Dynamic evaluation method for H2S transportation pipeline leakage range

Abstract

The application discloses a dynamic evaluation method for H2S transportation pipeline leakage range, and specifically comprises the following steps: step 1, collecting real-time data of H2S sensors on the H2S transportation pipeline and uploading to an industrial computer; step 2, judging whether the H2S transportation pipeline leaks or not and determining the leakage source; step 3, judging the leakage direction of the leakage source; step 4, establishing a dynamic evaluation model for H2S leakage diffusion to obtain the longest horizontal distance of H2S leakage; step 5, determining the actual leakage influence range of the leakage source; and step 6, emergency warning and personnel safety evacuation. The dynamic evaluation method for H2S transportation pipeline leakage range can timely detect whether H2S leaks during pipeline transportation and the position of the leakage source, and accurately evaluate the leakage range of the leakage source in real time, so that emergency warning can be carried out, personnel safety evacuation is organized and loss is reduced.

Description

Technical Field

[0001] This invention belongs to the technical field of leak gas safety assessment methods, specifically relating to a dynamic assessment method for the leakage range of H2S transportation pipelines. Background Technology

[0002] Pipeline transportation is the main route for transporting H2S gas. However, due to wear, pipeline pressure, corrosion, and other factors, pipelines are prone to leakage, posing a significant challenge to the safe transport of H2S gas. How to monitor sudden changes in H2S gas concentration along industrial transport pipelines in real time, dynamically assess the impact range of gas leaks, provide emergency warnings for hazardous areas, and assist emergency management departments in providing safe evacuation routes to avoid casualties are urgent problems that need to be solved.

[0003] Currently, regarding methods for calculating H2S leakage distance, Chinese patent "A Method for Estimating the Leakage Distance of Hydrogen Sulfide-Containing Equipment" (Publication No.: CN114441513A, Application Date: 20201103, Publication Date: 20220506) discloses a method for estimating the leakage distance of hydrogen sulfide-containing equipment. This method measures the H2S concentration in the equipment under test, calculates the potential H2S leakage, and establishes a relationship between the H2S concentration, potential leakage, and leakage distance, thus deriving the dangerous leakage distance for H2S concentrations of 20ppm and 100ppm. While this method can quickly estimate the safe leakage distance using H2S concentrations inside and outside the pipeline, it does not consider the influence of various conditions such as pipeline pressure and ambient wind speed, making it difficult to accurately assess the dangerous leakage distance of H2S pipelines. Therefore, a method is needed to dynamically and accurately assess and calculate the H2S leakage range in real time using H2S pipeline and environmental monitoring data, thereby ensuring the transportation safety of H2S-containing pipelines. Summary of the Invention

[0004] The purpose of this invention is to provide a dynamic evaluation method for the leakage range of H2S transportation pipelines. This method can detect the leakage source of H2S transportation pipelines in a timely manner and assess the leakage impact range of the leakage source in real time and accurately.

[0005] The technical solution adopted in this invention is a dynamic evaluation method for the leakage range of H2S transportation pipelines, which specifically includes the following steps:

[0006] Step 1: Collect real-time data from the H2S sensors on the H2S transport pipeline and upload it to the industrial control computer;

[0007] Step 2: Determine if there is a leak in the H2S transport pipeline and identify the source of the leak;

[0008] Step 3: Determine the direction of leakage from the leak source;

[0009] Step 4: Establish a dynamic evaluation model for H2S leakage and diffusion to obtain the longest horizontal distance of H2S leakage;

[0010] Step 5: Determine the actual leakage impact range of the leak source;

[0011] Step 6: Emergency warning and safe evacuation of personnel.

[0012] The invention is further characterized in that,

[0013] In step 1, there are n H2S sensors, which are evenly distributed on the H2S transport pipeline to monitor the H2S concentration at various points outside the pipeline in real time. The real-time data from the H2S sensors is uploaded to an industrial control computer equipped with a dynamic evaluation model of the H2S transport pipeline leakage range via RS485 communication.

