Heat exchanger surface accurate positioning and automatic cleaning method based on computer vision

Abstract

The invention discloses a heat exchanger surface accurate positioning and automatic cleaning method based on computer vision. The heat exchanger surface accurate positioning and automatic cleaning method comprises the following steps of S1, video acquisition and preprocessing; s2, performing pipeline target detection and positioning by using a YOLO algorithm, and optimizing convolution calculation efficiency by using a CSPNet structure; s3, position correction is conducted, specifically, the displacement amount between the cleaning equipment and the pipeline is calculated according to the obtained pipeline position; s4, dynamic adjustment is conducted, specifically, the position deviation is dynamically adjusted, and the position and angle of the cleaning equipment are adjusted according to the deviation value obtained through calculation; s5, confirming the position; and S6, a cleaning equipment control instruction is generated, and the pipeline is cleaned. Through the computer vision and deep learning technology, automatic and accurate positioning of the surface pipeline of the heat exchanger is achieved, and efficient cleaning is completed in combination with automatic equipment. Compared with a traditional method, the cleaning efficiency and accuracy are remarkably improved, manual intervention is reduced, and high-precision positioning can be kept in long-term work.

Description

Technical Field

[0001] The present invention relates to the technical field of heat exchanger cleaning, and in particular to a method for accurate positioning and automatic cleaning of a heat exchanger surface based on computer vision. Background Art

[0002] Traditional heat exchanger cleaning methods usually rely on manual operations, resulting in low efficiency and difficult construction. During long cleaning operations, positional deviation and poor cleaning effects are prone to occur. In addition, the existing semi-automatic equipment has limited accuracy when dealing with complex working conditions, which can easily cause error accumulation and affect the cleaning quality. Therefore, there is an urgent need for an automated solution based on computer vision that can identify the position of the pipes on the surface of the heat exchanger in real time to achieve precise positioning and efficient cleaning. Summary of the invention

[0003] In view of the above-mentioned technical deficiencies, the purpose of the present invention is to provide a method for precise positioning and automatic cleaning of the heat exchanger surface based on computer vision, which can achieve precise positioning of the pipeline and control the cleaning equipment for efficient cleaning, thereby solving the problem that manual cleaning or semi-automatic equipment has limited adaptability in complex environments, which easily leads to inadequate cleaning.

[0004] In order to solve the above technical problems, the present invention adopts the following technical solutions: The present invention provides a method for accurate positioning and automatic cleaning of a heat exchanger surface based on computer vision, which is characterized by comprising the following steps: S1. Video acquisition and preprocessing: Use a high-definition camera to collect real-time video streams of the heat exchanger surface, preprocess the collected heat exchanger surface images to remove noise, use edge detection algorithms to identify the shape edges of the heat exchanger pipes, and extract pipe features; S2. Use deep learning algorithms for pipeline target detection and positioning: Use the YOLO algorithm for pipeline target detection and positioning. The YOLO algorithm can quickly identify and mark the location of the pipeline in the video frame and generate accurate bounding box coordinates. The CSPNet (Cross-Stage Partial Network) structure is used to optimize the convolution calculation efficiency and generate multi-scale feature maps. S3, position correction: obtain the current position and status information of the cleaning equipment in real time, obtain feedback data, and calculate the position deviation between the cleaning equipment and the pipeline based on the feedback data obtained and the initially detected pipeline position; S4, dynamic adjustment: dynamically adjust the position deviation, and adjust the position and angle of the cleaning equipment according to the calculated deviation value; S5, position confirmation: After completing the position adjustment, the system confirms the alignment between the cleaning equipment and the pipeline through another test; S6. Generate cleaning equipment control instructions and clean the pipeline: The system generates cleaning equipment control instructions based on the identification and correction results, and controls the cleaning equipment to perform cleaning operations on the pipeline.

[0005] Preferably, in step S1, the Canny edge detection algorithm is used to detect the outer edge of the heat exchanger pipe, and the formula is as follows: ; Among them, G x and G y are the gradients of the image on the x-axis and y-axis respectively.

[0006] Preferably, in step S2, the formula for optimizing the convolution calculation efficiency using the CSPNet (cross-stage partial network) structure is: ; Among them, W is the convolution kernel, X is the input image feature map, b is the bias term, and σ is the activation function.

[0007] Preferably, step S3 uses the following formula to calculate the position deviation: ; in: is the position deviation; are the current coordinates of the device; are the detected pipe position coordinates.

