A leak detection method and system

Abstract

The present application relates to the field of refrigerator refrigeration system pipeline leakage detection, and particularly relates to a leakage detection method and system. The method comprises: calculating pressure data of a pipeline in a pressure maintaining process, wherein the pressure data comprises a plurality of pressures; performing smoothing processing on the pressure data by using an improved weighted moving average to obtain denoised pressure data; and if the average of the ratio of each pressure in the denoised pressure data to a rated pressure is less than a set value, then the pipeline has a leakage point. That is, the scheme of the present application improves the accuracy of pipeline leakage detection.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of refrigerator refrigeration system pipeline leakage detection. More specifically, the present application relates to a leakage detection method and system. BACKGROUND

[0002] The refrigeration of a refrigerator is achieved by the refrigerant in the internal pipeline of the refrigerator. Among them, the pipeline in the refrigeration system of the refrigerator is mainly composed of aluminum pipe, Bundy pipe (iron pipe) and copper pipe; among them, the copper pipe is usually used for the connection pipe of the refrigeration and freezing evaporator, which is pre-buried in the foaming layer and has the characteristics of corrosion resistance and strong durability; the aluminum pipe is usually used for the evaporator because of its excellent heat conduction performance and high refrigeration efficiency; the Bundy pipe is used for the condenser.

[0003] In order to avoid the leakage of refrigerant in the pipeline and ensure the quality of the refrigerator, leakage detection of the pipeline needs to be carried out when the refrigerator is assembled.

[0004] The existing pipeline leakage detection has the following methods:

[0005] (1) Bubble test method: the pipeline is immersed in a container filled with water solution for leakage detection. For example, in the related art, the patent application file of a kind of air conditioner copper pipe leakage detection method with publication number CN108088029A discloses a method for detecting the sealing performance of the copper pipe. Specifically, first, the copper pipe is fixed, then the air pressure equipment and the other end of the copper pipe are connected by a hose, and then the air pressure equipment is started. When air is transported, the pressure value indicated by the pressure gauge on the air pressure equipment is observed in real time. When the specified pressure value is reached, the pressure is stabilized for a period of time, the stability of the copper pipe and the interface connection is checked, and whether the air tightness is qualified is detected by observing whether the water surface in the container appears water bubbles.

[0006] The above method has low precision, and the test result depends on manual observation, which is subjective and difficult to quantify.

[0007] (2) Pressure detection method: usually, a leak detection gas is injected into the pipeline, and the pressure change of the pipeline after charging is detected to determine whether the pipeline has a leakage. Specifically, when the pressure in the pipeline reaches the test pressure, the charging is stopped, and the pressure change in the pipeline after stopping charging is monitored. If the pressure in the pipeline does not change, the pipeline has no leakage point; if the pressure in the pipeline changes, the pipeline has a leakage point.

[0008] When the above method detects the pressure, the pressure sensor will be affected by external vibration and other factors, so that there will be some noise in the collected pressure data, which will cause false judgment in the leakage detection of the pipeline.

[0009] Therefore, it is necessary to denoise the noise in the collected pressure data; specifically, the noise in the pressure data is usually smoothed using a weighted moving average, however, the weighted moving average only relies on historical pressure data, lacks consideration of environmental factors, and thus has certain limitations on noise removal of the collected pressure data, resulting in inaccurate detection.

[0010] In summary, the existing pipeline leakage detection has the problem of inaccuracy. SUMMARY

[0011] The purpose of the present application is to provide a leakage detection method and system to solve the problem of inaccurate pipeline leakage detection in the prior art; for this purpose, the present application provides solutions in the following two aspects.