[0014] Step 2 specifically includes the following steps:

[0015] Step 2.1, at the first read rate v 1. Read the values ​​from the H2S sensors at each location and record them as follows: S 1. S 2, ... S n-1 , S n The maximum concentration of H2S was obtained by performing maximum value processing. S max ; at the second read rate v 2. Take two consecutive readings of the H2S sensor values ​​at each location and calculate the concentration increment corresponding to the two consecutive readings at that location, denoted as Δ. S 1. Δ S 2、……、Δ S n-1 Δ S n And perform maximum value processing to obtain the maximum H2S concentration increment Δ S max ;

[0016] Step 2.2: Set the maximum safety threshold α and the maximum mutation safety threshold β for H2S. If S max ≥ɑ or Δ S max If ≥β, then a leak has occurred in the H2S transport pipeline, and the leak source is determined by the H2S sensor location where the leak occurred; proceed to step 3. S max <ɑ and Δ S max If <β, then no leak has occurred in the H2S transport pipeline; return to step 2.1.

[0017] In step 2.2, the maximum safety threshold α for H2S is 20 ppm, and the maximum mutation safety threshold β for H2S is 1.

[0018] The specific process for determining the leakage direction in step 3 is as follows: read the wind direction at the location of the leakage source in the meteorological monitoring station, and the leakage direction of the leakage source is opposite to the wind direction.

[0019] Step 4 specifically includes the following steps:

[0020] Step 4.1: In accordance with the provisions of the "Design Standard for Detection and Alarm of Combustible and Toxic Gases in Petrochemical Industry" and the "Guideline for Occupational Hazard Prevention of Hydrogen Sulfide", the leakage threshold of H2S is set to 20ppm, 100ppm and 500ppm.

[0021] Step 4.2: Use FLUENT simulation software to simulate leakage conditions corresponding to different leakage concentrations, pipeline pressures, and ambient wind speeds. The parameters are: 1% < leakage concentration < 20%, 0.5 MPa < pipeline pressure < 2.5 MPa, and 0.2 m / s < ambient wind speed < 2 m / s. This yields numerical simulation data for H2S gas leakage. Historical data on H2S leakage concentration, pipeline pressure, ambient wind speed, and H2S sensor locations at the moment of the H2S pipeline leak are extracted as on-site simulation data for H2S gas leakage.

[0022] Step 4.3: Use FLUENT simulation software to simulate the horizontal leakage distance corresponding to different leakage thresholds in Step 4.1. d Simulations were conducted using data from the field simulation, including leakage concentration, pipeline pressure, ambient wind speed, leakage threshold, and the corresponding horizontal distance from the leakage threshold. d Using the horizontal axis as the data source, the numerical simulation data and on-site simulation data of H2S gas leakage obtained in step 4.2 are randomly divided into five groups of data. These five groups of data are then used as the vertical axis to divide the H2S data into 5... 5. Feature data matrix;

[0023] Step 4.4: Input the feature data matrix obtained in Step 4.3 into the GAN network for feature fusion, and train to generate H2S dynamic evaluation model data with two data features: H2S gas leakage numerical simulation data and H2S gas leakage field simulation data.

[0024] Step 4.5: Use the H2S dynamic evaluation model data from Step 4.4 to train a two-layer GRNN network;

[0025] Step 4.6: Complete the training process of step 4.5 to obtain the H2S leakage diffusion dynamic evaluation model, thereby calculating the longest horizontal distance of H2S leakage in real time.

[0026] The specific training process of the two-layer GRNN network described in step 4.5 is as follows:

[0027] Step 4.5.1: Using the leakage concentration, pipeline pressure, ambient wind speed, and leakage threshold from the H2S dynamic evaluation model data obtained in Step 4.4 as input values, and the leakage horizontal distance corresponding to the leakage threshold... d As the output value, it is input into the first layer of the GRNN network to obtain the leakage horizontal distance coupling value of a single-layer GRNN network, and the leakage horizontal distance is calculated. d The absolute difference Δ between the leakage horizontal distance coupling value and that of a single-layer GRNN network d ;

[0028] Step 4.5.2: Using the leakage concentration, pipeline pressure, ambient wind speed, leakage threshold, and absolute difference Δ obtained in step 4.5.1 from the H2S dynamic evaluation model data obtained in step 4.4. d As input, the distance to the leakage level corresponding to the leakage threshold. d As the output value, it is input into the second-layer GRNN network to obtain the leakage horizontal distance coupling value of the two-layer GRNN network. D .

[0029] Step 4, the training process of the two-layer GRNN network, uses the particle swarm optimization algorithm to optimize the model training process and obtain the optimal smoothing factor of the model.

