Routing inspection robot-based hydrogen refueling station leakage detection and positioning method

Abstract

The invention discloses a hydrogen refueling station leakage detection and positioning method based on an inspection robot, and relates to the technical field of gas leakage detection and positioning. According to the method, an inspection robot with a binocular camera, an anemorumbometer and a pumping type hydrogen sensor is combined with a hydrogen concentration positioning algorithm (TD KDM + V algorithm), so that automatic inspection, hydrogen leakage identification and leakage source positioning of key areas of the hydrogen refueling station are realized. And after detecting a hydrogen signal, the robot dynamically adjusts an inspection path according to wind speed and wind direction information, draws a hydrogen concentration distribution cloud picture and determines a leakage position. The leakage detection accuracy and the positioning efficiency are improved, the system construction and maintenance cost is reduced, and the method is suitable for a large-scale open hydrogen refueling station environment.

Description

Technical Field

[0001] The present invention is a hydrogen station leak detection and positioning method based on a patrol robot. The present invention relates to the field of gas leakage source positioning and gas detection, in particular to the field of gas leakage positioning in a large-scale open environment. Background Art

[0002] Hydrogen, as a renewable energy source, is an alternative to fossil fuels. Currently, hydrogen is mostly used and stored in the form of high-pressure gas, and hydrogen-related infrastructure is being vigorously promoted. However, compared with conventional fossil fuels such as oil and natural gas, hydrogen has a lower ignition concentration and higher explosion potential. Furthermore, storing hydrogen in high-pressure gaseous form is more damaging to containers, making it more prone to leaks. Once hydrogen leaks and explodes upon contact with a fire source, it is bound to cause significant financial and human losses. Against this backdrop, the problem of hydrogen leak detection and location in large, open environments is a current research hotspot.

[0003] Currently, the most intuitive way to monitor hydrogen leaks at hydrogen refueling stations is to use hydrogen sensing technology, by placing hydrogen sensors in key areas to monitor whether the equipment is leaking. The cost of this method will increase as the scenario increases. If it is a large-scale integrated production-storage-hydrogen refueling station, a large number of hydrogen sensors will be required to detect the environment, which will increase the construction cost of the hydrogen refueling station. At the same time, too many sensors will also increase the complexity and maintenance difficulty of the system. Another monitoring method is to use inspection robots to monitor hydrogen refueling stations. Inspection robots usually use a single concentration threshold algorithm to alarm. This method has two disadvantages: First, it cannot determine the specific location of the leak. It can only alarm after hydrogen is detected. It cannot provide assistance for subsequent processing and requires manual secondary detection; second, the environmental wind direction will affect the gas diffusion path. Using a simple concentration threshold detection will cause regional misjudgment, delaying the valuable time for early accident processing. Summary of the Invention

[0004] In view of the shortcomings of the existing technology, the purpose of the present invention is to provide a hydrogen station leak detection and positioning method based on a patrol robot to solve the problems raised in the above background technology.

[0005] According to one aspect of the present application, a hydrogen refueling station leak detection and positioning method based on a patrol robot is provided. The patrol robot includes a patrol chassis, a binocular camera, a wind vane, and a pump-suction hydrogen sensor. The binocular camera, the wind vane, and the pump-suction hydrogen sensor are respectively installed on the patrol chassis. A positioning and navigation module is provided in the patrol chassis. The positioning and navigation module is provided with a signal processor and a positioning program and inputted with a hydrogen refueling station layout map. The positioning program is set based on a hydrogen concentration positioning algorithm.

[0006] The hydrogen refueling station leak detection and positioning method comprises the following steps:

[0007] S1. The inspection robot automatically plans the inspection route for key inspection equipment in the hydrogen refueling station based on the hydrogen refueling station layout through the positioning and navigation module;

[0008] S2. Turn on the pump-type hydrogen sensor and wind speed and direction meter, and start the inspection according to the inspection route;

[0009] S3, determining whether hydrogen is detected;

[0010] S4. When hydrogen is detected, the inspection robot receives and processes the wind direction and speed signals and hydrogen concentration signals through the signal processor, and automatically activates the positioning program. The inspection robot replans the inspection route in the hydrogen refueling station area where the hydrogen concentration is detected until it detects the area with the highest hydrogen concentration;

[0011] S5. The inspection robot draws a hydrogen concentration distribution cloud map of the hydrogen refueling station scene according to the hydrogen concentration positioning algorithm, determines the leakage location and sends it to the monitoring center.

