Onboard VOCs gas leakage visual inspection system

Abstract

The invention discloses an airborne VOCs gas leakage visual inspection system, which comprises an unmanned aerial vehicle platform, an integrated RTK positioning module and an obstacle avoidance sensor, the infrared imaging pod is mounted on the unmanned aerial vehicle platform and comprises a shell, and a 320 * 256 pixel II-type superlattice refrigeration infrared detector, a narrow-band filter, a three-axis image stabilization holder and a temperature and humidity sensor which are mounted on the shell; the data transmission unit comprises a 4G / 5G image transmission module and an AES encryption chip, ensures data security, and transmits the image data acquired by the infrared imaging pod to a ground control terminal through encryption; and the ground control terminal comprises a software system and hardware equipment. The invention discloses an airborne visual inspection system for VOCs gas leakage, and aims to solve the problem that VOCs gas leakage in an industrial scene is difficult to detect quickly, accurately and safely.

Description

Technical Field

[0001] This invention belongs to the field of environmental monitoring technology, specifically relating to an airborne VOCs gas leakage visualization inspection system. Background Technology

[0002] There are many technical challenges in detecting VOCs gas leaks in existing industrial settings: Low inspection efficiency: Traditional manual inspections or fixed equipment have limited coverage and are difficult to adapt to large-area and complex terrain scenarios such as petrochemical and natural gas storage and transportation. Insufficient real-time capability: Reliance on offline data collection makes it impossible to locate the leak source in real time and monitor it remotely, resulting in a delay in accident response; Insufficient sensitivity and accuracy: Uncooled infrared detectors have difficulty detecting trace VOC leaks (such as methane and benzene) and are easily affected by background interference; Poor adaptability to complex environments: Traditional equipment lacks stability in explosion-proof and corrosive environments (such as oil refineries and offshore platforms); Poor data synchronization: There is a lack of efficient data fusion and visualization processing capabilities when multiple devices conduct collaborative inspections.

[0003] Therefore, there is an urgent need for a VOCs leak detection system that can achieve wide-range, high-sensitivity, real-time, safe and reliable detection. Summary of the Invention

[0004] The main objective of this invention is to provide an airborne VOCs gas leak visualization inspection system, which aims to solve the problem of rapid, accurate, and safe detection of VOCs gas leaks in industrial settings.

[0005] To achieve the above objectives, this invention provides an airborne VOCs gas leak visualization inspection system, comprising a drone platform, an infrared imaging pod, a data transmission unit, and a ground control terminal, wherein: The drone platform integrates an RTK positioning module and obstacle avoidance sensors; An infrared imaging pod, mounted on an unmanned aerial vehicle platform, includes an outer shell and a 320×256 pixel Type II superlattice cooled infrared detector (3.2-3.5μm), a narrowband filter, a three-axis image-stabilized gimbal, and a temperature and humidity sensor mounted on the outer shell. The data transmission unit includes a 4G / 5G image transmission module (supporting 1080P transmission) and an AES encryption chip to ensure data security, transmitting encrypted image data collected by the infrared imaging pod to the ground control terminal. The ground control terminal includes a software system and hardware devices. The software system supports leak point coordinate calibration, AI algorithm-driven leak cloud tracking, and concentration inversion. The hardware devices are compatible with portable workstations and mobile terminals.

[0006] As a further preferred technical solution to the above technical solution, the workflow is as follows: Step 1: The inspection area is preset through the ground terminal, and the UAV platform plans the optimal path based on the GIS map and starts the flight; Step 2: The infrared imaging pod acquires infrared and visible light images in real time. The detector filters background noise through a 3.3μm filter and identifies VOCs characteristic radiation. Step 3: The data transmission unit encrypts the dual-light image and transmits it to the ground terminal. The AI ​​algorithm performs image fusion processing and dynamically renders the leaked cloud (red warning). Step 4: The system uses RTK to locate and mark the coordinates of the leak point (accuracy ±1 meter), calculates the leakage amount using the concentration inversion model, and triggers an alarm; Step 5: After the inspection is completed, the data is automatically synchronized to the LDAR system and the environmental protection platform to generate a compliance report.

[0007] As a further preferred technical solution to the above technical solution, the infrared imaging pod acquires infrared signals using an adaptive infrared signal sampling algorithm: First, dynamic sampling strategy: Assume ambient temperature Target gas temperature (Preset value), calculate the proportional coefficient ( m (For fixed parameters) when When the temperature difference is large, reduce the sampling frequency to reduce the amount of data; when the temperature difference is small, increase the sampling frequency to ensure accuracy. Total sampling time ( (Initial time), sampled data volume ( (for the initial frequency). Second, feedback verification mechanism: calculate the mean square error of the sampled data. ,like ( If the parameters are fixed, data resampling will be triggered to avoid false detections caused by environmental interference.

