Soil gas sensor arrangement and leakage source identification algorithm based on adjoint probability
Through soil gas sensor arrangement and leakage source identification algorithm based on accompanying probability, the problem of inefficient leakage detection efficiency in urban gas pipelines is solved, and rapid and accurate leakage source position determination and source strength calculation are achieved, which improves detection efficiency and accuracy.
Patent Information
- Application Number
- CN202411059890.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-02
- Publication Date
- 2024-11-29
AI Technical Summary
The existing urban gas pipeline leakage detection methods are inefficient, especially in complex underground space environments, which are difficult to accurately and quickly determine the leakage source location. The odor detection method is disturbed by the environment and has a slow response time, and the ultrasonic detection method is cumbersome to operate.
The soil gas sensor arrangement and leakage source recognition algorithm based on accompanying probability are adopted. By establishing a leakage calculation model and diffusion model, the leakage source position and source strength are calculated by using the multi-sensor combined accompanying probability equation and Tikhonov regularization algorithm.
It improves the accuracy and efficiency of determining the location of the leakage source, narrows the detection range, reduces financial and material investment, and realizes real-time monitoring and accurate identification.
Smart Images

Figure CN119047104A_ABST
Abstract
Description
(I) Technical field
[0001] The present invention belongs to the application of soil gas leakage source identification technology. The specific implementation is mainly to determine the concentration through sensor measurement, obtain the accompanying probability, determine the location of the leakage source in the soil, and finally calculate the source strength. This method can quickly and accurately determine the location of the leakage source and has certain potential in monitoring leakage sources in the future. (II) Background technology
[0002] As an important infrastructure of modern cities, urban gas pipelines are responsible for supplying natural gas to residents, industrial and commercial users. With the acceleration of my country's urbanization process, the coverage of urban gas pipeline networks is becoming wider and wider, playing a key role in ensuring the energy supply and environmental quality of cities.
[0003] The main function of urban gas pipelines is to provide safe, stable and efficient natural gas supply to cities. This special pipeline system needs to have extremely high safety, reliability and durability to cope with various complex environmental and climatic conditions. In addition, gas pipelines must also meet environmental protection standards to reduce the impact on the environment.
[0004] In order to ensure the safe and efficient operation of urban gas pipelines, a series of measures need to be taken. First, during the design and construction phase, it is necessary to follow relevant national and local standards and specifications to ensure the quality and safety of the pipeline system. Secondly, during the operation and maintenance phase, regular inspection and maintenance are required to promptly detect and repair potential problems to ensure the normal operation of the pipeline system. In addition, it is necessary to strengthen safety management and improve emergency response capabilities to deal with possible emergencies.
[0005] At present, my country mainly monitors gas in engineering through manual detection, using some specific instruments to detect leaks, and then further searches along this detection point until the location is found. Due to the complexity of underground space stations and the crisscrossing of pipelines, it is a relatively complicated and time-consuming task to find the exact leak point.
[0006] Regarding the detection of gas leaks, the more in-depth detection methods are mainly odor detection and ultrasonic detection. For the odor detection method, mainly because natural gas itself is colorless and odorless, in order to facilitate the detection of leaks, a special odorant is usually added to the gas to make it have a distinct rotten egg smell. When gas leaks, people will sense the odor through their sense of smell and take timely measures. As for the ultrasonic detection method, it is mainly a method of detecting gas leaks by using the propagation characteristics of ultrasound. By placing an ultrasonic transmitter near the gas pipeline, when gas leaks, the leaking gas will generate high-frequency noise. The ultrasonic receiver can receive these noise signals, analyze and judge them, and determine the location and extent of the leak.
[0007] In summary, there are many problems with the existing methods for determining the source of leakage. The odor detection method is limited by its environment and response time. The complex environment will interfere with the detection effect, making the result inaccurate; and some gases make the sensor response time slow and cannot be monitored in real time. Ultrasonic monitoring is mainly due to the complexity of materials and cumbersome operation. In order to overcome the above difficulties, this paper proposes a soil gas sensor layout and leakage source identification algorithm based on adjoint probability, which mainly utilizes the characteristics of gas diffusion, simplifies calculations, improves efficiency, and reduces the scope, thereby reducing a lot of financial and material resources. (III) Summary of the invention
[0008] Technical issues solved
[0009] As cities continue to expand, locating the source of leakage in the soil within urban blocks is very important to protect the health and safety of residents. Existing methods for detecting leakage sources are inefficient. In view of the shortcomings of existing methods, this paper provides a soil gas sensor deployment and leakage source identification algorithm based on adjoint probability.
