Hydraulic buffer detection system and method based on image recognition
Through the combination of image recognition and displacement sensing technology, the efficient and accurate detection of hydraulic buffers is achieved, and the existing detection methods are solved, with low efficiency, low accuracy and high cost, and detailed inspection reports and predictive maintenance are provided. It is recommended to improve the safety and reliability of the elevator.
Patent Information
- Application Number
- CN202411245927.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-06
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-09-06
AI Technical Summary
The existing hydraulic buffer detection methods are inefficient, low in accuracy and high in cost. They cannot comprehensively and accurately evaluate the performance of hydraulic buffers, especially in large-scale inspections, which are difficult to meet the needs.
The detection system based on image recognition and displacement sensing technology is adopted to identify and extract feature parameters through image acquisition, preprocessing, and deep learning algorithms, and comprehensive performance evaluation is carried out in combination with displacement sensor data to achieve automated and high-precision detection.
It significantly improves detection efficiency and accuracy, reduces inspection costs, provides detailed inspection reports and predictive maintenance suggestions, extends the service life of hydraulic buffers, and improves the safety and reliability of elevators.
Smart Images

Figure CN118762244B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of elevator hydraulic buffer detection, and in particular to a hydraulic buffer detection system and method based on image recognition. Background Art
[0002] Elevator hydraulic buffers are key safety components in elevator systems. Their main function is to reduce the impact force of the elevator through hydraulic damping when the elevator runs to an extreme position, thereby protecting the safety of passengers and equipment. With the continuous development and widespread application of elevator technology, the performance requirements for hydraulic buffers are becoming higher and higher. However, the existing detection methods have many shortcomings and cannot comprehensively and accurately evaluate the performance of hydraulic buffers. The specific performance is as follows:
[0003] Low detection efficiency: Traditional hydraulic buffer detection methods mainly rely on manual inspection and simple mechanical equipment. Manual inspection requires technicians to check the appearance, wear and deformation of the buffer one by one. This method is not only time-consuming and labor-intensive, but also the detection efficiency decreases significantly as the number of inspections increases. In addition, the detection speed of traditional mechanical equipment is slow and cannot meet the needs of large-scale inspections.
[0004] Low detection accuracy: Manual detection is easily affected by subjective factors of the inspector, such as lack of experience, fatigue and lack of concentration, which leads to inconsistent detection results. Traditional detection equipment has limitations in accuracy and sensitivity, and cannot fully and accurately detect various performance parameters of hydraulic buffers, especially in the detection of subtle appearance defects and internal wear.
[0005] High cost: Manual inspection and traditional equipment inspection require a lot of manpower, material resources and time, especially for multiple hydraulic buffers in large elevator systems, the inspection cost is high. In addition, the maintenance and update costs of traditional inspection equipment are also high, which increases the overall inspection cost.
[0006] Insufficient data analysis: Traditional testing methods usually only focus on a single indicator of the hydraulic buffer, such as appearance inspection and simple mechanical performance testing, and lack comprehensive analysis of multiple performance parameters. Due to the lack of comprehensive analysis and processing of data, it is difficult to accurately judge the overall performance status and service life of the hydraulic buffer, and it is impossible to provide strong data support for elevator maintenance and optimization.
[0007] In view of the above problems, there is an urgent need for a new type of elevator hydraulic buffer detection system and method, which can combine advanced image recognition technology and displacement sensing technology to perform efficient and accurate comprehensive performance detection of elevator hydraulic buffers. Therefore, the elevator hydraulic buffer detection system based on image recognition and displacement sensing technology designed by the present invention came into being. The system uses image acquisition and preprocessing, uses deep learning algorithms to perform image recognition, extracts characteristic parameters of the buffer, and combines the data of the displacement sensor to perform comprehensive performance evaluation, thereby realizing automated and high-precision detection of the hydraulic buffer. Summary of the invention
[0008] The purpose of the present invention is to solve the existing technical problems mentioned in the above background technology and provide a hydraulic buffer detection system and method based on image recognition.