[0012] Preferably, in step S5, the hydrogen concentration positioning algorithm adopts the TD KDM+V algorithm, which discretizes the inspection area into a network and uses a Gaussian kernel function to infer the gas distribution on the network. The importance of each sample is weighted by the Gaussian kernel, and the gas distribution is estimated based on the distance between each grid cell and the corresponding measurement point. By using time-varying recency weights, the time importance of the measurement is introduced into the spatial extrapolation algorithm, thereby achieving prediction of the time-varying gas distribution;

[0013] The specific calculation steps of the TD KDM+V algorithm are as follows:

[0014] Step 1: Calculate the time item

[0015]

[0016] Among them, t * is the moment when all measurements need to be unified, t i is the measured value r i timestamp, and t1<t i <t * , η is the scaling factor;

[0017] Step 2: Calculate spatial weight

[0018]

[0019] Among them, R co is the cutoff radius, xi Represents the measured value r i The spatial position of x j Represents the unmeasured value r j spatial location;

[0020] Step 3: The spatiotemporal weight function is expressed as:

[0021]

[0022] Step 4: The distribution mean of each grid cell is expressed as:

[0023]

[0024] Preferably, the inspection robot uses the TD KDM+V algorithm to calculate the distribution average value of each grid unit in the inspection area based on the obtained wind direction and speed signals and hydrogen concentration signal data, and draws a hydrogen concentration distribution cloud map of the hydrogen refueling station scene based on the original hydrogen refueling station layout diagram and predicts the leakage source. The high fluctuation area with prominent distribution average value will be found near the leakage source, thereby determining the location of the leakage source.

[0025] Preferably, in step S3, a hydrogen concentration threshold is set in the signal processor of the inspection robot. When the hydrogen concentration detected by the pump-suction hydrogen sensor does not exceed the set threshold, the inspection robot continues to inspect along the inspection route. When the detected hydrogen concentration exceeds the set threshold, the inspection robot issues an alarm and automatically activates the positioning program.

[0026] Preferably, the inspection chassis is further provided with an audible and visual alarm module, a wireless communication module and a battery pack.

[0027] Preferably, the anemometer is an ultrasonic wind speed and direction sensor.

[0028] Preferably, a plurality of the inspection robots are configured in the hydrogen refueling station.

[0029] Preferably, the inspection chassis is selected from one of a wheeled chassis, a crawler chassis and a four-legged chassis. If the inspection route of the hydrogen refueling station is a hardened road surface, a wheeled chassis is selected. If the inspection route of the hydrogen refueling station has obstacles and steps, a four-legged chassis is selected. If the inspection robot needs to carry a fire extinguishing agent, a crawler chassis is selected.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] 1. Compared with deploying a large number of sensors, the inspection and positioning method based on the inspection robot of the present invention is more economical for large-scale hydrogen refueling station scenarios and saves the cost of hydrogen refueling station safety monitoring equipment.

[0032] 2. The present invention adopts a hydrogen concentration positioning algorithm. Compared with the value alarm method used by existing inspection robots, this method can give the specific location of the leak, provide assistance for the staff's next on-site test, simplify the detection process, and save valuable processing time.

[0033] 3. The hydrogen concentration positioning algorithm of the present invention takes into account the influence of wind direction and wind speed on gas diffusion in the hydrogen refueling station space, thereby improving positioning accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a flow chart of a hydrogen station leak detection and positioning method based on a patrol robot according to an embodiment of the present application.

[0035] Figure 2 It is a structural schematic diagram of a patrol robot according to an embodiment of the present application, which is a method for leak detection and positioning of hydrogen refueling stations based on a patrol robot.

[0036] Figure 3 This is a schematic diagram of the layout of various areas of a hydrogen refueling station and the inspection route of the inspection robot according to a hydrogen refueling station leakage detection and positioning method based on an inspection robot according to one embodiment of the present application.

[0037] Figure 4 It is a hydrogen concentration distribution cloud map of a hydrogen refueling station scene according to a hydrogen refueling station leak detection and positioning method based on a patrol robot according to one embodiment of the present application. DETAILED DESCRIPTION

[0038] In order to make the contents of this application easier to understand, the technical solutions in the embodiments of this application will be described clearly and completely below in conjunction with the drawings in the embodiments of this application.

