the medium time period is at least ten times greater than the short time period, and the long time period is at least ten times greater than the long time period. For example, when the gas density data collection period is one minute, the three time scales are five minutes, one hour, and twenty-four hours.

[0037] Step 3: Calculate the average gas density of each time period after dividing the gas density data according to three time scales. For example, when the gas density data is 30 days of gas density data, if the time scale is 5 minutes, the average gas density data of 8640 can be calculated, if the time scale is one hour, the average gas density data of 720 can be calculated, and if the time scale is 24 hours, the average gas density data of 30 can be calculated.

[0038] Step 4: Perform regression analysis on the average values ​​of gas density data at each time scale. Through regression analysis, the average values ​​of gas density data at each time scale are fitted to obtain the corresponding straight lines at each time scale. When the slope of the straight line is positive, it means that there is no gas leakage in the high-voltage switchgear and the leakage rate is zero. When the slope of the straight line is negative, the absolute value of the slope of the straight line is the leakage rate at each time scale. For example, when there are three time scales, the average values ​​of gas density data at the three time scales are fitted to obtain three straight lines respectively, and then the three leakage rates are calculated.

[0039] Step 5: If the calculated leakage rate is greater than or equal to the set leakage rate threshold of the remote density meter of the acquisition device, it is determined that a gas leakage that affects the operation of the high-voltage switchgear occurs. When the leakage rate is less than the set leakage rate threshold, it is determined that the leakage rate is within the allowable range and can be ignored.

[0040] When determining that a gas leakage has occurred that affects the operation of the high-voltage switchgear, the time for the gas density to leak to the set density alarm value is calculated based on the maximum leakage rate among the leakage rates corresponding to each time scale. If all the calculated leakage rates do not exceed the error range of the remote density meter of the acquisition device, it is determined that there is no gas leakage in the high-voltage switchgear.

[0041] The difference between Example 2 of the gas leakage prediction method for high-voltage switchgear and Example 1 of the gas leakage prediction method for high-voltage switchgear is that there is no judgment on the existence of gas leakage risk. In Example 1 of the gas leakage prediction method for high-voltage switchgear, it is only necessary to judge whether the range of the average value of the gas density data under the minimum time scale is greater than the set range threshold. If it is not greater than the set range threshold, it is judged that there is no gas leakage risk and no subsequent calculation is required. Although when the range is greater than the set range threshold and there is a gas leakage risk, subsequent leakage rate calculation is required, and there is an increase in the amount of calculation, but in the case of unknown high-voltage equipment gas leakage, some calculation amount is reduced relative to Example 2 of the gas leakage prediction method for high-voltage switchgear.

[0042] Computer device embodiment 1:

[0043] The present invention also provides a computer device, including a processor, the processor is used to execute a computer program to implement the steps of the high-voltage switchgear gas leakage prediction method. The specific process, principle and effect have been described in detail in Embodiment 1 of the high-voltage switchgear gas leakage prediction method, and will not be repeated here.

[0044] Computer device embodiment 2:

[0045] The present invention also provides a computer device, including a processor, the processor is used to execute a computer program to implement the steps of the high-voltage switchgear gas leakage prediction method. The specific process, principle and effect have been described in detail in Embodiment 2 of the high-voltage switchgear gas leakage prediction method, and will not be repeated here.

[0046] Specific implementation methods are given above, but the present invention is not limited to the described implementation methods. The basic idea of ​​the present invention lies in the above basic scheme. For ordinary technicians in this field, it does not take creative work to design various deformed models, formulas, and parameters according to the teachings of the present invention. Changes, modifications, substitutions, and variations of the implementation methods without departing from the principles and spirit of the present invention still fall within the scope of protection of the present invention.

Claims

1. A method for predicting gas leakage of high-voltage switchgear, characterized in that: The method comprises the following steps: If it is determined that there is a risk of gas leakage in the high-voltage switchgear within the predicted time period, the gas density data of the high-voltage switchgear in the predicted time period is segmented according to at least two different time scales, and the gas density average value of the gas density data of each time period at different time scales is calculated, and then the leakage rate corresponding to each time scale is calculated based on the gas density average value. If there is a leakage rate exceeding the set leakage rate threshold, it is determined that a gas leakage that affects the operation of the high-voltage switchgear has occurred.

2. The method for predicting gas leakage of high-voltage switchgear according to claim 1, characterized in that: The specific process of determining whether there is a gas leakage risk in the high-voltage switchgear within the predicted time period is as follows: based on the gas density average value of the gas density data for each period of time at the minimum time scale among the at least two different time scales, the range of the gas density average value at the minimum time scale is calculated; if the range is greater than the set range threshold, it is determined that there is a gas leakage risk.

3. The method for predicting gas leakage of high-voltage switchgear according to claim 1, characterized in that: The specific process of calculating the leakage rate based on the average value of gas density is as follows: fit the average value of gas density at a certain time scale into a straight line. When the slope of the straight line is negative, the absolute value of the slope of the straight line is the corresponding leakage rate at the time scale.

4. The method for predicting gas leakage of high-voltage switchgear according to claim 1, characterized in that: The different time scales include three time scales, which are respectively a first time scale, a second time scale and a third time scale. The third time scale is at least ten times greater than the second time scale, and the second time scale is at least ten times greater than the first time scale.

5. The method for predicting gas leakage of high-voltage switchgear according to claim 1, characterized in that: The method also includes: when determining that a gas leakage that affects the operation of the high-voltage switchgear occurs, calculating the time from gas leakage to gas density reaching a set density alarm value according to the maximum leakage rate among the leakage rates corresponding to each time scale.

6. The method for predicting gas leakage of high-voltage switchgear according to claim 2, characterized in that: The set extreme difference threshold is determined according to the type of high-voltage switchgear and its density table error.

7. A method for predicting gas leakage of high-voltage switchgear, characterized in that: The method comprises the following steps: segmenting the gas density data of the high-voltage switchgear in the prediction time period according to at least two different time scales, and calculating the gas density average value of the gas density data of each time period under different time scales, and then calculating the leakage rate corresponding to each time scale according to the gas density average value of the gas density data of each time period under different time scales, and if there is a leakage rate exceeding the set leakage rate threshold, it is determined that a gas leakage that affects the operation of the high-voltage switchgear occurs.

8. The method for predicting gas leakage of high-voltage switchgear according to claim 7, characterized in that: The specific process of calculating the leakage rate based on the average value of gas density is as follows: fit the average value of gas density at a certain time scale into a straight line. When the slope of the straight line is negative, the absolute value of the slope of the straight line is the corresponding leakage rate at the time scale.

9. A computer device comprising a processor, characterized in that: The processor is used to execute a computer program to implement the steps of the high-voltage switchgear gas leakage prediction method according to any one of claims 1 to 6.

10. A computer device comprising a processor, characterized in that: The processor is used to execute a computer program to implement the steps of the high-voltage switchgear gas leakage prediction method according to any one of claims 7 to 8.

Images