Multi-specification pressure instrument automatic verification system and intelligent scheduling method

Abstract

The application discloses a multi-specification pressure instrument automatic verification system and an intelligent scheduling method, belongs to the technical field of pressure measurement verification, and comprises a conveying and transferring module, a visual recognition module, a robot cooperation module, a pressure gauge automatic measurement module and a control module: based on material information of the pressure gauge transmitted by the visual recognition module and occupation conditions of pressure gauge detection stations transmitted by the pressure gauge automatic measurement module, a distributed constraint optimization algorithm is used to generate a dynamic optimal mapping distribution scheme of pressure gauges to be detected and the pressure gauge detection stations, the robot cooperation module is controlled to place the pressure gauges on corresponding pressure gauge detection stations according to the dynamic optimal mapping distribution scheme of the pressure gauges to be detected and the pressure gauge detection stations, and the robot cooperation module is controlled to clean, install and dismount and sort the pressure gauges to be detected; the system realizes continuous verification of multi-specification and large-batch pressure gauges, has high accuracy, strong anti-interference capability and optimal scheduling.

Description

Technical Field

[0001] This invention belongs to the field of pressure measurement and verification technology, and relates to an automatic verification system and intelligent scheduling method for multi-specification pressure instruments. Background Technology

[0002] Pressure gauges are ubiquitous in aerospace, oil and gas transportation, and water and gas supply systems, primarily used for safety protection and equipment operation monitoring. Prolonged use of pressure gauges causes wear and tear on their internal elastic materials and mechanical components, leading to decreased measurement accuracy, impacting industrial production, and even causing safety accidents. Therefore, national laws require pressure gauges used for safety protection to undergo periodic verification. Traditional verification methods rely on manual operation, which is inefficient and prone to subjective errors. In recent years, automated verification systems have gradually emerged, but most systems only automate the verification process; preliminary preparation and post-processing still require manual intervention, and their recognition models have poor generalization capabilities, failing to adapt to the verification needs of multiple specifications and models of pressure gauges. Some systems focus only on single technological breakthroughs, such as improving image recognition algorithms or integrating single hardware components, failing to address issues such as full-process automation, multi-specification compatibility, and anomaly handling at the system level. Low system integration and limited intelligence levels result in overall performance bottlenecks.

[0003] Existing automated inspection systems mostly employ robot loading, visual reading, single-station or same-specification inspection methods, which have significant drawbacks: 1) Only supports pressure gauges of a single specification / range, resulting in poor compatibility; 2) There is no optimization design or real-time intelligent scheduling solution based on historical data, resulting in low workstation utilization; 3) The entire process of warehousing, transportation, inspection, and archiving has not been closed-loop; 4) Existing patents do not include real-time intelligent dynamic scheduling during the testing process, making it impossible to achieve optimal workstation allocation for mixed testing of multiple specifications; 5) The automatic detection process does not include the necessary cleaning process for the pressure gauge connectors and other practical application steps, making it difficult to guarantee the stability and continuity of the overall system; 6) There is a lack of effective information exchange with the laboratory information management system, making it impossible to automatically generate original records and certificates. Summary of the Invention