[0030] Step 5 specifically includes the following steps:

[0031] Step 5.1, at the first read rate v 1. Real-time reading of ambient wind speed, ambient temperature, pipeline pressure, and values ​​of H2S sensors at various points on the H2S transport pipeline;

[0032] Step 5.2: Input the ambient wind speed, ambient temperature, pipeline pressure, and H2S sensor values ​​obtained in Step 5.1, along with the values ​​from each point on the transport pipeline, into the H2S pipeline leakage distance dynamic evaluation model obtained in Step 4, to obtain the leakage horizontal distance coupling value corresponding to different leakage thresholds. D ;

[0033] Step 5.3: Based on the multimodal monitoring values ​​read in Step 5.1 and the modified Gaussian plume model of the longest horizontal distance of H2S leakage calculated in Step 4, further calculate the actual influence range of the leak source on the windward side.

[0034] Step 5.4: Using the leak source as the center and one-third of the actual influence range on the windward side of the leak source obtained in Step 5.3 as the radius, obtain the dangerous range of the H2S gas leak center.

[0035] Step 5.5: Superimpose the actual impact range on the downwind side obtained in Step 5.3 with the actual impact range on the upwind side to obtain the actual leakage impact range of the leakage source.

[0036] First read rate v 1 represents 1 read / second; second read rate v 2 means 2 times / second.

[0037] Step 5.3 specifically includes the following steps:

[0038] Step 5.3.1: Since the hydrogen sulfide on the downwind side will be directly blown away by the wind after the hydrogen sulfide leak, no leakage range will form on the downwind side of the hydrogen sulfide leak source, that is, the actual influence range on the downwind side is 0; based on the multimodal monitoring values ​​read in Step 5.1, the influence range on the upwind side of the leak source is calculated using Equations (5.1) and (5.2):

[0039]

[0040] In the formula:

[0041] Q Leakage rate, mg / s;

[0042] μ The average wind speed in the environment at the time of the leak, in m / s;

[0043] H The effective height of the leak source, in meters (m).

[0044] σ y , σ z The diffusion systems for crosswind and vertical wind are respectively. Based on atmospheric and topographical conditions, atmospheric stability can be divided into six levels: A, B, C, D, E, and F. The corresponding crosswind and vertical wind diffusion systems are shown in Table 1.

[0045] Table 1. Diffusion coefficients for crosswind and vertical wind directions at different atmospheric stability levels.

[0046]

[0047] x , y , z These represent the distances to the downwind direction, crosswind direction, and vertical wind direction, respectively, in meters (m).

[0048] C For a point in space ( x , y , z The concentration of toxic substances in the air, mg / m³ 3 ;

[0049] r Let the leakage orifice diameter be in meters (m).

[0050] λ The adiabatic coefficient of the gas;

[0051] P The average pipeline pressure is expressed in Pa.

[0052] M The value represents the molar mass of the leaked gas, expressed in kg / kmol.

[0053] R is the leakage gas constant;

[0054] T Temperature of the leaked gas, in °C;

[0055] Step 5.3.2: For the two-dimensional plane of the leak, integrate the maximum H2S concentration value obtained in Step 5.1. S max ,and x , y , z ≠0, the calculation formula for the Gaussian plume model is as follows:

[0056]

[0057] Step 5.3.3: Correct the Gaussian plume model obtained in step 5.3.2 using the longest horizontal distance of H2S leakage calculated in step 4. Specifically, select the corresponding horizontal distance coupling values ​​for leakage concentrations of 20ppm and 100ppm. D 1. D 2. An improved Gaussian plume model was obtained through two-layer GRNN optimization, and the actual influence range on the upwind side was calculated. The calculation formula is as follows:

[0058]

[0059] The beneficial effects of this invention are:

[0060] (1) The dynamic evaluation method for H2S transport pipeline leakage range of the present invention has multiple H2S sensors evenly arranged on the H2S transport pipeline for real-time monitoring of H2S concentration at various points outside the pipeline. Different H2S sensors correspond to different fixed numbers, so that the location of the leakage source can be determined in time when H2S leakage occurs.

[0061] (2) The present invention mainly establishes a dynamic evaluation model for H2S leakage diffusion by combining GAN and two-layer GRNN and further obtains the leakage influence range of the leakage source by combining Gaussian plume model. At the same time, when establishing the dynamic evaluation model for H2S leakage diffusion, the influence of pipeline pressure and environmental wind speed are comprehensively considered, thereby improving the real-time performance and accuracy of leakage influence range assessment. Attached Figure Description

[0062] Figure 1 This is a flowchart of the dynamic evaluation method for the leakage range of H2S transportation pipelines according to the present invention;

[0063] Figure 2 This is a flowchart for determining whether an H2S transport pipeline has leaked, part of the dynamic evaluation method for the leakage range of H2S transport pipelines in this invention.