[0012] In the first aspect, the present application provides a leakage detection method, comprising:

[0013] obtaining pressure data of the pipeline during the pressure maintaining process using a pressure sensor, the pressure data comprising a plurality of pressures;

[0014] smoothing the pressure data using an improved weighted moving average to obtain denoised pressure data;

[0015] if the average of the ratio of each pressure in the denoised pressure data to the rated pressure is less than a set value, then the pipeline has a leakage point;

[0016] the weight of each pressure in the improved weighted moving average is inversely related to the degree of abnormal change of the corresponding pressure; the degree of abnormal change of the pressure is is: is the average of the degree of abnormality of the pressure data of the pipeline during the inflation process, P is the rated pressure, and and are the jth and j-1th pressures during the pressure maintaining process, respectively, is the average of all pressures before the jth pressure during the pressure maintaining process, and T is the state stability.

[0017] In the above solution, the degree of abnormality of each pressure in the pressure data of the pipeline during the inflation process is used to capture the fluctuation characteristics of the pressure data during the inflation process, and then the average of the degree of abnormality of the pressure data during the inflation process, the pressure data during the pressure maintaining process, and the acceleration data are used to calculate the degree of abnormal change of each pressure in the pressure data during the pressure maintaining process, so as to distinguish the noise data during the pressure maintaining process, and then different weights are applied to different pressures during the pressure maintaining process. The weighted moving average of the pressure data during the pressure maintaining process can reduce the influence of noise data on the result of the weighted moving average, thereby reducing the influence of noise data on the pipeline leakage detection, and making the pipeline leakage detection more accurate.

[0018] Optionally, the state stability is a maximum value of ratios in three directions; the ratio is a ratio of a jth acceleration of the pressure sensor in an m direction during the pressure maintaining process to a mean value of the acceleration data of the pressure sensor in the m direction during the inflation process; wherein the three directions are an x-axis direction, a y-axis direction and a z-axis direction, and the m direction is any one of the x-axis direction, the y-axis direction and the z-axis direction.

[0019] Optionally, the weight is:

[0020]

[0021] wherein w′ j-k+1 is a weight of a j-k+1th pressure during the pressure maintaining process, is an abnormal change degree of the j-k+1th pressure during the pressure maintaining process, exp() is an exponential function, and K is a size of a sliding window.

[0022] In the above scheme, the abnormal change degree of each pressure during the pressure maintaining process is calculated, and a weight is given to the corresponding pressure for subsequent pressure correction.

[0023] Optionally, the weight is:

[0024] w′ j-k+1 is a weight of a j-k+1th pressure during the pressure maintaining process, is a difference between the abnormal change degree of the j-k+1th pressure and the abnormal change degree of a j-kth pressure during the pressure maintaining process, exp() is an exponential function, and K is a size of a sliding window.

[0025] In the above scheme, the difference between the abnormal change degrees of adjacent two pressures during the pressure maintaining process is considered, and a suitable weight is assigned to the corresponding pressure.

[0026] Optionally, the abnormal degree is:

[0027] wherein w i is an abnormal degree of an ith pressure of the pressure data of the pipeline during the inflation process, i>1, p i and p i-1 are the ith and an i-1th pressure during the inflation process, respectively, P is a rated pressure, and I is a total number of pressures during the inflation process.

[0028] In the above scheme, the abnormal degree of each pressure of the pressure data during the inflation process is obtained, and data support is provided for the abnormal change degree of the pressure data during the subsequent pressure maintaining process.

[0029] Optionally, the specific process of smoothing the pressure data by using the improved weighted moving average to obtain the denoised pressure data is as follows:

[0030] a sliding window of the preset weighted moving average;

[0031] performing weighted average on the sliding window in which the target pressure is located to obtain the denoised pressure corresponding to the target pressure, and further obtain the denoised pressure data in the pressure maintaining process; the target pressure is any pressure in the pressure data in the pressure maintaining process, and the target pressure is located at the end of the corresponding sliding window;

[0032] the denoised pressure is: w′ j-k+1 is the weight of the j-k+1th pressure in the pressure maintaining process, is the j-k+1th pressure in the pressure maintaining process, and K is the size of the sliding window.

[0033] In the above scheme, by obtaining the weight of each pressure in the pressure maintaining process, the influence of noise data on the pipeline leakage judgment can be reduced.

