LNG station leakage detection method based on multi-source data

Abstract

The invention discloses an LNG station leakage detection method based on multi-source data, and the method comprises the following steps: 1, carrying out multi-source sensing, turning on an infrared thermal imaging sensor in a monitoring system, capturing fog clusters formed by low-temperature leakage through employing the absorption characteristics of methane at a specific infrared band, turning on a sonic sensor installed on a storage tank, and carrying out the detection of the fog clusters through the sonic sensor; the method comprises the steps of capturing pressure fluctuation signals caused by LNG leakage, collecting environment data through a temperature and humidity sensor and a wind speed sensor, conducting edge preprocessing, eliminating snowflake noisy points of an infrared image through median filtering, conducting risk assessment and dividing risk grades, conducting graded alarming, and conducting alarming according to the risk grades, and conducting data storage and management. According to the invention, through cooperative work of the multi-source sensors, the timeliness and reliability of leakage discovery are significantly improved, and a reliable evidence chain is provided for accident analysis, responsibility tracing and process optimization through long-term storage of videos and detection records in a high-leakage-risk time period by the cloud and with assistance of access control for tampering prevention.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of leakage detection, in particular to a LNG station leakage detection method based on multi-source data. BACKGROUND

[0002] LNG is a clean energy formed by liquefying natural gas under ultra-low temperature conditions, mainly composed of methane, a small amount of ethane, propane and a trace amount of nitrogen and other impurities. After purification and dehydration, it is cooled and liquefied at -162℃ under normal pressure, and the volume is reduced to about 1 / 625 of the gaseous state. After combustion, the pollutant emissions are significantly lower than coal and oil. After leakage, it quickly vaporizes and diffuses, and is not easy to accumulate. LNG has become a key transitional fuel for energy transformation due to its high-efficiency compression, clean and low-carbon, and flexible application. China, as the world's largest LNG importer, is promoting the replacement of high-carbon energy in the fields of electricity, transportation and industry to help achieve the "double carbon" goal.

[0003] In the utility model with application number 202320244918.5, an LNG filling station anti-leakage detection device is disclosed. By setting a flashing warning board on the equipment, the working efficiency during the later maintenance of the equipment can be significantly improved. Compared with the introduced equipment, the flashing warning board can prompt the maintenance personnel and avoid secondary leakage caused by bumping during maintenance. By setting an electric control plugging column on the equipment, the loss can be minimized. Compared with the introduced equipment, the electric control plugging column includes a plugging block and first and second electric control columns. After receiving the leakage signal, the electric control power block sends a control signal to the first and second electric control columns, so that the plugging block plugs the pipeline, controls the leakage, and facilitates the later maintenance.

[0004] However, the above-mentioned LNG filling station anti-leakage detection device judges whether gas leakage occurs by the change of the internal pressure of the pipeline, which cannot accurately locate the leakage position in time and needs the on-site investigation of the staff. The above-mentioned LNG filling station anti-leakage detection device judges whether gas leakage occurs by the change of the internal pressure of the pipeline, which is prone to misjudgment and cannot detect leakage based on multi-source data. Therefore, the present application proposes a LNG station leakage detection method based on multi-source data to solve the problems in the prior art. SUMMARY

[0005] In view of the above problems, the purpose of the present application is to provide a LNG station leakage detection method based on multi-source data, which cooperates with multi-source sensors to capture early leakage signs from physical, chemical and environmental dimensions, significantly improving the timeliness and reliability of leakage discovery, and providing reliable evidence chain for accident analysis, responsibility tracing and process optimization by storing video and detection records of high leakage risk period in the cloud for a long time, supplemented by access control to prevent tampering.