[0009] The above-mentioned object of the present invention is achieved by the following technical solutions:
[0010] The hydraulic buffer detection system based on image recognition includes:
[0011] Image acquisition module: used to acquire images of elevator hydraulic buffers;
[0012] Image preprocessing module: used to preprocess the collected images, including denoising, enhancement and grayscale transformation;
[0013] Image recognition module: used to recognize the pre-processed image using a deep learning algorithm and extract the characteristic parameters of the elevator hydraulic buffer;
[0014] Displacement detection module: used to monitor the displacement of the elevator hydraulic buffer in real time;
[0015] Data analysis module: used to conduct comprehensive analysis on the identified characteristic parameters and displacement to determine the performance status of the elevator hydraulic buffer;
[0016] Display and control module: used to display the test results and classify the elevator hydraulic buffers according to the analysis results.
[0017] As a preferred technical solution of the present invention, the image acquisition module includes one or more high-resolution industrial cameras for acquiring images of the elevator hydraulic buffer from different angles.
[0018] As a preferred technical solution of the present invention, the image preprocessing module includes a denoising unit, an enhancement unit and a grayscale conversion unit, which are used to denoise, enhance and grayscale convert the image to improve the image quality.
[0019] As a preferred technical solution of the present invention, the image recognition module uses a convolutional neural network (CNN) to extract and recognize features of the preprocessed image, and the feature parameters include appearance defects, degree of wear, and deformation.
[0020] As a preferred technical solution of the present invention, the convolutional neural network (CNN) includes multiple convolutional layers, pooling layers and fully connected layers, which are used to extract multi-level features of the image, and identify the appearance defects, wear degree and deformation characteristics of the hydraulic buffer by comparing the differences between the pre-trained model and the actual detection image.
[0021] As a preferred technical solution of the present invention, the displacement detection module includes a linear variable differential transformer (LVDT) and a displacement sensor, which is used to accurately measure the displacement of the elevator hydraulic buffer in a working state.
[0022] As a preferred technical solution of the present invention, the data analysis module evaluates the performance of the elevator hydraulic buffer based on the identified characteristic parameters and displacement through the following formula:
[0023] ;
[0024] in, is the comprehensive performance index, and are the maximum and minimum deformation values detected, and are the maximum and minimum wear levels detected, and are the maximum and minimum appearance defect levels detected, is the displacement of the hydraulic buffer, , , and is the weight coefficient of each parameter, and the weight coefficient can be between 0.1 and 1.0 according to experimental data and specific application requirements.
[0025] As a preferred technical solution of the present invention, the data analysis module further calculates the buffer energy according to the detected parameters , the calculation formula is as follows:
[0026] ;
[0027] in, To buffer energy, is the elastic coefficient of the hydraulic buffer, is the detected displacement;
[0028] Among them, the data analysis module further calculates the wear rate based on the detected parameters , the calculation formula is as follows:
[0029] ;
[0030] in, is the wear rate, and are the maximum and minimum wear levels detected, respectively;
[0031] Among them, the data analysis module further calculates the deformation rate based on the detected parameters , the calculation formula is as follows:
[0032] ;
[0033] in, is the deformation rate, and are the maximum and minimum deformation values detected, respectively.
[0034] Wherein, the display and control module includes a display unit and a control unit, the display unit is used to display the detection results, and the control unit is used to classify the elevator hydraulic buffer according to the detection results and generate relevant report data.
[0035] As a preferred technical solution of the present invention, it also includes a database module for storing a large number of elevator hydraulic buffer images and corresponding detection results to improve the recognition accuracy and analysis capabilities of the system. The database module can also perform historical data analysis to predict the service life and maintenance requirements of the hydraulic buffer.
[0036] The present invention also provides a detection method of a hydraulic buffer detection system based on image recognition, comprising the following steps:
[0037] S1, collecting images of elevator hydraulic buffers;
[0038] S2, preprocessing the collected images, including denoising, enhancement and grayscale conversion;
[0039] S3, using a deep learning algorithm to recognize the preprocessed image and extract characteristic parameters of the elevator hydraulic buffer;
[0040] S4, monitor the displacement of the elevator hydraulic buffer and measure it in real time through a linear variable differential transformer (LVDT) and a displacement sensor;
[0041] S5, analyzing the identified characteristic parameters and displacement, and calculating the performance status of the hydraulic buffer by using a comprehensive performance index formula;
[0042] S6. Display the test results, classify the elevator hydraulic buffers according to the analysis results, and generate report data;
[0043] S7. Store the test results and image data in a database, and analyze the historical data to predict the service life and maintenance requirements of the hydraulic buffer.