[0034] Optionally, the pressure charging process is a time period between the moment when the leak detection gas is delivered and the moment when the rated pressure is reached and the delivery of the leak detection gas is stopped; and the pressure maintaining process is a set time period starting from the moment when the delivery of the leak detection gas is stopped.

[0035] Optionally, the acceleration data represents the vibration condition of the pipeline in the inflation process.

[0036] In a second aspect, a leakage detection system comprises:

[0037] a processor;

[0038] a memory storing computer instructions for leakage detection, when the computer instructions are run by the processor, the system executes the above-mentioned leakage detection method.

[0039] The beneficial effects of the present application are:

[0040] The scheme of the present application can reduce the influence of noise data on the weighted moving average result by applying different weights to different pressures in the pressure maintaining process and using the weights to perform weighted moving average on the pressure data in the pressure maintaining process, thereby reducing the influence of noise data on the pipeline leakage detection and making the pipeline leakage detection more accurate. BRIEF DESCRIPTION OF DRAWINGS

[0041] The above and other objects, features and advantages of the present embodiments will become more apparent from the following detailed description read in conjunction with the accompanying drawings. In the drawings, several embodiments of the present application are illustrated by way of example and not limitation in which like reference numerals refer to similar elements throughout:

[0042] Figure 1 A step flow chart of a leakage detection method in the embodiment is schematically shown;

[0043] Figure 2 A structure block diagram of a leakage detection system in the embodiment is schematically shown. DETAILED DESCRIPTION

[0044] The present application is directed to the pipeline in the refrigerator refrigeration system, that is, in order to avoid the leakage of refrigerant in the pipeline, the quality of the refrigerator is ensured, and all the pipelines in the refrigerator refrigeration system need to be detected for leakage. Taking any pipeline in a certain refrigerator refrigeration system as an example, a leakage detection method in the embodiment is introduced.

[0045] Specifically, as shown in Figure 1 A leakage detection method includes the following steps:

[0046] Step S1, obtaining pressure data of the pipeline during the pressure maintaining process.

[0047] In the embodiment, taking a copper pipe as an example, the copper pipe is fixed, one end of the copper pipe is sealed, and the other end is inflated by an inflation device; when the leakage detection gas is delivered, the pressure data of the copper pipe during the inflation process is obtained by a pressure sensor, and the acceleration data of the position where the pressure sensor is located during the inflation process is collected by an acceleration sensor; until the pressure value on the pressure sensor reaches the rated pressure, stop inflating, and stabilize for a period of time, then obtain the pressure data and acceleration data of the copper pipe during the pressure maintaining process.

[0048] The gas filled into the pipeline can be air or nitrogen.

[0049] The pressure data of the inflation process and the pressure maintaining process both contain multiple pressures. When collecting the pressure data, sampling can be performed at a set interval, such as 1s, 2s or 5s, etc.

[0050] The inflation process is the time period between the moment of delivering the leakage detection gas and the moment of rising to the rated pressure and stopping delivering the leakage detection gas; the pressure maintaining process is a set time period starting from the moment of stopping delivering the leakage detection gas; the set time period can be 30 minutes or 60 minutes.

[0051] Any pipe in the embodiment can be an aluminum pipe, a Bundy pipe (iron pipe) or a copper pipe in a refrigeration system of a refrigerator. In step S2, the pressure data is smoothed by using the improved weighted moving average to obtain denoised pressure data.

[0052] The weighted moving average is a method of assigning different weights to observation values, obtaining moving average values according to different weights, and determining a prediction value based on the last moving average value. The weighted moving average is a statistical method mainly used for calculating average values in a period of time, in which the weights of each data point are different, and the closer the data point, the greater the weight.

[0053] The weight is generally set according to experience at present, and in the embodiment, the weight in the weighted moving average is adaptively set according to the change of the pressure data collected in the inflation process and the pressure maintaining process, so as to smooth the pressure data in the pressure maintaining process.

[0054] Specifically, the specific process of smoothing the pressure data by using the improved weighted moving average is as follows:

[0055] Firstly, a sliding window of the weighted moving average is preset. In the embodiment, the size of the sliding window is set to 8, and there are 8 pressure data in the sliding window each time the sliding window slides.