[0064] Figure 3 This is a flowchart of the dynamic evaluation model for H2S leakage diffusion in the dynamic evaluation method for H2S transportation pipeline leakage range of the present invention;

[0065] Figure 4 This is a flowchart illustrating the dynamic evaluation method for determining the actual leakage impact range of a leakage source in the H2S transportation pipeline leakage range of this invention. Detailed Implementation

[0066] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0067] The flowchart of the dynamic evaluation method for the leakage range of H2S transportation pipelines in this invention is as follows: Figure 1 As shown, the main process involves H2S sensors collecting H2S concentration data from the surface of the H2S transport pipeline. The signals are then uploaded to an industrial control computer equipped with a dynamic evaluation model of the H2S transport pipeline leakage range for real-time calculation. This provides timely warnings about whether a gas leak has occurred in the H2S transport pipeline and the extent of the leak. The specific steps include:

[0068] Step 1: Collect real-time data from n H2S sensors evenly distributed on the H2S transport pipeline and upload it to the industrial control computer via the industrial ring network;

[0069] The H2S sensor is used to monitor the H2S concentration at various points outside the pipeline in real time. The number of H2S sensors, n, can be determined according to actual needs, and each sensor corresponds to 1 to n different fixed numbers to determine the location of the corresponding sensor. This invention mainly uses RS485 communication to output signals to external devices. The sensor monitoring signal is converted from RS485 to optical signal through an RS optical transceiver and transmitted to the industrial ring network. The signal is converted into a 485 signal through the RS optical transceiver receiver in the industrial ring network and received by the industrial control computer equipped with a dynamic evaluation model of the H2S transportation pipeline leakage range. The computer then performs real-time calculations on the collected data.

[0070] Step 2: Determine if there is a leak in the H2S transport pipeline and identify the source of the leak. The process for determining whether there is a leak in the H2S transport pipeline is as follows: Figure 2 As shown, the specific steps include:

[0071] Step 2.1, at the first read rate v 1. Read the values ​​from the H2S sensors at each location and record them as follows: S 1. S 2, ... S n-1 , S n The maximum concentration of H2S was obtained by performing maximum value processing. S max ; at the second read rate v 2. Take two consecutive readings of the H2S sensor values ​​at each location and calculate the concentration increment corresponding to the two consecutive readings at that location, denoted as Δ. S 1. Δ S 2、……、Δ S n-1 Δ S n And perform maximum value processing to obtain the maximum H2S concentration increment Δ S max ;

[0072] Step 2.2: Set the maximum safety threshold α and the maximum mutation safety threshold β for H2S. If S max ≥ɑ or Δ S max If ≥β, then a leak has occurred in the H2S transport pipeline, and the leak source is determined by the location of the H2S sensor at which the leak occurred; proceed to step 3. S max <ɑ and Δ S max If <β, then no leak has occurred in the H2S transport pipeline, and we return to step 2.1.

[0073] In accordance with the "Design Standard for Detection and Alarm of Combustible and Toxic Gases in Petrochemical Industry" and the "Guidelines for Occupational Hazard Prevention of Hydrogen Sulfide" (H2S-related specifications and standards), the H2S hazard concentration classification index is set to three types: acceptable exposure safety limit of 20 ppm, critical hazardous concentration of 100 ppm, and life-threatening concentration of 500 ppm. Therefore, the maximum safety threshold α of this invention is set to 20 ppm, and the maximum sudden change safety threshold β is set to 1; the first reading rate... v 1 represents 1 read / second; second read rate v 2 means 2 times / second.

[0074] Step 3: Determine the leakage direction of the leak source. Specifically, read the wind direction at the location of the leak source in the meteorological monitoring station. The leakage direction of the leak source is opposite to the wind direction.