[0004] To address the problems of poor compatibility, inefficient scheduling, insufficient robustness in identification, and lack of closed-loop processing in existing automated calibration systems, the present invention adopts the following technical solution: an automated calibration system for multi-specification pressure instruments, comprising: Conveying and transferring module: used to transport pressure gauges of different specifications that are laid flat to be tested; Visual recognition module: used to read and recognize the information from the pressure gauges transmitted by the conveying and transfer module; Robot collaboration module: used for grasping, cleaning connectors, installing and disassembling, and sorting pressure gauges to be tested conveyed by the transport module; Automatic pressure gauge metering module: used to monitor the occupancy status of N pressure gauge testing stations; The control module, based on the material information of multiple pressure gauges transmitted by the vision recognition module and the occupancy status of the pressure gauge testing stations transmitted by the automatic pressure gauge metering module, generates a dynamic optimal sequence of each pressure gauge to be tested and its testing station. It determines the grasping order and verification path, controls the robot collaboration module to grasp the pressure gauges according to the dynamic optimal sequence and mapping scheme, cleans the joints of the pressure gauges to be tested, places the pressure gauges on the corresponding testing station, and then controls the automatic pressure gauge metering module to run. Based on the completion signal, it sends action commands to the robot collaboration module to execute the pressure gauge disassembly and sorting operations. The process of generating the dynamic optimal sequence of each pressure gauge to be tested and the pressure gauge testing station, and determining the grabbing order and verification path is as follows: Construct a dictionary object to map the pressure gauge testing stations within the physical workshop, and assign physical constraint boundaries to each station; Based on constructing a dictionary object, connecting to the pressure gauge testing database, obtaining historical pressure gauge calibration records as training samples, and constructing a dynamic time prediction model for pressure gauge testing; Establish an independent daemon thread to monitor the status and collect data at the pressure gauge detection station; Based on the status monitoring and data of the pressure gauge testing station and the time consumption prediction during the pressure gauge testing process, the matching relationship between the pressure gauge testing task and the pressure gauge testing station is established, constraints are set, and multi-dimensional constraint solving is performed to obtain the pressure gauge grabbing order and verification path.

[0005] Furthermore, the automatic pressure gauge metering module includes: Testing station submodule: includes N pressure gauge testing stations, used to place the pressure gauges to be tested; Pressure Standard Submodule: Set up at the pressure gauge testing station in the testing station submodule as a standard instrument for pressure gauge testing, used to calibrate the pressure gauge to be tested; Pressure gauge tapping test submodule: Set up on the pressure gauge test station in the test station submodule, for performing tapping tests on the pressure gauges under test as required by the procedure. Visual recognition monitoring submodule: used to identify pointer pressure gauge readings in real time during calibration, convert analog quantities into digital quantities, and photograph and store all detection information. It can also monitor the occupancy of the pressure gauge detection station in the detection station submodule in real time. Pressure controller: Used to control the pressure of the pressure gauge under test; Automatic pressure gauge calibration system: This system controls the automatic calibration of pressure gauges connected to the pressure gauge testing station, outputs the required pressure for the pressure gauge to be tested, controls the pressure standard submodule to calibrate the pressure gauge to be tested, and controls the pressure gauge tapping test submodule to perform tapping tests on the pressure gauge. Based on the visual recognition monitoring submodule, the system outputs the digital data of the pressure gauge test during the pressure gauge testing process, forming a synchronously stored file containing images, data, and timestamps.

[0006] Furthermore, the process for determining the range of the pressure standard corresponding to the pressure gauge testing station is as follows: Acquire historical data of pressure gauge testing over multiple cycles, including pressure gauge range, accuracy class, and the number of historical pressure gauge tests within each range. The proportion of pressure gauges in each range is calculated, the maximum permissible error of pressure gauges in each range is calculated, and then the requirements for standard instruments in JJG52-2013 Verification Procedure for Elastic Element Type General Pressure Gauges, Pressure Vacuum Gauges and Vacuum Gauges are combined with the requirement that the absolute value of the maximum permissible error of the standard instrument should not be greater than 1 / 4 of the absolute value of the maximum permissible error of the pressure gauge under test, to obtain the optimal pressure standard instrument combination scheme for each pressure gauge testing station.

[0007] Furthermore, the optimal pressure standard combination scheme includes: For the first range segment where the proportion of pressure gauges being tested is greater than the preset threshold, at least two pressure gauge testing stations should be configured. For the second range where the proportion of pressure gauges being measured is less than the preset threshold, at least one pressure gauge testing station shall be configured. Furthermore, for the K ranges that rank first in terms of quantity in historical testing data, all pressure gauge testing stations are configured to meet their testing requirements.

[0008] Furthermore, the independent daemon thread includes: The device fault guardian thread is used to send read holding register requests to the PLC controller at fixed intervals to monitor hardware fault codes. The underlying feedback listening thread is used to continuously traverse the workstation configuration dictionary object, filter out workstations with the status of being occupied, and read the level of their completion feedback coil. The external interface polling thread is used to initiate polling requests to the LIMS system via the HTTP protocol at fixed intervals to obtain the pressure gauge tasks to be inspected.