[0008] As a further preferred technical solution to the above technical solution, gas leak detection is performed using AI algorithms: First, dual-light fusion and target recognition: Visible light images provide scene structure information, infrared images extract gas radiation features, and pixel-level fusion enhances the contrast of the leak area; A convolutional neural network (CNN) is used to identify the morphology of leaked cloud clusters, dynamically label their locations, and calculate their diffusion trends. Second, the concentration inversion model: Based on the radiative transfer equation, combined with the gas absorption coefficient and path length, the VOCs concentration distribution is inverted to generate a quantitative report.

[0009] As a further preferred embodiment of the above technical solution, the center wavelength of the narrowband filter is 3.3 μm.

[0010] The beneficial effects of this invention are as follows: Highly efficient coverage: The area covered by a single inspection is 5-10 times that of traditional methods; Real-time response: Real-time image transmission of dual-light technology, with leak point location accuracy up to ±1 meter; High sensitivity: The minimum detectable flow rate is superior to similar international products; Adaptable to complex environments: Meets explosion-proof, waterproof, and corrosion-resistant requirements, suitable for high-risk scenarios; Intelligent analysis: Supports dynamic tracking of leaked cloud clusters, concentration inversion, and synchronous analysis of multi-source data. Attached Figure Description

[0011] Figure 1 This is a schematic diagram of the present invention. Detailed Implementation

[0012] The following description is intended to disclose the present invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art. The basic principles of the invention defined in the following description can be applied to other embodiments, modifications, improvements, equivalents, and other technical solutions that do not depart from the spirit and scope of the invention.

[0013] In the preferred embodiments of the present invention, those skilled in the art should note that the VOCs gases involved in the present invention can be considered as prior art.

[0014] Preferred embodiment.

[0015] like Figure 1 As shown, this invention discloses an airborne VOCs gas leak visualization inspection system, including a UAV platform, an infrared imaging pod, a data transmission unit, and a ground control terminal, wherein: The drone platform integrates an RTK positioning module and obstacle avoidance sensors; An infrared imaging pod, mounted on an unmanned aerial vehicle platform, includes an outer shell and a 320×256 pixel Type II superlattice cooled infrared detector (3.2-3.5μm), a narrowband filter, a three-axis image-stabilized gimbal, and a temperature and humidity sensor mounted on the outer shell. The data transmission unit includes a 4G / 5G image transmission module (supporting 1080P transmission) and an AES encryption chip to ensure data security and transmit encrypted image data collected by the infrared imaging pod to the ground control terminal. The ground control terminal includes a software system and hardware devices. The software system supports leak point coordinate calibration, AI algorithm-driven leak cloud tracking, and concentration inversion. The hardware devices are compatible with portable workstations and mobile terminals.

[0016] Specifically, the workflow is as follows: Step 1: The inspection area is preset through the ground terminal, and the UAV platform plans the optimal path based on the GIS map and starts the flight; Step 2: The infrared imaging pod acquires infrared and visible light images in real time. The detector filters background noise through a 3.3μm filter and identifies VOCs characteristic radiation. Step 3: The data transmission unit encrypts the dual-light image and transmits it to the ground terminal. The AI ​​algorithm performs image fusion processing and dynamically renders the leaked cloud (red warning). Step 4: The system uses RTK to locate and mark the coordinates of the leak point (accuracy ±1 meter), calculates the leakage amount using the concentration inversion model, and triggers an alarm; Step 5: After the inspection is completed, the data is automatically synchronized to the LDAR system and the environmental protection platform to generate a compliance report.

[0017] More specifically, the infrared imaging pod acquires infrared signals using an adaptive infrared signal sampling algorithm: First, dynamic sampling strategy: Assume ambient temperature Target gas temperature (Preset value), calculate the proportional coefficient ( m (For fixed parameters) when When the temperature difference is large, reduce the sampling frequency to reduce the amount of data; when the temperature difference is small, increase the sampling frequency to ensure accuracy. Total sampling time ( (Initial time), sampled data volume ( (for the initial frequency). Second, feedback verification mechanism: calculate the mean square error of the sampled data. ,like ( If the parameters are fixed, data resampling will be triggered to avoid false detections caused by environmental interference.