[0044] Beneficial effects:
[0045] In summary, the above technical solution of the present invention has the following beneficial effects compared with the prior art:
[0046] 1. Improve detection efficiency: The elevator hydraulic buffer detection system based on image recognition and displacement sensing technology can automatically and quickly collect and process the image and displacement data of the hydraulic buffer. Through the combination of high-resolution industrial cameras and linear variable differential transformers (LVDT), the system can complete a comprehensive detection of the hydraulic buffer in a short time, significantly improving the detection efficiency and avoiding the time-consuming and labor-intensive problems in traditional manual and mechanical detection;
[0047] 2. Improve detection accuracy: The system uses advanced convolutional neural network (CNN) models for image recognition, which can accurately extract the characteristic parameters of the hydraulic buffer, such as appearance defects, wear degree and deformation. At the same time, a high-precision displacement sensor is used to monitor the displacement of the hydraulic buffer in real time to ensure the accuracy and reliability of the detection data. Through multi-angle and multi-position image acquisition and comprehensive performance index Q The system can comprehensively and accurately evaluate the performance of hydraulic buffers.
[0048] 3. Reduce inspection costs: The automated inspection system reduces reliance on manual inspection, reduces labor costs and operational risks. The system can complete a large number of hydraulic buffer inspection tasks in a relatively short period of time, significantly reducing inspection costs. In addition, the system adopts a modular design, with good compatibility and scalability between modules, low maintenance and upgrade costs, and helps to reduce overall operating costs in the long term;
[0049] 4. Enhanced data analysis capabilities: The system can not only display the test results in real time, but also has powerful data analysis and storage capabilities. Through the data analysis module, the system can conduct a comprehensive analysis of the identified characteristic parameters and displacements, and generate detailed test reports and maintenance recommendations. The database module stores a large amount of test data, supports historical data analysis and predictive maintenance, provides a scientific basis for the management and optimization of elevator hydraulic buffers, helps to extend the service life of hydraulic buffers, and improves the safety and reliability of elevators. BRIEF DESCRIPTION OF THE DRAWINGS
[0050] Figure 1This is a system block diagram of the hydraulic buffer detection system based on image recognition proposed by the present invention;
[0051] Figure 2 This is a flowchart of the detection method of the hydraulic buffer detection system based on image recognition proposed by the present invention. DETAILED DESCRIPTION
[0052] The following will be combined with 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.
[0053] Example 1: Basic detection system
[0054] See also Figure 1 and Figure 2 As shown, the image acquisition module:
[0055] The system uses two high-resolution industrial cameras, mounted on both sides of the hydraulic buffer, to capture images of the hydraulic buffer from different angles. The cameras are kept at a fixed distance from the hydraulic buffer and adjusted in position through a slide rail system to ensure full coverage of the hydraulic buffer surface.
[0056] Image preprocessing module:
[0057] Perform denoising on the collected images and use the median filter algorithm to remove the noise in the images;
[0058] Use contrast enhancement algorithms to improve image clarity;
[0059] Convert the image to grayscale for subsequent processing.
[0060] Image recognition module:
[0061] Use a pre-trained convolutional neural network (CNN) model to extract features from the pre-processed images;
[0062] Identify the appearance defects (such as cracks, dents), wear and deformation of hydraulic buffers;
[0063] Extract characteristic parameters including defect area, wear depth and deformation degree.
[0064] Displacement detection module: Uses a linear variable differential transformer (LVDT) to monitor the displacement of the hydraulic buffer in real time D ;
[0065] The displacement data is transmitted to the data analysis module for processing.
[0066] Data analysis module: Calculates comprehensive performance index based on identified characteristic parameters and displacement ,
[0067] ;
[0068] in, is the comprehensive performance index, and are the maximum and minimum deformation values detected, and are the maximum and minimum wear levels detected, and are the maximum and minimum appearance defect levels detected, is the displacement of the hydraulic buffer, , , and is the weight coefficient of each parameter, and the weight coefficient can be between 0.1 and 1.0 according to experimental data and specific application requirements.
[0069] Display and control module: Displays test results, including comprehensive performance index and various characteristic parameters;
[0070] According to the comprehensive performance index Classify the hydraulic buffers into qualified, repairable and scrapped ones;
[0071] Generate a test report and store it in the database.
[0072] Example 2: Advanced Detection System
[0073] See also Figure 1 and Figure 2 As shown, based on the basic detection system of Example 1, the detection accuracy and automation level are further improved.