[0009] Furthermore, the constraints include those for establishing a matching relationship between pressure gauge testing tasks and pressure gauge testing stations, comprising the following three dimensions: Pressure constraint: The range attribute of the pressure gauge under test belongs to the closed interval formed by the minimum pressure threshold and the maximum pressure threshold of the workstation. Mechanical constraints: The connector type of the pressure gauge to be tested is available in the compatible connector specification list at the workstation; State constraint: The current lifecycle state of the workstation is idle; The execution priority order of the constraint solving is as follows: First priority: workstation status constraint, prioritize workstations in the waiting state to execute unloading and sorting tasks; Second priority: Pressure constraints and mechanical constraints are checked simultaneously when selecting idle workstations to load new tasks.

[0010] Furthermore, the conveying and transfer module includes: Standardized tray: Used to support and flatten multiple pressure gauges of different ranges to be tested; The trolley module is used to receive multiple standardized pallets and perform long-distance transfers within a predetermined area; the trolley module uses multiple AGV trolleys. Conveying mechanism: Used to sequentially place multiple trays to the detection area after the trolley module arrives at the designated position.

[0011] Furthermore, it also includes a pre-processing module for mechanical cleaning and debris recovery of the pressure gauge connector threads and sealing surfaces; The pre-processing module before verification includes: Rotary brush head: Used to clean the sealing surface of the pressure gauge connector. Drive unit: used to provide power for rotating the rotating brush head; Negative pressure suction module: used to collect and clean the waste generated during the cleaning process of the rotating brush head.

[0012] An intelligent scheduling method for an automatic calibration system for multi-specification pressure instruments includes the following steps: The pressure gauges to be tested are carried by AGV carts in a gauge-pallet mode. The AGV carts carry multiple pallets and use laser navigation and ground positioning landmarks to deliver each pallet to the designated position on the shelf. The multiple pallets are then transferred to the testing area in sequence. The visual recognition module scans and identifies the unique identifiers of all pressure gauges in the tray within the detection area, obtains information on the pressure gauge size, range, accuracy class, connector type, and dial diameter, and transmits it to the control module. Based on the real-time identified pressure gauge information, the occupancy status of each pressure gauge testing station, and the range matching degree of the pressure controller, a distributed constraint optimization algorithm is used to generate the dynamic optimal sequence and mapping allocation scheme between each pressure gauge to be tested and the pressure gauge testing station, thereby completing the dynamic optimal mapping between the pressure gauge and the pressure gauge testing station and determining the grabbing order and verification path. The robot collaboration module, following the scheduling instructions and intelligent scheduling sequence, picks up pressure gauges from the tray to clean the connectors and uses a rotating brush head to mechanically clean the threads. After cleaning, the robot collaboration module moves the pressure gauge to the mapped and assigned pressure gauge testing station, achieving precise docking between the connector and the calibration interface, automatically completing locking and sealing, and performing airtightness pre-testing; The automatic pressure gauge metering module automatically executes the test process according to the JJG52-2013 verification procedure; it determines whether the test is qualified or unqualified according to the procedure, and simultaneously stores the verification results, process images, timestamps, and equipment numbers in the local database and cloud platform to achieve full traceability. The verification data is synchronized to the laboratory information management system in real time via an interface, automatically generating original records, verification results and certificates; The robot collaboration module removes the inspected pressure gauges from the pressure gauge testing station and automatically sorts them according to the pass or fail results; the AGV carts transfer the inspected pallets to the inspected area shelves, complete the warehousing and archiving, and realize continuous and uninterrupted automated inspection.