[0075] Step 4: Establish a dynamic evaluation model for H2S leakage and diffusion by fusing GAN (Generative Adversarial Network) and two-layer GRNN (Generalized Regressive Neural Network) to obtain the longest horizontal distance of H2S leakage. The process for establishing the dynamic evaluation model for H2S leakage and diffusion is as follows: Figure 3 As shown, the specific steps include:

[0076] Step 4.1: Based on the "Design Standard for Detection and Alarm of Combustible and Toxic Gases in Petrochemical Industry", "Guidelines for Occupational Hazard Prevention of Hydrogen Sulfide", and other relevant H2S specifications and standards, the leakage thresholds of H2S are set to 20ppm, 100ppm, and 500ppm, which correspond to the acceptable safety limit for personnel exposure, the critical concentration of danger, and the concentration that threatens life and health, respectively.

[0077] Step 4.2: Use FLUENT simulation software to simulate the leakage situation corresponding to different leakage concentrations, pipeline pressures, and ambient wind speeds. Among them, 1% < leakage concentration < 20%, 0.5 MPa < pipeline pressure < 2.5 MPa, and 0.2 m / s < ambient wind speed < 2 m / s to obtain numerical simulation data of H2S gas leakage. Extract the leakage concentration, pipeline pressure, ambient wind speed, and historical data of H2S sensors at each location at the moment of H2S pipeline leakage as on-site simulation data of H2S gas leakage.

[0078] Step 4.3: Use FLUENT simulation software to simulate the horizontal distance of leakage corresponding to different leakage thresholds in Step 4.1. d Simulations were conducted using data from the field simulation, including leakage concentration, pipeline pressure, ambient wind speed, leakage threshold, and the corresponding horizontal distance from the leakage threshold. d Using the horizontal axis as the data source, the numerical simulation data and on-site simulation data of H2S gas leakage obtained in step 4.2 are randomly divided into five groups of data. These five groups of data are then used as the vertical axis to divide the H2S data into 5... 5. Feature data matrix;

[0079] Step 4.4: Input the feature data matrix obtained in Step 4.3 into the GAN network for feature fusion, and train to generate H2S dynamic evaluation model data with two data features: H2S gas leakage numerical simulation data and H2S gas leakage field simulation data.

[0080] Step 4.5: Using the H2S dynamic evaluation model data from Step 4.4, train a two-layer GRNN network. Optimize the model training process using the Particle Swarm Optimization (PSO) algorithm, determine the optimal smoothing factor, and set initial parameters such as smoothing factor, number of iterations, inertia weights, learning rate, and iteration step size. In each iteration of model training, calculate the fitness value of each particle, assigning better fitness values ​​to individual and swarm extremes, and updating the optimal parameter solution. Continue until the required number of iterations is completed or the model's set accuracy is reached, at which point model training stops, and the PSO algorithm completes the model update. Specifically, this includes the following steps:

[0081] Step 4.5.1: Using the leakage concentration, pipeline pressure, ambient wind speed, and leakage threshold from the H2S dynamic evaluation model data obtained in Step 4.4 as input values, and the leakage horizontal distance corresponding to the leakage threshold... d As the output value, it is input into the first layer of the GRNN network to obtain the leakage horizontal distance coupling value of a single-layer GRNN network, and the leakage horizontal distance is calculated. d The absolute difference Δ between the leakage horizontal distance coupling value and that of a single-layer GRNN network d ;

[0082] Step 4.5.2: Using the leakage concentration, pipeline pressure, ambient wind speed, leakage threshold, and absolute difference Δ obtained in step 4.5.1 from the H2S dynamic evaluation model data obtained in step 4.4. d As input, the distance to the leakage level corresponding to the leakage threshold. d As the output value, it is input into the second-layer GRNN network to obtain the leakage horizontal distance coupling value of the two-layer GRNN network. D ;

[0083] Step 4.6: Complete the training process of step 4.5 to obtain the H2S leakage diffusion dynamic evaluation model, thereby calculating the longest horizontal distance of H2S leakage in real time.