[0056] Of course, as other embodiments, the size of the sliding window can also be set according to actual conditions, such as 4, 6, etc.

[0057] The step length of the above sliding can be set to 1, 2, 3 or 6, etc., and of course it can also be set according to actual conditions.

[0058] Secondly, the weight of each pressure data in any sliding window is obtained, the pressure data in the sliding window is weighted and averaged by using the weight, the weighted average value is obtained, the weighted average value is taken as the denoised pressure of the pressure corresponding to the end of the window, and the denoised pressure data is obtained.

[0059] The above weight obtaining process includes steps S21-S24, and specifically includes:

[0060] In step S21, the pressure data of the pipe in the inflation process is obtained, and the abnormality degree of each pressure in the pressure data in the inflation process is calculated.

[0061] In the embodiment, when the tracer gas is injected into the pipeline rapidly, the change of the flow rate of the gas can cause the transient pressure fluctuation in the pipeline, which can cause the slight vibration of the pipeline, and then the vibration can be transmitted to the pressure sensor through the pipeline. The vibration can disturb the data acquisition of the pressure sensor, and thus the acquisition of the pressure data can be affected. Therefore, the abnormality degree of each pressure in the pressure data of the pipeline during the inflation process also needs to be calculated.

[0062] In the embodiment, the process from the start of the inflation of the pipeline to the first time when the pressure data of the pipeline reaches the rated pressure is the inflation process (the inflation of the pipeline is not performed after the inflation process). The pressure data of the pipeline during the inflation process is obtained, the difference between each pressure and the previous pressure is calculated, and the abnormality degree of each pressure in the pressure data of the pipeline during the inflation process is calculated based on the difference and the average level of the pressure change.

[0063] Specifically, the abnormality degree of each pressure in the pressure data during the inflation process is calculated as follows:

[0064]

[0065] wherein w i is the abnormality degree of the i th pressure in the pressure data of the pipeline during the inflation process, i > 1, p i and p i-1 are the i th and the i-1 th pressure during the inflation process, respectively, P is the rated pressure, and I is the total number of the pressures during the inflation process.

[0066] represents the average value of the pressure change in the pipeline during the inflation process, and reflects the average level of the pressure change in the pipeline during the inflation process. The greater |p i -p i-1 represents the real-time change amount of the adjacent two pressures during the inflation process. Since the change of the environment in which the pressure sensor is located can cause a certain noise in the acquired pressure data, the real-time change amount of the pressure data deviates from the average level of the pressure change, and thus, The greater the value is, the more likely the i th pressure in the pressure data of the pipeline has an abnormal change, and thus the greater the abnormality degree of the i th pressure is. The smaller the value is, the less likely the i th pressure in the pressure data of the pipeline has an abnormal change, and thus the smaller the abnormality degree of the i th pressure is.

[0067] In the embodiment, in order to facilitate the calculation, the P is normalized.

[0068] ​Step S22, respectively, obtain the acceleration data of the pipeline in the inflation process and the pressure maintaining process; and calculate the abnormal change degree of each pressure in the pressure data in the pressure maintaining process according to the mean value of the abnormal degree of the pressure data in the inflation process, the pressure data in the pressure maintaining process and the acceleration data.

[0069] In this embodiment, it is considered that the steady state of the pressure sensor will change due to the compression and flow of the gas in the pipeline inflation process, and the state of the probe gas is relatively stable in the pressure maintaining process after the inflation process, so that the state of the pressure sensor in the pressure maintaining process is relatively stable compared with the inflation process. Therefore, the change of the pressure data in the pressure maintaining process can be further analyzed by the characteristics to calculate the abnormal change degree of each pressure in the pressure data in the pressure maintaining process.

[0070] The acceleration data is composed of the acceleration of the pressure sensor at different time in the inflation process and the pressure maintaining process. The pressure maintaining process is a period of time starting from the end of the inflation.