[0018] Furthermore, AI algorithms are used for gas leak detection: First, dual-light fusion and target recognition: Visible light images provide scene structure information, infrared images extract gas radiation features, and pixel-level fusion enhances the contrast of the leak area; A convolutional neural network (CNN) is used to identify the morphology of leaked cloud clusters, dynamically label their locations, and calculate their diffusion trends. Second, the concentration inversion model: Based on the radiative transfer equation, combined with the gas absorption coefficient and path length, the VOCs concentration distribution is inverted to generate a quantitative report.

[0019] Furthermore, the center wavelength of the narrowband filter is 3.3 μm.

[0020] In a pilot test at an oil refinery, the system covered 10 square kilometers in a single inspection, identifying three benzene micro-leaks (0.5 g / min) within 8 seconds, reducing detection costs by 60% compared to traditional methods. In offshore oil and gas platform applications, the system has a false alarm rate of less than 0.1% in strong wind and high humidity environments, and the accident prevention rate is improved by 90%.

[0021] It is worth mentioning that the VOCs gas and other technical features involved in this patent application should be regarded as prior art. The specific structure, working principle and possible control methods and spatial arrangement of these technical features can be adopted using conventional choices in the field, and should not be regarded as the inventive point of this patent. This patent will not be further elaborated in detail.

[0022] For those skilled in the art, modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this invention should be included within the protection scope of this invention.

Claims

1. An airborne VOCs gas leak visualization inspection system, characterized in that, It includes an unmanned aerial vehicle (UAV) platform, an infrared imaging pod, a data transmission unit, and a ground control terminal, among which: The drone platform integrates an RTK positioning module and obstacle avoidance sensors; An infrared imaging pod is installed on an unmanned aerial vehicle platform and includes an outer shell and a 320×256 pixel Type II superlattice cooled infrared detector, a narrowband filter, a three-axis image-stabilized gimbal and a temperature and humidity sensor installed on the outer shell. The data transmission unit includes a 4G / 5G image transmission module and an AES encryption chip to ensure data security and transmit encrypted image data collected by the infrared imaging pod to the ground control terminal. The ground control terminal includes a software system and hardware devices. The software system supports leak point coordinate calibration, AI algorithm-driven leak cloud tracking, and concentration inversion. The hardware devices are compatible with portable workstations and mobile terminals.

2. The airborne VOCs gas leak visualization inspection system according to claim 1, characterized in that, The workflow is as follows: Step 1: The inspection area is preset through the ground terminal, and the UAV platform plans the optimal path based on the GIS map and starts the flight; Step 2: The infrared imaging pod acquires infrared and visible light images in real time. The detector filters background noise through a 3.3μm filter and identifies VOCs characteristic radiation. Step 3: The data transmission unit encrypts the dual-light image and transmits it to the ground terminal. The AI ​​algorithm performs image fusion processing and dynamically renders the leaked cloud. Step 4: The system uses RTK to locate and mark the coordinates of the leak point, calculates the leakage amount using the concentration inversion model, and triggers an alarm; Step 5: After the inspection is completed, the data is automatically synchronized to the LDAR system and the environmental protection platform to generate a compliance report.

3. The airborne VOCs gas leak visualization inspection system according to claim 2, characterized in that, For infrared imaging pods, infrared signals are acquired using an adaptive infrared signal sampling algorithm: First, dynamic sampling strategy: Assume ambient temperature Target gas temperature Calculate the proportionality coefficient ; when When the temperature difference is large, reduce the sampling frequency to reduce the amount of data; when the temperature difference is small, increase the sampling frequency to ensure accuracy. Total sampling time Sample data volume ; Second, feedback verification mechanism: calculate the mean square error of the sampled data. ,like If this occurs, data resampling will be triggered to avoid false detections caused by environmental interference.

4. The airborne VOCs gas leak visualization inspection system according to claim 3, characterized in that, Gas leak detection using AI algorithms: First, dual-light fusion and target recognition: Visible light images provide scene structure information, infrared images extract gas radiation features, and pixel-level fusion enhances the contrast of the leak area; A convolutional neural network was used to identify the morphology of the leaking cloud, dynamically label its location, and calculate its diffusion trend. Second, the concentration inversion model: Based on the radiative transfer equation, combined with the gas absorption coefficient and path length, the VOCs concentration distribution is inverted to generate a quantitative report.

5. The airborne VOCs gas leak visualization inspection system according to claim 4, characterized in that, The center wavelength of the narrowband filter is 3.3 μm.

Images