[0084] Step 5: Determine the actual leakage impact range of the leak source. This mainly involves calculating the two-dimensional impact range of the H2S leak using the Gaussian plume model and the longest horizontal distance of the H2S leak calculated in Step 4. The Gaussian plume model is then improved, and the original model is corrected using the H2S leakage diffusion dynamic evaluation model. The process for determining the actual leakage impact range of the leak source is as follows: Figure 4 As shown, the specific steps include:

[0085] Step 5.1, at the first read rate v 1. Real-time reading of ambient wind speed, ambient temperature, pipeline pressure, and H2S sensor values ​​at various points along the H2S transport pipeline, wherein the first reading rate... v 1 means 1 time / second;

[0086] Step 5.2: Input the ambient wind speed, ambient temperature, pipeline pressure, and H2S sensor values ​​obtained in Step 5.1, along with the values ​​from each point on the transport pipeline, into the H2S leakage diffusion dynamic evaluation model obtained in Step 4, to obtain the leakage horizontal distance coupling value corresponding to different leakage thresholds. D ;

[0087] Step 5.3: Based on the multimodal monitoring values ​​read in Step 5.1 and the corrected Gaussian plume model of the longest horizontal distance of H2S leakage calculated in Step 4, further calculate the actual influence range of the leak source on the windward side, specifically including the following steps:

[0088] Step 5.3.1: Since the hydrogen sulfide on the downwind side will be directly blown away by the wind after the hydrogen sulfide leak, no leakage range will form on the downwind side of the hydrogen sulfide leak source, that is, the actual influence range on the downwind side is 0; based on the multimodal monitoring values ​​read in Step 5.1, the influence range on the upwind side of the leak source is calculated using Equations (5.1) and (5.2):

[0089]

[0090] In the formula:

[0091] Q Leakage rate, mg / s;

[0092] μ The average wind speed in the environment at the time of the leak, in m / s;

[0093] H The effective height of the leak source, in meters (m).

[0094] σ y , σ z The diffusion systems for crosswind and vertical wind are respectively. Based on atmospheric and topographical conditions, atmospheric stability can be divided into six levels: A, B, C, D, E, and F. The corresponding crosswind and vertical wind diffusion systems are shown in Table 1.

[0095] Table 1. Diffusion coefficients for crosswind and vertical wind directions at different atmospheric stability levels.

[0096]

[0097] x ,y , z These represent the distances to the downwind direction, crosswind direction, and vertical wind direction, respectively, in meters (m).

[0098] C For a point in space ( x , y , z The concentration of toxic substances in the air, mg / m³ 3 ;

[0099] r Let the leakage orifice diameter be in meters (m).

[0100] λ The adiabatic coefficient of the gas;

[0101] P The average pressure inside the pipeline, in Pa;

[0102] M The value represents the molar mass of the leaked gas, expressed in kg / kmol.

[0103] R is the leakage gas constant;

[0104] T The temperature of the leaked gas is ℃.

[0105] Step 5.3.2: For the two-dimensional plane of the leak, integrate the maximum H2S concentration value obtained in Step 5.1. S max ,and x , y , z ≠0, the calculation formula for the Gaussian plume model is as follows:

[0106]

[0107] Step 5.3.3: Correct the Gaussian plume model obtained in step 5.3.2 using the longest horizontal distance of H2S leakage calculated in step 4. Specifically, select the corresponding horizontal distance coupling values ​​for leakage concentrations of 20ppm and 100ppm. D 1. D 2. An improved Gaussian plume model was obtained through two-layer GRNN optimization, and the actual influence range on the upwind side was calculated. The calculation formula is as follows:

[0108]

[0109] Step 5.4: Using the leak source as the center and one-third of the actual influence range on the windward side of the leak source obtained in Step 5.3 as the radius, obtain the dangerous range of the H2S gas leak center.

[0110] Step 5.5: Superimpose the actual impact range on the downwind side obtained in Step 5.3 with the actual impact range on the upwind side to obtain the actual leakage impact range of the leakage source.

[0111] Step 6: Emergency Warning and Personnel Safety Evacuation. When an H2S leak is detected using the dynamic evaluation method for the H2S transport pipeline leakage range of this invention, the chemical industrial park will be notified in a timely manner to issue an audible and visual alarm and notify the local fire department for emergency response. Workers and residents within the leakage impact range will be safely evacuated.