[0071] Specifically, the calculation method of the abnormal change degree of each pressure in the pressure data in the pressure maintaining process is as follows:

[0072] The mean value of the abnormal degree of the pressure data of the pipeline in the inflation process, P is the rated pressure, and The jth and j-1th pressure in the pressure maintaining process, The mean value of all pressures before the jth pressure in the pressure maintaining process, and T is the state stability.

[0073] Because the flow and compression of the probe gas in the inflation process will cause the pipeline to vibrate, and then the collection state of the pressure sensor is unstable, at this time the abnormal change degree of the pressure data may be relatively high; and in the pressure maintaining process, the probe gas no longer flows violently, so that the collection state of the pressure sensor is relatively stable, and at this time the abnormal change degree of the pressure data is relatively low, therefore, the mean value of the abnormal change degree of the pressure data in the inflation process is taken as the reference value of the abnormal change degree of the pressure data of the pipeline in the pressure maintaining process, to evaluate whether the pressure data in the pressure maintaining process has abnormal change.

[0074] When , it is proved that the next pressure does not increase relative to the previous pressure, which indicates that there may be a leakage point in the pipeline to reduce the pressure data, or (the pipeline has no leakage) the change of the data is caused by noise. Because the aperture of the pipeline leakage is very small, the pressure drop of the pipeline with leakage point in the pressure maintaining process is small, and the noise has randomness, so the noise data may cause the pressure data to have a large drop, therefore ​The greater the abnormal change degree of the jth pressure in the pressure maintaining process is, the greater the change of the pressure data is likely to be caused by the noise data. The smaller the abnormal change degree of the jth pressure in the pressure maintaining process is, the greater the change of the pressure data is likely to be caused by the pipeline leakage.

[0075] The state stability degree is the maximum value of the ratios in three directions; the ratio is the ratio of the jth acceleration of the pressure sensor in the m direction in the pressure maintaining process to the average of the acceleration data of the pressure sensor in the m direction in the inflation process; the three directions are the x-axis direction, the y-axis direction and the z-axis direction, and the m direction is any one of the x-axis direction, the y-axis direction and the z-axis direction.

[0076] Specifically, The average of the acceleration data of the pressure sensor in the m direction in the inflation process is a j,m The jth acceleration of the pressure sensor in the m direction in the pressure maintaining process is a

[0077] T represents the state stability degree of the pressure sensor corresponding to the jth pressure in the pressure maintaining process; the greater T is, the more unstable the state of the position of the pressure sensor is, and the more likely there is noise in the collected pressure data, resulting in abnormal change of the pressure data; the smaller T is, the more stable the state of the position of the pressure sensor is, and the less likely there is noise in the collected data, and the smaller the possibility of abnormal change of the pressure data is. Therefore, T is corrected by T When T is greater, the smaller the abnormal change degree of the jth pressure in the pressure maintaining process is, the greater the change of the pressure data is likely to be caused by the pipeline leakage. the more likely it is less than 1, and the greater the upward correction degree of T to is, so that the abnormal change degree of the jth pressure in the pressure maintaining process is greater; the smaller T is, the greater it is, this time the more likely it is greater than 1, and the greater the downward correction degree of T to is, so that the abnormal change degree of the jth pressure in the pressure maintaining process is smaller.

[0078] When , it is proved that the latter pressure in the pressure maintaining process is greater than the former pressure, and since the inflation into the pipeline has been stopped at this time, the actual pressure in the pipeline cannot increase, so must be caused by abnormal change of the pressure data, and at this time, the abnormal change degree of the pressure data in the pressure maintaining process is the greatest, so when , the abnormal change degree of the pressure data is 1.

[0079] Step S23: Calculate the weight of each pressure based on the degree of abnormal change in the pressure data during the pressure holding process, and then perform a weighted moving average on multiple pressures within the sliding window using the weights to obtain the denoised pressure at the end of the window.

[0080] In one embodiment, the weights of each pressure within any sliding window are:

[0081]

[0082] Among them, w′ j-k+1 The weight of the (j-k+1)th pressure during the pressure holding process, To represent the degree of abnormal change in the pressure at the (j-k+1)th pressure during the pressure holding process, exp() is an exponential function with the natural constant e as the base, K is the size of the sliding window, and k is the index of the pressure within the sliding window.