[0008] Preferably, the current position and status information of the cleaning device in step S3 includes the coordinates of the device, the nozzle angle and the position deviation relative to the pipeline. Preferably, the cleaning equipment control instructions in step S6 include cleaning position, force and spray direction.

[0009] The beneficial effects of the present invention are as follows: the present invention realizes automatic and precise positioning of the surface pipes of the heat exchanger through computer vision and deep learning technology, and completes efficient cleaning in combination with automated equipment; compared with traditional methods, the present invention significantly improves the efficiency and accuracy of cleaning, reduces manual intervention, and can maintain high-precision positioning during long-term work. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0011] Figure 1A flow chart of a method for accurate positioning and automatic cleaning of a heat exchanger surface based on computer vision provided in an embodiment of the present invention; Figure 2 A system architecture diagram of a method for precise positioning and automatic cleaning of a heat exchanger surface based on computer vision provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0012] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0013] Example Take industrial heat exchanger cleaning as an example: Suppose an industrial enterprise uses a heat exchanger for a cooling system. The surface of the heat exchanger is covered with pipes. As time goes by, dirt will accumulate inside the pipes, affecting the heat exchange efficiency, and regular cleaning is required.

[0014] Operation process: The server receives the video stream from the camera and passes it to the YOLO target detection model.

[0015] like Figure 1 As shown, the cleaning steps are as follows: 1. System architecture and equipment configuration (such as Figure 2 shown) In order to realize the accurate positioning and automatic cleaning method of the heat exchanger surface based on computer vision, the system includes the following main hardware components: Network switch: responsible for connecting various devices to ensure stable transmission of data flow, including servers, cameras, PLCs, cleaning devices, etc.

[0016] Server: used to process video image data, run deep learning models, and calculate adjustment instructions and position corrections for cleaning equipment.

[0017] High-definition camera: installed on the cleaning equipment, used to collect image data on the surface of the heat exchanger in real time.

[0018] PLC (Programmable Logic Controller): Controls the operation of the cleaning equipment, receives control instructions from the server, and drives the movement and operation of the cleaning device.

[0019] Cleaning device: includes a nozzle and a robotic arm, which can adjust the position according to instructions, spray water or cleaning agent, and perform the actual cleaning work.

[0020] Monitoring screen: used to display system operation status, monitor video and cleaning progress in real time, and provide manual intervention and management interface.

[0021] 2. Video acquisition and preprocessing Equipment configuration: Install a high-definition camera on the robot arm or fixed position of the cleaning equipment to ensure that the image of the heat exchanger surface can be clearly captured.

[0022] Operation process: The high-definition camera collects images of the heat exchanger surface in real time and transmits the image data to the server. After receiving the image, the server uses a preprocessing algorithm (such as Canny edge detection) to remove noise and extract the outer edges of the pipe, providing accurate image input for subsequent target detection and positioning.

[0023] 3. Pipeline target detection and positioning Device configuration: Run the YOLO algorithm on the server for target detection to identify the location of the pipes on the surface of the heat exchanger. The YOLO algorithm uses a deep learning model to quickly and accurately locate the pipes in the image and generate a bounding box.

[0024] YOLO compares the training data set with the real-time image to identify the boundary of the pipeline and output the precise coordinate information of the pipeline. At this time, the positioning results of the image can be displayed on the monitoring screen for manual real-time viewing.

[0025] 4. Position Correction Equipment configuration: Get the current position information and nozzle angle of the cleaning equipment (such as a robotic arm), which is provided by sensors installed on the equipment. The PLC obtains feedback data from the equipment in real time by communicating with the server.

[0026] Operation process: The server calculates the deviation between the pipeline position detected by YOLO and the current position of the cleaning device. For example, if the detected pipeline position is offset by 10 cm, the server will calculate the position deviation that needs to be adjusted, and the calculation result will be transmitted to the cleaning device through PLC.

[0027] 5. Dynamic Adjustment Equipment configuration: The cleaning device (robotic arm, nozzle) is connected to the PLC and receives control instructions from the server for dynamic adjustment.

[0028] Operation process: According to the result of position correction, PLC controls the cleaning equipment to adjust the position and nozzle angle. For example, if the system calculates that the cleaning equipment is 10 cm away from the pipe position, PLC will drive the robot arm to move and adjust the nozzle angle to ensure the alignment of the cleaning target.

[0029] PLC instructions: Dynamic adjustment commands (including position coordinates, angle adjustment, etc.) are executed by the PLC to control the cleaning equipment. After the adjustment is completed, the PLC will feedback the equipment status to the server to ensure accurate position correction.