Claims

1. A dynamic evaluation method for the leakage range of H2S transportation pipelines, characterized in that, Specifically, the following steps are included: Step 1: Collect real-time data from the H2S sensors on the H2S transport pipeline and upload it to the industrial control computer; Step 2: Determine if there is a leak in the H2S transport pipeline and identify the source of the leak; Step 3: Determine the direction of leakage from the leak source; Step 4: Establish a dynamic evaluation model for H2S leakage and diffusion to obtain the longest horizontal distance of H2S leakage; Step 5: Determine the actual leakage impact range of the leak source; Step 6: Emergency Warning and Safe Evacuation of Personnel; Step 4 specifically includes the following steps: Step 4.1: In accordance with the provisions of the "Design Standard for Detection and Alarm of Combustible and Toxic Gases in Petrochemical Industry" and the "Guideline for Occupational Hazard Prevention of Hydrogen Sulfide", the leakage threshold of H2S is set to 20ppm, 100ppm and 500ppm. Step 4.2: Use FLUENT simulation software to simulate leakage conditions corresponding to different leakage concentrations, pipeline pressures, and ambient wind speeds. The parameters are: 1% < leakage concentration < 20%, 0.5 MPa < pipeline pressure < 2.5 MPa, and 0.2 m / s < ambient wind speed < 2 m / s. This yields numerical simulation data for H2S gas leakage. Historical data on H2S leakage concentration, pipeline pressure, ambient wind speed, and H2S sensor locations at the moment of the H2S pipeline leak are extracted as on-site simulation data for H2S gas leakage. Step 4.3: Use FLUENT simulation software to simulate the horizontal leakage distance corresponding to different leakage thresholds in Step 4.

1. d Simulations were conducted using data from the field simulation, including leakage concentration, pipeline pressure, ambient wind speed, leakage threshold, and the corresponding horizontal distance from the leakage threshold. d Using the horizontal axis, the numerical simulation data and field simulation data of H2S gas leakage obtained in step 4.2 are randomly divided into five groups of data, and these five groups of data are used as the vertical axis to divide the H2S data into a 5*5 feature data matrix. Step 4.4: Input the feature data matrix obtained in Step 4.3 into the GAN network for feature fusion, and train to generate H2S dynamic evaluation model data with two data features: H2S gas leakage numerical simulation data and H2S gas leakage field simulation data. Step 4.5: Use the H2S dynamic evaluation model data from Step 4.4 to train a two-layer GRNN network; Step 4.6: Complete the training process of Step 4.5 to obtain the H2S leakage diffusion dynamic evaluation model, thereby calculating the longest horizontal distance of H2S leakage in real time. Step 5 specifically includes the following steps: Step 5.1, at the first read rate v 1. Real-time reading of ambient wind speed, ambient temperature, pipeline pressure, and values ​​of H2S sensors at various points on the H2S transport pipeline; Step 5.2: Input the ambient wind speed, ambient temperature, pipeline pressure, and H2S sensor values ​​obtained in Step 5.1, along with the values ​​from each point on the transport pipeline, into the H2S leakage diffusion dynamic evaluation model obtained in Step 4, to obtain the leakage horizontal distance coupling value corresponding to different leakage thresholds. D ; Step 5.3: Based on the multimodal monitoring values ​​read in Step 5.1 and the modified Gaussian plume model of the longest horizontal distance of H2S leakage calculated in Step 4, further calculate the actual influence range of the leak source on the windward side. Step 5.4: Using the leak source as the center and one-third of the actual influence range on the windward side of the leak source obtained in Step 5.3 as the radius, obtain the dangerous range of the H2S gas leak center. Step 5.5: Superimpose the actual impact range on the downwind side obtained in Step 5.3 with the actual impact range on the upwind side to obtain the actual leakage impact range of the leakage source; Step 5.3 specifically includes the following steps: Step 5.3.1: Since the hydrogen sulfide on the downwind side will be directly blown away by the wind after the hydrogen sulfide leak, no leakage range will form on the downwind side of the hydrogen sulfide leak source, that is, the actual influence range on the downwind side is 0; based on the multimodal monitoring values ​​read in Step 5.1, the influence range on the upwind side of the leak source is calculated using Equations (5.1) and (5.2): In the formula: Q Leakage rate, mg / s; μ The average wind speed in the environment at the time of the leak, in m / s; H The effective height of the leak source, in meters (m). σ y , σ z The diffusion systems for crosswind and vertical wind are respectively. Based on atmospheric and topographical conditions, atmospheric stability can be divided into six levels: A, B, C, D, E, and F. The corresponding crosswind and vertical wind diffusion systems are shown in Table 1. Table 1. Diffusion coefficients for crosswind and vertical wind directions at different atmospheric stability levels. x , y , z These represent the distances to the downwind direction, crosswind direction, and vertical wind direction, respectively, in meters (m). C For a point in space ( x , y , z The concentration of toxic substances in the air, mg / m³ 3 ; r Let the leakage orifice diameter be in meters (m). λ The adiabatic coefficient of the gas; P The average pipeline pressure is expressed in Pa. M The value represents the molar mass of the leaked gas, expressed in kg / kmol. R is the leakage gas constant; T Temperature of the leaked gas, in °C; Step 5.3.2: For the two-dimensional plane of the leak, integrate the maximum H2S concentration value obtained in Step 5.