[0006] To achieve the purpose of the present application, the present application realizes the following technical scheme: a LNG station leakage detection method based on multi-source data, comprising the following steps:

[0007] Step one, multi-source perception, turn on the infrared thermal imaging sensor in the monitoring system, capture the fog cluster formed by low-temperature leakage by using the absorption characteristics of methane in a specific infrared wave band, turn on the sound wave sensor installed on the storage tank, capture the pressure fluctuation signal caused by LNG leakage, and collect environmental data by using temperature, humidity and wind speed sensors;

[0008] Step two, edge preprocessing, adopt median filtering to eliminate snowflake noise of infrared image, and adopt histogram equalization to enhance;

[0009] Step three, risk assessment, determine potential risk patterns and risk indicators, divide risk levels, train the model by using monitoring data, evaluate the leakage diffusion path according to the environmental data, and dynamically judge the leakage risk level;

[0010] Step four, graded alarm, start corresponding alarm according to the risk level evaluated in step four;

[0011] Step five, data storage and management, transmit and store video files and detection records to the cloud disk, and adopt access control mechanism to prevent data leakage and malicious tampering.

[0012] Further improvement lies in that the infrared thermal imaging sensor detects the methane characteristic absorption peak wave band as 1653nm in step one, and the sound wave sensor detects the range of 20-100kHz.

[0013] Further improvement lies in that the monitoring system data collects LNG leakage images as a training set in step one, and labels them for training machine learning model.

[0014] Further improvement lies in that the graded alarm in step four is divided into first-level alarm and second-level alarm, the first-level alarm is sound-light alarm and positioning information push alarm, and the second-level alarm is conveying valve closing and evacuation alarm.

[0015] Further improvement lies in that the high-leakage-level data corresponding to the time recording file and the detection record stored in the step five are long-term stored, and the low-risk data under normal working conditions are locally cleaned and deleted in time.

[0016] Further improvement lies in that the monitoring system in the step one is used for monitoring the change of the LNG station storage tank in real time, including an optical imaging module, an acoustic sensing module, a liquid level monitoring module and an environment monitoring module, the optical imaging module is used for monitoring and observing the LNG station storage tank in real time through an optical camera, the acoustic sensing module is used for capturing the tiny vibration and pressure fluctuation signal of the storage tank through an acoustic sensor, the liquid level monitoring module is used for indirectly judging whether there is leakage by monitoring the liquid level change in the storage tank, and the environment monitoring module is used for monitoring the environmental data in real time through a temperature and humidity sensor and a wind speed sensor, so as to provide data support for predicting the leakage diffusion path.

[0017] Further improvement lies in that the optical imaging module includes an infrared thermal imaging unit and a visible light camera unit, the infrared thermal imaging unit is used for detecting the methane characteristic absorption peak through an infrared thermal imaging sensor, realizing the visualization of the methane gas cloud, and the visible light camera unit is used for capturing the image of the leakage area through a visible light camera, assisting in positioning the leakage point.

[0018] The beneficial effects of the present application are that: the present application cooperates with multiple source sensors, captures the weak signs of leakage in early stage from the physical, chemical and environmental dimensions, significantly improves the timeliness and reliability of leakage discovery, stores the video and detection record of the high-leakage-risk period in the cloud for a long time, and provides a reliable evidence chain for accident analysis, responsibility tracing and process optimization, at the same time, the local low-risk normal working condition data is cleaned regularly, the storage resource utilization rate is effectively optimized, and the storage cost and management burden are reduced. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 The step flow chart of the present application is shown in the figure.

[0020] Figure 2 The monitoring system architecture diagram of the present application is shown in the figure. DETAILED DESCRIPTION

[0021] In order to deepen the understanding of the present application, the present application will be further described in combination with the embodiments below, and the present embodiment is only used to explain the present application, and does not constitute the limitation to the protection scope of the present application.