[0074] Image acquisition module: Add three high-resolution industrial cameras, which are installed on the top, bottom and rear side of the hydraulic buffer;
[0075] A multi-angle and multi-position image acquisition solution is adopted to ensure all-round coverage of the hydraulic buffer.
[0076] Image preprocessing module: Use adaptive histogram equalization algorithm to further enhance image contrast;
[0077] Use edge detection algorithms to highlight crack and defect edges in images.
[0078] Image recognition module: Use an improved deep convolutional neural network (DCNN) model, increase the number of network layers, and improve the recognition accuracy of the model;
[0079] Increase recognition of three-dimensional deformation and extract deformation parameters.
[0080] Displacement detection module: add multi-point displacement sensors to monitor the displacement of the hydraulic buffer at different stress points in real time;
[0081] The multi-point displacement data are transmitted to the data analysis module for comprehensive analysis.
[0082] Data analysis module: Use multiple regression analysis method to combine multi-point displacement data and characteristic parameters to calculate the comprehensive performance index ,
[0083] ;
[0084] in, is the comprehensive performance index, and are the maximum and minimum deformation values detected, and are the maximum and minimum wear levels detected, and are the maximum and minimum appearance defect levels detected, is the displacement of the hydraulic buffer, , , and is the weight coefficient of each parameter, and the weight coefficient can be between 0.1 and 1.0 according to experimental data and specific application requirements.
[0085] Display and control module: displays detailed test results, including displacement of each stress point, characteristic parameters at multiple angles and comprehensive performance index;
[0086] Maintenance suggestions and optimization plans are automatically generated, stored in the database and pushed to maintenance personnel.
[0087] Example 3: Predictive Maintenance System
[0088] See also Figure 1 and Figure 2 As shown, based on the advanced detection system of Example 2, a predictive maintenance function is added.
[0089] Image acquisition module: Use a combination of laser scanner and high-resolution industrial camera to perform high-precision three-dimensional image acquisition.
[0090] Image preprocessing module: combines laser scanning data and camera images to generate high-precision 3D models;
[0091] De-noise and enhance 3D models.
[0092] Image recognition module: Uses a 3D convolutional neural network (3D-CNN) model to identify appearance defects, wear and deformation in 3D models;
[0093] Extract three-dimensional feature parameters.
[0094] Displacement detection module: uses high-precision laser displacement sensors to monitor the three-dimensional displacement of the hydraulic buffer at different force points in real time.
[0095] Data analysis module: Use machine learning algorithms and historical data to analyze the detected characteristic parameters and displacements and calculate the comprehensive performance index and Predictive Maintenance Index :
[0096] ;
[0097] in, and are the reference characteristic parameters and displacement, and is the weight coefficient.
[0098] Example 4: The above comprehensive performance index is analyzed by combining some examples. Specific calculation description:
[0099] We select the following key performance indicators to evaluate the performance of elevator hydraulic buffers: deformation Maximum and minimum deformation of the hydraulic buffer during use; degree of wear :Wear on the surface of hydraulic buffer; Appearance defects : The degree of defects in the appearance of the hydraulic buffer; displacement The displacement of the hydraulic buffer.
[0100] Normalization: In order to unify the performance indicators of different dimensions, we need to normalize each indicator. The normalization formula is as follows:
[0101] ;
[0102] in, Indicates deformation , Degree of wear Or appearance defects After normalization, the normalized values of deformation, wear degree and appearance defects are:
[0103] ;
[0104] ;
[0105] ;
[0106] Weighted average calculation: To obtain a comprehensive performance index , we need to perform weighted average on each normalized index. The specific formula is as follows:
[0107] .
[0108] in, 、 , γ and They are the weight coefficients of deformation, wear degree, appearance defect and displacement respectively.
[0109] The values of these weight coefficients range from 0.1 to 1.0 based on experimental data and specific application requirements.
[0110] Specific calculation steps: Data collection: Deformation from the elevator hydraulic buffer and , Degree of wear and , appearance defects and And the displacement .
[0111] Example calculation: Assume that the collected data is as follows: Normalization: Use the normalization formula to calculate 、 and , calculate the comprehensive performance index: use the weighted average formula to calculate the comprehensive performance index . Deformation: 、 ; Degree of wear: 、 ; Appearance defects: 、 ; Displacement: ; The normalized value is:
[0112] ;
[0113] ;
[0114] ;
[0115] Assume the weight coefficient is: , , , ;
[0116] Then, the comprehensive performance index for: That is, the above result of this embodiment reflects the comprehensive performance status of the elevator hydraulic buffer.