[0039] like Figures 1 to 4As shown, a leak detection and positioning method for hydrogen refueling station based on an inspection robot is shown. The inspection robot includes an inspection chassis, a binocular camera, a wind vane and a pump-suction hydrogen sensor. The inspection chassis can be a wheeled chassis, a crawler chassis or a four-legged chassis according to the specific design of the hydrogen refueling station. If the inspection route of the hydrogen refueling station is a hardened road, a wheeled chassis is selected. If the inspection route of the hydrogen refueling station has obstacles or steps, a four-legged chassis is selected. If the inspection robot needs to carry a fire extinguishing agent, a crawler chassis is selected. It is equipped with binocular cameras, wind vanes and pump-suction hydrogen sensors. The binocular cameras can provide image information for the inspection robot. When the inspection robot moves along the predetermined route, it can avoid obstacles when it finds them ahead. The wind vane uses an ultrasonic wind speed and direction sensor. The ultrasonic wind speed and direction sensor uses ultrasonic time difference transmission to measure wind speed and direction. It has high measurement accuracy, reliable performance, and is easy to carry. It does not have any moving parts and does not require maintenance or on-site calibration. Provide real-time wind speed and direction information for the inspection robot; the pump-suction hydrogen sensor can sense hydrogen in the environment, and the sensor can be used in multiple ranges: a high-precision sensor with a range of 1ppm-1000ppm can ensure that hydrogen leaks are detected at a farther distance, and a hydrogen sensor with a range of 0-30% can help the positioning system find the source of the leak; the inspection chassis is equipped with a positioning and navigation module, an audio and visual alarm module, a wireless communication module and a battery pack. The positioning and navigation module is equipped with a signal processor and a positioning program and inputs the hydrogen station layout map. The positioning program is set based on the hydrogen concentration positioning algorithm. The signal processor receives and processes wind direction and speed signals and hydrogen concentration signal data and can activate the positioning program. The positioning and navigation module can realize automatic cruising and path planning and guide the inspection robot to move along the predetermined path according to the satellite navigation signal. In addition, multiple inspection robots are configured in the hydrogen station to adapt to large, open hydrogen station scenarios, increase the inspection frequency, and one inspection robot can only inspect a fixed area.

[0040] The hydrogen refueling station leak detection and location method includes the following steps:

[0041] S1. The inspection robot uses the positioning and navigation module to automatically plan an inspection route around key inspection equipment in the hydrogen refueling station, such as hydrogen storage tanks and compressors, based on the hydrogen refueling station layout;

[0042] S2. The inspection robot begins inspection along the inspection route and simultaneously turns on the pump-suction hydrogen sensor and ultrasonic wind speed and direction sensor to provide the inspection robot with real-time hydrogen concentration information and wind speed and direction information data;

[0043] S3. Determine whether hydrogen is detected. A hydrogen concentration threshold, such as 20 ppm, is set in the signal processor of the inspection robot. When the hydrogen concentration detected by the pump-suction hydrogen sensor does not exceed the set threshold, the inspection robot continues inspection along the inspection route. When the hydrogen sensor detects that the hydrogen concentration in the environment exceeds the set threshold, the inspection robot's sound and light alarm module sounds an alarm and automatically activates the positioning program.

[0044] S4. When the inspection robot determines that hydrogen is detected, the inspection robot receives and processes the wind direction and speed signals and the hydrogen concentration signal through the signal processor, and automatically activates the positioning program. The inspection robot replans the inspection route in the hydrogen refueling station area where the hydrogen concentration is detected until it detects the area with the highest hydrogen concentration;

[0045] S5. The inspection robot draws a hydrogen concentration distribution cloud map of the hydrogen refueling station scene according to the hydrogen concentration positioning algorithm of the positioning program, determines the leakage location and sends it to the monitoring center through the wireless communication module to guide the staff to conduct inspections.

[0046] Specifically, the hydrogen concentration positioning algorithm adopts the TD KDM+V algorithm. The TD KDM+V algorithm discretizes the inspection area into a network and uses a Gaussian kernel function to infer the gas distribution on the network. The importance of each sample is weighted by the Gaussian kernel, and the gas distribution is estimated based on the distance between each grid cell and the corresponding measurement point. By using time-varying recency weights, the temporal importance of the measurement is introduced into the spatial extrapolation algorithm, thereby realizing the prediction of time-varying gas distribution.

[0047] The specific calculation steps of the TD KDM+V algorithm are as follows:

[0048] Step 1: Calculate the time item

[0049]

[0050] Among them, t * is the moment when all measurements need to be unified, t i is the measured value r i timestamp, and t1<t i <t * , η is the scaling factor;

[0051] Step 2: Calculate spatial weight

[0052]

[0053] Among them, R co is the cutoff radius, x i Represents the measured value r i The spatial position of xj Represents the unmeasured value r j spatial location;

[0054] Step 3: The spatiotemporal weight function is expressed as:

[0055]

[0056] Step 4: The distribution mean of each grid cell is expressed as:

[0057]

[0058] Based on the real-time wind direction and speed signals and hydrogen concentration signal data, the inspection robot uses the TD KDM+V algorithm to calculate the distribution average of each grid cell in the inspection area. Based on the original hydrogen station layout diagram, it draws a hydrogen concentration distribution cloud map of the hydrogen station scene and predicts the source of the leak. Among them, high-fluctuation areas with prominent distribution average values ​​are usually found near the location of the leak gas source, thereby determining the location of the leak source.