[0083] The larger the value, the more likely the corresponding pressure is to be noise data, and the smaller the impact of that pressure should be on the subsequent weighted moving average result. The smaller the value, the better. The less likely the data is to be noisy, the greater the impact of the corresponding pressure on the weighted moving average result should be. Therefore, using... As The weight, This represents the (j-k+1)th pressure during the pressure holding process.

[0084] In another embodiment, since the occurrence of noise data is sporadic, and there is a certain gap between the degree of abnormal change in pressure data accompanied by noise and the degree of abnormal change in normal pressure data, the weight is calculated by utilizing the difference between two adjacent pressures.

[0085] Specifically, the weights are:

[0086] w′ j-k+1 The weight of the (j-k+1)th pressure during the pressure holding process, Let exp(()) be the difference between the abnormal change in the (j-k+1)th pressure and the abnormal change in the (j-k)th pressure during the pressure holding process. exp(()) is an exponential function, K is the size of the sliding window, and k is the index of the pressure within the sliding window.

[0087] when When it is larger, and There may be noisy data, and The data is relatively likely to be noisy, and should be made... The smaller the weight, when The smaller, and The noise data can not exist in the pressure data, The lower the possibility of the noise data is, the pressure data should be has a greater weight, so that w′ j-k+1 as the weight of the pressure data, and are the j-kth and j-k+1th pressure data in the pressure maintaining process, respectively.

[0088] Based on the above weight, for any pressure data at the end of the sliding window, the corresponding denoised pressure data is: is the jth denoised pressure data in the pressure maintaining process, w′ j-k+1 is the weight of the j-k+1th pressure data in the pressure maintaining process, is the j-k+1th pressure data in the pressure maintaining process, and K is the size of the sliding window.

[0089] The reason for the above weighted average of the pressure data in the sliding window by the weight is that due to the state change of the pressure sensor in the process of collecting the pressure data, there is certain noise data in the collected pressure data, so in order to weaken the influence of the noise data on the judgment of whether the pipeline leaks, the weight of the pressure data is calculated according to the abnormal change degree of the pressure data in the pressure maintaining process, so as to smooth each pressure data in the pressure data.

[0090] In step S3, if the average of the ratio of each pressure data in the denoised pressure data to the rated pressure is less than a set value, the pipeline has a leakage point.

[0091] In this embodiment, the ratio of each pressure data in the denoised pressure data to the rated pressure is obtained, and the average of all ratios is calculated. When the average is less than a set value, it is considered that the pipeline has a leakage point.

[0092] The value of the above set value is 0.998, and of course it can also be determined according to the actual situation.

[0093] Up to now, the leakage detection of all pipelines in the refrigerator refrigeration system can be realized.

[0094] The scheme of the present application calculates the abnormal degree of each pressure data in the pressure data in the inflation process through the pressure data of the pipeline in the inflation process; calculates the abnormal change degree of the pressure data in the pressure maintaining process through the abnormal degree of the pressure data in the pipeline inflation process and the pressure data and acceleration data in the pressure maintaining process. The weight of each pressure data in the pressure data in the pressure maintaining process is calculated through the abnormal change degree of the pressure data in the pressure maintaining process, and the pressure data is weighted and moving averaged through the weight; through the weighted moving average value of the pressure data, the influence of the noise data in the pressure data on the pipeline leakage detection can be reduced, and the false detection rate of the pipeline leakage can be reduced. ​

[0095] The present application also provides a leakage detection system. As shown in Figure 2 the system comprises a processor and a memory storing computer program instructions which, when executed by the processor, implement a leakage detection method according to the above description of the present application.

[0096] The system also comprises a communication bus and a communication interface and other components well known to those skilled in the art, the settings and functions of which are known in the art, and thus will not be described here.