[0117] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.
[0118] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. A hydraulic buffer detection system based on image recognition, characterized in that: include: Image acquisition module: used to acquire images of elevator hydraulic buffers; Image preprocessing module: used to preprocess the collected images, including denoising, enhancement and grayscale conversion; Image recognition module: used to recognize the pre-processed image using a deep learning algorithm and extract the characteristic parameters of the elevator hydraulic buffer; The image recognition module uses a convolutional neural network to extract and recognize features of the preprocessed image, wherein the feature parameters include appearance defects, degree of wear, and deformation; Displacement detection module: used to monitor the displacement of the elevator hydraulic buffer in real time; the displacement detection module includes a linear variable differential transformer and a displacement sensor, which is used to accurately measure the displacement of the elevator hydraulic buffer in the working state; Data analysis module: used to comprehensively analyze the identified characteristic parameters and displacement to determine the performance status of the elevator hydraulic buffer; the data analysis module evaluates the performance of the elevator hydraulic buffer based on the identified characteristic parameters and displacement using the following formula: Among them, Q is the comprehensive performance index, D max and D min are the maximum and minimum deformation values detected, T max and T min are the maximum and minimum wear levels detected, W max and W min are the maximum and minimum appearance defect levels detected, respectively, D is the displacement of the hydraulic buffer, α, β, γ and δ are weight coefficients of each parameter, and the weight coefficients are between 0.1 and 1.0 according to experimental data and specific application requirements; The data analysis module further calculates the buffer energy E based on the detected parameters, and the calculation formula is as follows: Where E is the buffer energy, k is the elastic coefficient of the hydraulic buffer, and D is the detected displacement; The data analysis module further calculates the wear rate W based on the detected parameters. r , the calculation formula is as follows: Among them, W r is the wear rate, T max and T min are the maximum and minimum wear levels detected, respectively; The data analysis module further calculates the deformation rate D according to the detected parameters. r , the calculation formula is as follows: Among them, D r is the deformation rate, D max and D min are the maximum and minimum deformation values detected, respectively; Display and control module: used to display the test results; wherein, the display and control module includes a display unit and a control unit, the display unit is used to display the test results, and the control unit is used to classify the elevator hydraulic buffer according to the test results and generate relevant report data.
2. The hydraulic buffer detection system based on image recognition according to claim 1 is characterized in that: The image acquisition module includes one or more high-resolution industrial cameras for acquiring images of the elevator hydraulic buffer from different angles.
3. The hydraulic buffer detection system based on image recognition according to claim 1, characterized in that: The image preprocessing module comprises a denoising unit, an enhancing unit and a grayscale transforming unit, and is used for performing denoising, enhancing and grayscale transforming on the image to improve the image quality.
4. The hydraulic buffer detection system based on image recognition according to claim 1, characterized in that: The convolutional neural network includes multiple convolutional layers, pooling layers and fully connected layers, which are used to extract multi-level features of the image and identify the appearance defects, wear degree and deformation characteristics of the hydraulic buffer by comparing the differences between the pre-trained model and the actual detection image.
5. The hydraulic buffer detection system based on image recognition according to claim 1, characterized in that: It also includes a database module for storing a large number of elevator hydraulic buffer images and corresponding detection results to improve the recognition accuracy and analysis capabilities of the system. The database module also performs historical data analysis to predict the service life and maintenance requirements of the hydraulic buffer.
6. A detection method for a hydraulic buffer detection system based on image recognition, which is applicable to the hydraulic buffer detection system based on image recognition according to any one of claims 1 to 5, characterized in that: The following steps are involved: S1, collecting images of elevator hydraulic buffers; S2, preprocessing the collected images, including denoising, enhancement and grayscale conversion; S3, using a deep learning algorithm to recognize the preprocessed image and extract characteristic parameters of the elevator hydraulic buffer; S4, monitoring the displacement of the elevator hydraulic buffer, and performing real-time measurement through a linear variable differential transformer and a displacement sensor; S5, analyzing the identified characteristic parameters and displacement, and calculating the performance status of the hydraulic buffer by using a comprehensive performance index formula; S6. Display the test results, classify the elevator hydraulic buffers according to the analysis results, and generate report data; S7. Store the test results and image data in a database, and analyze the historical data to predict the service life and maintenance requirements of the hydraulic buffer.