[0059] The above embodiments are intended only to illustrate the technical solutions of the embodiments of the present application and are not intended to limit them. Although the embodiments of the present application have been described in detail with reference to the aforementioned embodiments, it should be understood by those skilled in the art that, without departing from the spirit and scope defined by the claims of the present application, they may still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein with equivalents.

Claims

1. A hydrogen station leak detection and positioning method based on a patrol robot, characterized in that: The inspection robot includes an inspection chassis, a binocular camera, a wind vane, and a pump-suction hydrogen sensor. The inspection chassis is respectively equipped with a binocular camera, a wind vane, and a pump-suction hydrogen sensor. A positioning and navigation module is provided in the inspection chassis. The positioning and navigation module is provided with a signal processor and a positioning program, and a layout diagram of the hydrogen refueling station is inputted therein. The positioning program is set based on a hydrogen concentration positioning algorithm. The hydrogen refueling station leak detection and positioning method comprises the following steps: S1. The inspection robot automatically plans an inspection route around key inspection equipment in the hydrogen refueling station based on the hydrogen refueling station layout through the positioning and navigation module; S2. Turn on the pump-type hydrogen sensor and wind speed and direction meter, and start the inspection according to the inspection route; S3, determining whether hydrogen is detected; S4. When hydrogen is detected, the inspection robot receives and processes the wind direction and speed signals and hydrogen concentration signals through the signal processor, and automatically activates the positioning program. The inspection robot replans the inspection route in the hydrogen refueling station area where the hydrogen concentration is detected until it detects the area with the highest hydrogen concentration; S5. The inspection robot draws a hydrogen concentration distribution cloud map of the hydrogen refueling station scene according to the hydrogen concentration positioning algorithm, determines the leakage location and sends it to the monitoring center.

2. A hydrogen station leak detection and positioning method based on a patrol robot according to claim 1, characterized in that: In step S5, the hydrogen concentration positioning algorithm adopts the TD KDM+V algorithm. The TD KDM+V algorithm discretizes the inspection area into a network and uses a Gaussian kernel function to infer the gas distribution on the network. The importance of each sample is weighted by the Gaussian kernel, and the gas distribution is estimated based on the distance between each grid cell and the corresponding measurement point. By using time-varying recency weights, the temporal importance of the measurement is introduced into the spatial extrapolation algorithm, thereby achieving the prediction of time-varying gas distribution. The specific calculation steps of the TD KDM+V algorithm are as follows: Step 1: Calculate the time item Among them, t * is the moment when all measurements need to be unified, t i is the measured value r i timestamp, and t1<t i <t * , η is the scaling factor; Step 2: Calculate spatial weight Among them, R co is the cutoff radius, x i Represents the measured value r i The spatial position of x j Represents the unmeasured value r j spatial location; Step 3: The spatiotemporal weight function is expressed as: Step 4: The distribution mean of each grid cell is expressed as:

3. A hydrogen station leak detection and positioning method based on a patrol robot according to claim 2, characterized in that: The inspection robot uses the TD KDM+V algorithm to calculate the distribution average of each grid cell in the inspection area based on the obtained wind direction and speed signals and hydrogen concentration signal data, and draws a hydrogen concentration distribution cloud map of the hydrogen refueling station scene based on the original hydrogen refueling station layout diagram to predict the leakage source. Highly fluctuating areas with prominent distribution average values ​​will be found near the leakage source, thereby determining the location of the leakage source.

4. The method for leak detection and positioning of a hydrogen refueling station based on a patrol robot according to claim 1, characterized in that: In step S3, a hydrogen concentration threshold is set in the signal processor of the inspection robot. When the hydrogen concentration detected by the pump-suction hydrogen sensor does not exceed the set threshold, the inspection robot continues to inspect along the inspection route. When the detected hydrogen concentration exceeds the set threshold, the inspection robot issues an alarm and automatically activates the positioning program.

5. The method for leak detection and positioning of a hydrogen refueling station based on a patrol robot according to claim 1, characterized in that: The inspection chassis is also provided with an audible and visual alarm module, a wireless communication module and a battery pack.

6. The method for leak detection and positioning of a hydrogen refueling station based on a patrol robot according to claim 1, characterized in that: The anemometer is an ultrasonic wind speed and direction sensor.

7. The method for leak detection and positioning of a hydrogen refueling station based on a patrol robot according to claim 1, characterized in that: A plurality of the inspection robots are configured in the hydrogen refueling station.

8. The method for leak detection and positioning of a hydrogen refueling station based on a patrol robot according to claim 1, characterized in that: The inspection chassis is selected from one of a wheeled chassis, a crawler chassis and a four-legged chassis. If the inspection route of the hydrogen refueling station is a hardened road, a wheeled chassis is selected. If the inspection route of the hydrogen refueling station has obstacles and steps, a four-legged chassis is selected. If the inspection robot needs to carry a fire extinguishing agent, a crawler chassis is selected.

Images