[0010] Technical Solution
[0011] 1. A soil gas sensor arrangement and leakage source identification algorithm based on adjoint probability, characterized in that it includes the following steps:
[0012] Step (1): Soil is a porous medium. First, determine the type of soil in the area based on the data provided by the local land bureau, and then derive the effective relative diffusion coefficient of the leaked gas in the soil as D m ;
[0013] Step (2): Establish a leakage calculation model and a diffusion model of leaked gas in the soil. With the potential leakage point as the center and the area above the leakage point as the control volume, establish a spatial rectangular coordinate system. The diffusion model in the soil is:
[0014]
[0015] Where q is the leakage mass flow rate of the leaking gas, kg / s; r is the distance from the leaking point, m; D m is the effective relative diffusion coefficient of leaked gas in soil, m 2 / s; C is the forward diffusion concentration, ppm;
[0016] Step (3): Arrange the sensors. First, arrange the first sensor at a certain distance above the pipeline and observe the diffusion concentration C displayed by the sensor. When the diffusion concentration is detected, select the position of the second sensor. Take the first sensor as the center and the distance from the pipeline section by 3 times to obtain the front and back positions. Select the position with the larger detected concentration as the setting position of the second sensor.
[0017] Step (4): After installing the sensors, the monitoring data of the two sensors are extracted, and the historical concentration information of the pollutants detected by the two sensors and the location coordinates thereof are recorded as S1 and S2;
[0018] Step (5): The information of the two pollutant sensors in step (4) is integrated through the multi-sensor joint adjoint probability equation to calculate the adjoint probability of the area to be measured. The multi-sensor joint adjoint probability equation used is: in is the accompanying concentration of the sensor, ppm; r1, r2 are the distances from the sensor, m; D m is the effective relative diffusion coefficient of gas in soil, m 2 / s, c1, c2 are sensor concentrations, ppm;
[0019] Step (6): Calculate the size of the accompanying probability according to step (5), find the maximum accompanying probability, and determine the location of the leakage source.
[0020] Step (7): The information of the maximum accompanying probability position can be used through equation (1-1) to obtain the response matrix A corresponding to the monitoring point. The source strength q of the potential pollution source can be obtained by using the monitor information S1 and the response matrix A through a regularization algorithm.
[0021] 2. The regularization algorithm in step (7) adopts the Tikhonov regularization algorithm, and the specific equation is as follows:
[0022] min q (||Aq-b|| 2 +λ||q|| 2 ) where A is the response matrix which can be calculated by equation (1-1), q is the source strength to be estimated, b is the sensor data which can be measured by the sensor, λ is the regularization parameter which is obtained by the L-Curve method, and ||.|| represents the Euclidean norm (that is, the 2 norm of the vector). In order to conveniently calculate the source strength q, the equation can be differentiated with respect to q to determine the calculated source strength. The derivative equation is q=(A T A+λI) -1 (A T b), where A is the response matrix, I is the identity matrix, λ is the regularization parameter, and b is the sensor data. (IV) Description of the drawings
[0023] Figure 1 Flowchart of a soil gas sensor placement and leakage source identification algorithm based on adjoint probability
[0024] Figure 2 Implementation Case Site Layout
[0025] Figure 3 Distribution diagram of leakage source accompanying probability along the pipeline in the implementation case
[0026] Figure 4 3D schematic diagram of gas leakage source in soil (V) Specific implementation methods
[0027] The specific embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention. After reading the content taught by the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms also fall within the scope defined by the claims attached to this application.