[0022] LNG receiving station is the core energy hub for receiving, storing and regasifying marine imported liquefied natural gas, usually built in coastal ports, composed of unloading area, storage tank area, process device area, etc., through special equipment to realize LNG unloading, low-temperature storage, boil-off gas treatment and gasification, to ensure the safety of natural gas supply. The core of LNG station leakage detection is the integration of multiple technologies and intelligent analysis, mainly focusing on high-risk positions, usually using double-layer tank structure to block leakage, supplemented by fire and explosion suppression system, and repairing sealing point defects regularly through LDAR technology;

[0023] Based on this, according to Figure 1 、 Figure 2 The embodiment provides a LNG station leakage detection method based on multi-source data, which comprises the following steps:

[0024] Step one, multi-source perception, turn on the infrared thermal imaging sensor in the monitoring system, capture the fog cluster formed by low-temperature leakage by using the absorption characteristics of methane in a specific infrared wave band, turn on the sound wave sensor installed on the storage tank, capture the pressure fluctuation signal caused by LNG leakage, collect environmental data by using temperature and humidity and wind speed sensors, the infrared thermal imaging sensor detects the methane characteristic absorption peak wave band as 1653nm, and the sound wave sensor detects the range of 20-100kHz;

[0025] A variety of types of sensors are used to simultaneously and comprehensively capture physical signals and environmental parameters directly or indirectly related to LNG leakage. Responsible for obtaining raw data input, it is the basis and "senses" of the entire detection system, ensuring that various signs of early leakage such as fog cluster, pressure wave and abnormal liquid level can be detected, and key environmental information for model training and diffusion analysis can be collected;

[0026] The monitoring system collects LNG leakage images as a training set and labels them for training machine learning models.

[0027] Step two, edge preprocessing, median filter is used to eliminate snowflake noise in infrared images, and histogram equalization is used to enhance;

[0028] On the edge computing device close to the sensor, the originally collected data is preliminarily cleaned and optimized. Its core goal is to eliminate noise interference and enhance the recognizability of key features, to provide clearer and more reliable input data for subsequent analysis steps, and to improve the accuracy of subsequent steps;

[0029] Step three, risk assessment, determine potential risk patterns and risk indicators, divide risk levels, train models using monitoring data, evaluate leakage diffusion paths according to environmental data, and dynamically judge the risk level of leakage;

[0030] After comprehensive analysis and processing of multi-source sensor data, the pre-trained machine learning model is used to identify leakage features, and combined with real-time environmental data to predict the diffusion path of the leakage gas. The core task is to dynamically determine whether there is a leakage risk, how big the risk is, and the possible impact range, and output a quantitative risk level accordingly, providing key basis for subsequent response decisions.

[0031] Step four, hierarchical alarm, according to the risk level evaluated in step four, start the corresponding alarm, hierarchical alarm is divided into first-level alarm and second-level alarm, first-level alarm is sound and light alarm and positioning information push alarm, second-level alarm is delivery valve closing and evacuation alarm;

[0032] According to the risk level obtained by the risk assessment step, trigger the response measures of different levels matched with it. The core purpose is to timely and effectively convert risk information into specific action instructions: first-level alarm such as sound and light alarm, positioning push for early warning and prompt for lower risk, second-level alarm (such as valve closing, evacuation) for emergency disposal and personnel safety protection for high risk, to achieve "the greater the risk, the stronger the response" of precise emergency management.

[0033] Step five, data storage and management, video files and detection records are transmitted and stored to the cloud disk, and access control mechanism is used to prevent data leakage and malicious tampering;

[0034] The video files and detection records corresponding to the time of the stored high leakage level data are saved for a long time, and the low risk data under normal working conditions are deleted locally at regular intervals;

[0035] Systematically process all data generated during the detection process. On the one hand, store key evidence safely and permanently, and upload video and records of high leakage risk period to the cloud for accident tracing and analysis; on the other hand, efficiently manage storage resources, regularly clean up local data with low risk or no risk. At the same time, through the implementation of access control and other security mechanisms, the integrity, confidentiality and compliance of data are guaranteed, and data loss, leakage or tampering are prevented, serving long-term security audit and system optimization.