[0097] In the present application, the aforementioned memory can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device. For example, the computer readable storage medium can be any appropriate magnetic storage medium or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc., or any other medium that can be used to store the desired information and that can be accessed by an application, module, or both. Any such computer storage media can be part of the device or accessible or connectable thereto. Any application or module described in the present application can be implemented by computer readable / executable instructions stored or otherwise held by such computer readable media.

[0098] Although the present application has been shown and described with respect to several embodiments thereof, it will be apparent that those skilled in the art will be able to make many modifications, changes and alterations to this application without departing from the spirit and scope thereof. It should be understood that various alternatives to the embodiments of the application described herein can be employed in practicing the application.

Claims

1. A leak detection method, characterized in that, include: Pressure data of the pipeline during the pressure holding process is acquired using a pressure sensor, and the pressure data includes multiple pressures. The pressure data is smoothed using an improved weighted moving average to obtain denoised pressure data. If the average ratio of each pressure to the rated pressure in the denoised pressure data is less than the set value, then there is a leak in the pipeline. In the improved weighted moving average, the weight of each pressure is inversely correlated with the degree of abnormal change in the corresponding pressure; the degree of abnormal change for: ; This represents the average degree of anomalies in the pipeline pressure data during the inflation process. For rated pressure, and These are the j-th and (j-1)-th pressures during the pressure holding process, respectively. Let T be the average of all pressures before the j-th pressure during the pressure holding process, and let T be the state stability. State stability is the maximum of the ratios in the three directions; The ratio is the ratio of the j-th acceleration of the pressure sensor in the m-direction during the pressure holding process to the average value of the acceleration data of the pressure sensor in the m-direction during the inflation process; where the three directions are the x-axis direction, y-axis direction, and z-axis direction, and the m-direction is any one of the x-axis direction, y-axis direction, and z-axis direction.

2. The leakage detection method according to claim 1, characterized in that, The weights are: ； in, The weight of the (j-k+1)th pressure during the pressure holding process, To represent the degree of abnormal pressure change at the (j-k+1)th pressure during the pressure holding process, exp() is an exponential function, K is the size of the sliding window, and k is the index of the pressure within the sliding window.

3. The leakage detection method according to claim 1, characterized in that, The weights are: ; The weight of the (j-k+1)th pressure during the pressure holding process, Let exp() be the difference between the abnormal change in the (j-k+1)th pressure and the abnormal change in the jkth pressure during the pressure holding process. exp() is an exponential function, K is the size of the sliding window, and k is the index of the pressure within the sliding window.

4. The leakage detection method according to claim 1, characterized in that, The degree of abnormality is: ; in, The degree of anomaly in the i-th pressure data of the pipeline during inflation, where i>

1. and These are the i-th and (i-1)-th pressures during inflation, respectively. I represents the rated pressure, and I represents the total pressure during the inflation process.

5. A leakage detection method according to claim 2 or 3, characterized in that, The specific process of smoothing the pressure data using an improved weighted moving average to obtain denoised pressure data is as follows: A preset sliding window for weighted moving averages; A weighted average is applied to the sliding window containing the target pressure to obtain the denoised pressure corresponding to the target pressure, thereby obtaining the denoised pressure data during the pressure holding process; the target pressure is any pressure in the pressure data during the pressure holding process, wherein the target pressure is located at the end of the corresponding sliding window; The pressure after noise reduction for: ; The weight of the (j-k+1)th pressure during the pressure holding process, K represents the (j-k+1)th pressure during the pressure holding process, and K is the size of the sliding window.

6. The leakage detection method according to claim 1, characterized in that, The pressurization process is the time interval between the moment when the leak detection gas is supplied and the moment when the pressure is raised to the rated pressure and the supply of leak detection gas is stopped; the pressure holding process is a set time interval starting from the moment the supply of leak detection gas is stopped.

7. The leakage detection method according to claim 1, characterized in that, The acceleration data characterizes the vibration of the pipeline during the inflation process.

8. A leak detection system, characterized in that, include: processor; A memory storing computer instructions for leak detection, which, when executed by the processor, cause the system to perform a leak detection method according to any one of claims 1-7.

Images