[0028] (1) According to the data provided by the local land bureau, find out the type of local soil, and derive the relative diffusion coefficients of the soil and the effective relative diffusion coefficient of gas in the soil as D. m =5.2×10 -6 m 2 / s;
[0029] (2) Test site Figure 2 , the leak source coordinates are (0, 0), and the source strength is constant at 1kg / s. Two sensors are selected, named S1 and S2 respectively; because the diffusion of gas in the soil is isotropic, the installation of sensors needs to consider both the lower limit of the sensor's monitoring and the lower limit of the gas concentration in different areas close to the leak point. The coordinates of sensor 1 are set to (-3, 3), and the distance from the second sensor is 3 times the length of the pipe section, that is, 9m, to obtain the front and back coordinates. The position with the larger detected concentration is selected as the second sensor setting position, and the coordinates are (6, 3). The measured gas reached stability after 23s of leakage, and the concentrations detected by the two sensors were 160ppm and 120ppm respectively;
[0030] (3) Substitute the set area coordinates, the concentration recorded by the sensor, and the time of gas leakage into equation (1-1) to obtain the adjoint equations of the two sensors respectively;
[0031] (4) Integrate the adjoint equation in (3) to obtain the adjoint probability equation, and then substitute the experimental data information detected by sensors S1 and S2 into the multi-sensor joint adjoint probability equation After calculating the regional leakage source probability distribution, the concentration forms two protrusions as shown in Figure 3 ;
[0032] (5) Applying the accompanying concentration map in (4) to the three-dimensional image, the probability location of the leakage source can be easily observed, such as Figure 4 As shown, according to Figure 4 The maximum associated probability position is (-3, 0). This position is about 3m away from the leakage source (0, 0) on the pipeline, so the location of the leakage source can be determined. The response matrix A corresponding to the position (-3, 0) can be obtained. Then, through the Tikhonov regularization equation min q (||Aq-b|| 2 +λ||q|| 2 ) and his derivative equation q=(A T A+λI) -1 (A T b) The source strength is calculated to be 0.5 kg / s, where the Tikhonov regularization parameter λ is 0.1. Therefore, it can be determined that the source strength at the maximum probability position is 0.5 kg / s.
Claims
1. A soil gas sensor placement and leakage source identification algorithm based on adjoint probability, It is characterized in that The following steps are involved: Step (1): Soil is a porous medium. First, the type of soil in the area is determined based on the data provided by the local land bureau. Then, the effective relative diffusion coefficient of the leaked gas in the soil is obtained as D m ; Step (2): Establish a leakage calculation model and a diffusion model of leaked gas in the soil. With the potential leakage point as the center and the area above the leakage point as the control volume, establish a spatial rectangular coordinate system. The diffusion model in the soil is: Where q is the leakage mass flow rate of the leaking gas, kg / s; r is the distance from the leaking point, m; D m is the effective relative diffusion coefficient of leaked gas in soil, m 2 / s; C is the forward diffusion concentration, ppm; Step (3): Arrange the sensors. First, arrange the first sensor at a certain distance above the pipeline and observe the diffusion concentration C displayed by the sensor. When the diffusion concentration is detected, select the position of the second sensor. Take the first sensor as the center and the distance from the pipeline section by 3 times to obtain the front and back positions. Select the position with the larger detected concentration as the setting position of the second sensor. Step (4): After installing the sensors, extract the monitoring data of the two sensors and record the historical concentration information and location coordinates of the pollutants detected by the two sensors as S 1 and S 2 ; Step (5): The information of the two pollutant sensors in step (4) is integrated through the multi-sensor joint adjoint probability equation to calculate the adjoint probability of the area to be measured. The multi-sensor joint adjoint probability equation used is: in is the accompanying concentration of the sensor, ppm; r 1 , r 2 is the distance from the sensor, m; D m is the effective relative diffusion coefficient of gas in soil, m 2 / s,c 1 , c 2 is the sensor concentration, ppm; Step (6): Calculate the size of the accompanying probability according to step (5), find the maximum accompanying probability, and determine the location of the leakage source. Step (7): The information of the maximum accompanying probability position can be used through equation (1-1) to obtain the response matrix A corresponding to the monitoring point. The source strength q of the potential pollution source can be obtained by using the monitor information S1 and the response matrix A through a regularization algorithm.
2. The regularization algorithm in step (7) adopts the Tikhonov regularization algorithm, and the specific equation is as follows: min q (||Aq-b|| 2 +λ||q|| 2 ), where A is the response matrix that can be calculated by equation (1-1), q is the source strength to be estimated, b is the sensor data that can be measured by the sensor, λ is the regularization parameter, which is obtained here by the L-Curve method, and ||·|| represents the Euclidean norm (that is, the 2 norm of the vector). In order to conveniently calculate the source strength q, the equation is differentiated with respect to q to determine the obtained source strength. The derivative equation is q=(A T A+λI) -1 (A T b), where A is the response matrix, I is the identity matrix, λ is the regularization parameter, and b is the sensor data.
Citation Information
Cited By
Gas leakage source identification system and method based on multiple sensors and storage medium
CN119377534A
Multi-Sensor Based Gas Leak Source Identification System, Method and Storage Medium
CN119377534B
Leakage detecting and positioning method for water supply pipeline
CN122286188A