[0036] The monitoring system is used to monitor the changes of the LNG station in real time, including optical imaging module, acoustic sensing module, liquid level monitoring module and environmental monitoring module. The optical imaging module is used to monitor the LNG station in real time through optical camera, the acoustic sensing module is used to capture the micro-vibration and pressure fluctuation signal of the tank through acoustic sensor, the liquid level monitoring module is used to indirectly determine whether there is leakage by monitoring the change of liquid level in the tank, and the environmental monitoring module is used to monitor the environmental data in real time through temperature and humidity sensor and wind speed sensor, providing data support for predicting the diffusion path of the leakage.

[0037] The optical imaging module comprises an infrared thermal imaging unit and a visible light camera unit, the infrared thermal imaging unit is used for detecting a methane characteristic absorption peak through an infrared thermal imaging sensor to realize visualization of a methane gas cloud, and the visible light camera unit is used for capturing an image of a leakage area through a visible light camera to assist in positioning a leakage point.

[0038] The above shows and describes the basic principles, main features and advantages of the present application. Those skilled in the art should understand that the present application is not limited to the above-mentioned embodiments, and the above-mentioned embodiments and descriptions in the specification are only to illustrate the principles of the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the claimed present application. The scope of protection of the present application is defined by the appended claims and their equivalents.

Claims

1. A LNG station leakage detection method based on multi-source data, comprising the following steps: Step one, multi-source perception, turn on the infrared thermal imaging sensor in the monitoring system, capture the fog cluster formed by low-temperature leakage by using the absorption characteristics of methane in a specific infrared wave band, turn on the sound wave sensor installed on the storage tank, capture the pressure fluctuation signal caused by LNG leakage, collect environmental data by using temperature and humidity and wind speed sensors; Step two, edge preprocessing, adopt median filtering to eliminate snowflake noise of infrared image, and adopt histogram equalization to enhance; Step three, risk assessment, determine potential risk mode and risk index, divide risk level, train model by using monitoring data, evaluate leakage diffusion path according to environmental data, and dynamically judge leakage risk level; Step four, graded alarm, start corresponding alarm according to the risk level evaluated in step four; Step five, data storage and management, transmit and store video files and detection records to cloud disk, and at the same time, adopt access control mechanism to prevent data leakage and malicious tampering.

2. The LNG station leakage detection method based on multi-source data according to claim 1, characterized in that: The infrared thermal imaging sensor detects the methane characteristic absorption peak wave band as 1653 nm in the step one, and the sound wave sensor detects the range of 20-100 kHz.

3. The LNG station leakage detection method based on multi-source data according to claim 1, characterized in that: The monitoring system data collects LNG leakage image as a training set in the step one, and labels it for training machine learning model.

4. The LNG station leakage detection method based on multi-source data according to claim 1, characterized in that: The graded alarm in the step four is divided into first-level alarm and second-level alarm, the first-level alarm is sound and light alarm and positioning information push alarm, and the second-level alarm is conveying valve closing and evacuation alarm.

5. The method of claim 1, wherein: The video files and detection records corresponding to the time of high leakage level data are stored for a long time in the step five, and the low-risk data under normal working conditions are deleted by local timing cleaning.

6. The method of claim 1, wherein: The monitoring system in the step one is used for real-time monitoring of the change of the storage tank in the LNG station, which includes an optical imaging module, an acoustic sensing module, a liquid level monitoring module and an environmental monitoring module, the optical imaging module is used for real-time monitoring and observation of the LNG station storage tank through an optical camera, the acoustic sensing module is used for capturing the micro-vibration and pressure fluctuation signal of the storage tank through a sound wave sensor, the liquid level monitoring module is used for indirectly judging whether there is leakage by monitoring the liquid level change in the storage tank, and the environmental monitoring module is used for real-time monitoring of environmental data by temperature and humidity sensor and wind speed sensor, which provides data support for predicting the leakage diffusion path.

7. The method of claim 1, wherein: The optical imaging module includes an infrared thermal imaging unit and a visible light camera unit, the infrared thermal imaging unit is used for detecting methane characteristic absorption peak by infrared thermal imaging sensor to realize methane gas cloud visualization, and the visible light camera unit is used for capturing leakage area image by visible light camera to assist in locating the leakage point.

Images