[0030] 6. Location confirmation Equipment configuration: Use cameras and sensors to double-check the alignment of the cleaning equipment with the pipe.

[0031] Operation process: After the position is adjusted, the system will use the camera to capture the pipeline image again and re-run the target detection algorithm (YOLO) to confirm whether the cleaning equipment has been accurately aligned with the pipeline. If the position is not accurate, the system will issue a position adjustment instruction again until it is aligned.

[0032] 7. Generation and execution of cleaning equipment control instructions Equipment configuration: PLC controls the cleaning device to perform spraying and cleaning operations. The control instructions include parameters such as cleaning position, cleaning force, and spraying direction.

[0033] Operation process: When the system confirms that the cleaning equipment has been accurately aligned with the pipeline, the server generates cleaning control instructions, including information such as cleaning position, force, and spray direction. These instructions are transmitted to the cleaning device (such as a high-pressure water gun nozzle or a cleaning agent spray device) through the PLC to start the actual cleaning operation.

[0034] 8. Monitoring and feedback Equipment configuration: The cleaning progress, equipment status and pipeline cleaning effect are displayed in real time on the monitoring screen. It can display real-time images, equipment location, cleaning status and other information to provide operators with supervision and management.

[0035] Operation process: By monitoring the large screen, the operator can view the real-time progress of the cleaning task, equipment status feedback, and cleaning results. If an abnormality occurs during the cleaning process, the operator can manually intervene or adjust.

[0036] Through the above implementation process, the computer vision-based heat exchanger surface precise positioning and automatic cleaning method can be combined with specific hardware (such as network switches, cameras, PLCs, cleaning devices, etc.) to achieve precise automatic cleaning operations. The system can automatically identify the position of the pipes on the surface of the heat exchanger, locate and correct them through deep learning algorithms, dynamically adjust the position of the cleaning equipment, and generate cleaning instructions to perform efficient cleaning tasks. Finally, the entire process is monitored in real time through the large monitoring screen to ensure the accuracy and efficiency of the cleaning process.

[0037] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A heat exchanger surface precise positioning and automatic cleaning method based on computer vision, characterized in that: The following steps are involved: S1. Video acquisition and preprocessing: Use a high-definition camera to collect real-time video streams of the heat exchanger surface, preprocess the collected heat exchanger surface images to remove noise, use edge detection algorithms to identify the shape edges of the heat exchanger pipes, and extract pipe features; S2. Use deep learning algorithms for pipeline target detection and positioning: Use the YOLO algorithm for pipeline target detection and positioning, use the CSPNet structure to optimize convolution calculation efficiency, and generate multi-scale feature maps; S3, position correction: obtain the current position and status information of the cleaning equipment in real time, obtain feedback data, and calculate the position deviation between the cleaning equipment and the pipeline based on the feedback data obtained and the initially detected pipeline position; S4, dynamic adjustment: dynamically adjust the position deviation, and adjust the position and angle of the cleaning equipment according to the calculated deviation value; S5, position confirmation: After completing the position adjustment, the system confirms the alignment between the cleaning equipment and the pipeline through another test; S6. Generate cleaning equipment control instructions and clean the pipeline: The system generates cleaning equipment control instructions based on the identification and correction results, and controls the cleaning equipment to perform cleaning operations on the pipeline.

2. The computer vision-based heat exchanger surface precise positioning and automatic cleaning method according to claim 1, characterized in that: In step S1, the Canny edge detection algorithm is used to detect the outer edge of the heat exchanger pipe. The formula is as follows: ； Among them, G x and G y are the gradients of the image on the x-axis and y-axis respectively.

3. The computer vision-based heat exchanger surface precise positioning and automatic cleaning method according to claim 1, characterized in that: The formula for optimizing the convolution calculation efficiency using the CSPNet structure in step S2 is: ； Among them, W is the convolution kernel, X is the input image feature map, b is the bias term, and σ is the activation function.

4. The computer vision-based heat exchanger surface precise positioning and automatic cleaning method according to claim 1, characterized in that: Step S3 uses the following formula to calculate the position deviation: ； in: is the position deviation; are the current coordinates of the device; are the detected pipe position coordinates.

5. The computer vision-based heat exchanger surface precise positioning and automatic cleaning method according to claim 1, characterized in that: The current position and status information of the cleaning device in step S3 includes the coordinates of the device, the nozzle angle and the position deviation relative to the pipeline.

6. The computer vision-based heat exchanger surface precise positioning and automatic cleaning method according to claim 1, characterized in that: The cleaning equipment control instructions in step S6 include cleaning position, force and spray direction.

Images