1. S max ,and x , y , z ≠0, the calculation formula for the Gaussian plume model is as follows: Step 5.3.3: Correct the Gaussian plume model obtained in step 5.3.2 using the longest horizontal distance of H2S leakage calculated in step 4. Specifically, select the corresponding horizontal distance coupling values ​​for leakage concentrations of 20ppm and 100ppm. D 1. D 2. An improved Gaussian plume model was obtained through two-layer GRNN optimization, and the actual influence range on the upwind side was calculated. The calculation formula is as follows:

2. The dynamic evaluation method for the leakage range of H2S transportation pipelines according to claim 1, characterized in that, In step 1, there are n H2S sensors, which are evenly distributed on the H2S transport pipeline to monitor the H2S concentration at various points outside the pipeline in real time. The real-time data from the H2S sensors is uploaded to an industrial control computer equipped with a dynamic evaluation model of the H2S transport pipeline leakage range via RS485 communication.

3. The dynamic evaluation method for the leakage range of H2S transportation pipelines according to claim 1 or 2, characterized in that, Step 2 specifically includes the following steps: Step 2.1, at the first read rate v 1. Read the values ​​from the H2S sensors at each location and record them as follows: S 1. S 2, ... S n-1 , S n The maximum concentration of H2S was obtained by performing maximum value processing. S max ; at the second read rate v 2. Take two consecutive readings of the H2S sensor values ​​at each location and calculate the concentration increment corresponding to the two consecutive readings at that location, denoted as Δ. S 1. Δ S 2、……、Δ S n-1 Δ S n And perform maximum value processing to obtain the maximum H2S concentration increment Δ S max ; Step 2.2: Set the maximum safety threshold α and the maximum mutation safety threshold β for H2S. If S max ≥ɑ or Δ S max If ≥β, then a leak has occurred in the H2S transport pipeline, and the leak source is determined by the H2S sensor location where the leak occurred; proceed to step 3. S max <ɑ and Δ S max If <β, then no leak has occurred in the H2S transport pipeline; return to step 2.

1. In step 2.2, the maximum safety threshold α for H2S is 20 ppm, and the maximum mutation safety threshold β for H2S is 1.

4. The dynamic evaluation method for the leakage range of H2S transportation pipelines according to claim 3, characterized in that, The specific process for determining the leakage direction in step 3 is as follows: read the wind direction at the location of the leakage source in the meteorological monitoring station, where the leakage direction of the leakage source is opposite to the wind direction.

5. The dynamic evaluation method for the leakage range of H2S transportation pipelines according to claim 4, characterized in that, The specific training process of the two-layer GRNN network described in step 4.5 is as follows: Step 4.5.1: Using the leakage concentration, pipeline pressure, ambient wind speed, and leakage threshold from the H2S dynamic evaluation model data obtained in Step 4.4 as input values, and the leakage horizontal distance corresponding to the leakage threshold... d As the output value, it is input into the first layer of the GRNN network to obtain the leakage horizontal distance coupling value of a single-layer GRNN network, and the leakage horizontal distance is calculated. d The absolute difference Δ between the leakage horizontal distance coupling value and that of a single-layer GRNN network d ; Step 4.5.2: Using the leakage concentration, pipeline pressure, ambient wind speed, leakage threshold, and absolute difference Δ obtained in step 4.5.1 from the H2S dynamic evaluation model data obtained in step 4.

4. d As input, the distance to the leakage level corresponding to the leakage threshold. d As the output value, it is input into the second-layer GRNN network to obtain the leakage horizontal distance coupling value of the two-layer GRNN network. D .

6. The dynamic evaluation method for the leakage range of H2S transportation pipelines according to claim 1, characterized in that, Step 4 describes the training process of the two-layer GRNN network, which utilizes the particle swarm optimization algorithm to optimize the model training process and obtain the optimal smoothing factor for the model.

7. The dynamic evaluation method for the leakage range of H2S transportation pipelines according to claim 3, characterized in that, First read rate v 1 represents 1 read / second; the second read rate v 2 means 2 times / second.

Images