[0013] This invention provides an automatic verification system and intelligent scheduling method for multi-specification pressure gauges, which solves the problems of poor compatibility, inefficient scheduling, insufficient identification robustness, and lack of closed-loop in the entire process of existing automatic verification systems. It can realize continuous unattended verification of multi-specification and large-batch pressure gauges, with high identification accuracy, strong anti-interference ability, optimal scheduling, and traceable data. It is suitable for automated metrological verification and calibration of pressure gauges in metrology institutions and industrial enterprises. Based on multi-cycle historical data, the system optimizes the configuration of metrological standard modules and implements a real-time dynamic optimization scheduling method during the testing process. This enables the system to flexibly adapt to the verification needs of different batches and multiple specifications of instruments, maximize resource utilization, and match the optimal design scheme. The collaborative mode between the transport and transfer module and the robot cooperation module, along with the intelligent scheduling scheme for multiple workstations, is dynamically optimized to maximize system throughput and improve testing efficiency.

[0014] An additional pre-processing module is added to automatically clean the pressure gauge connectors before automatic verification begins. This ensures connector cleanliness, improves sealing success rate and verification data reliability, prevents impurities from entering the quick-connect mechanism and piping system, reduces the risk of blockage, wear, and leakage, and lowers the frequency of equipment downtime for maintenance. This significantly reduces the failure rate of automated verification, enhancing the system's reliability and practical value.

[0015] The entire process is unmanned and closed-loop: from the instrument entering the automatic testing laboratory to the automatic execution of the verification process, and then to the outbound management after the verification is completed, the whole process achieves a high degree of automation without human intervention. Multi-specification instrument strong compatibility design: The system can flexibly support mixed detection of various instruments with different diameters, multiple measurement ranges and from different manufacturers, with wide adaptability; Multi-station parallel synchronous verification capability: By configuring multiple independent pressure gauge testing stations, the system can realize parallel metrological verification and calibration testing of multiple instruments without interference, which greatly improves the overall verification efficiency. Data closed-loop integration with the laboratory information management system: All data in the verification process is fully recorded and stored, forming a traceable electronic record, which fully meets the strict requirements of data traceability and reliability in the metrology management system.

[0016] Intelligent scheduling: The visual recognition module scans the unique identifier of the pressure gauge in the laboratory information management system to obtain the preset information of all the instruments under inspection in the current tray in the laboratory information management system, and sends it to the control module. The control module will generate the optimal inspection sequence based on the information of all the pressure gauges under inspection in this group, as well as the occupancy status of each pressure gauge inspection station, the range matching degree of the pressure controller, and the estimated inspection time, etc., through a distributed constraint optimization algorithm. It will optimally sort all the pressure gauges in the tray and complete the dynamic optimal mapping between each pressure gauge and the corresponding pressure gauge inspection station, and determine the grabbing order and inspection path. Joint cleaning: This addresses the root cause of problems such as poor sealing, leakage, and pressure failure caused by impurities. Multi-specification compatibility: Combination of multi-range pressure modules and multiple quick connectors supports mixed testing of instruments with different diameters, ranges, dial sizes, and manufacturers. Full-process traceability: The entire process of detection, including images, data, timestamps, and equipment information, is recorded to meet metrological management requirements.

[0017] The present invention provides an automatic calibration system and intelligent method for multi-specification pressure gauges, which has the following beneficial effects: The entire process is automated, reducing manual intervention by more than 90%. It can perform parallel testing of pressure gauges at multiple workstations, which is highly efficient and can be continuously calibrated without human intervention 24 hours a day. It supports mixed inspection of multiple specifications (multiple sizes, multiple ranges), has strong versatility, and is widely applicable; It includes optimized layout design based on historical big data, combined with real-time dynamic intelligent scheduling, resulting in a significant improvement in capacity and utilization. Including practical application steps such as the necessary cleaning process for pressure gauge connectors, the overall system has good stability and continuity and strong environmental adaptability; It enables efficient information interaction with the laboratory information management system, ensuring complete and traceable data. Images, data, timestamps, and equipment information are all recorded throughout the process, and original records and certificates can be automatically generated to meet the requirements of the metrology management system. Modular structure, easy to deploy, maintain and expand. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a diagram showing the module composition of a pressure gauge calibration system; Figure 2 This is a diagram of the communication and multi-threaded architecture of the control module of the verification system; Figure 3 This is a flowchart of the testing process. Detailed Implementation

[0020] It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of the present invention can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and embodiments.

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the present invention or its application or use. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Figure 1 This is a diagram showing the module composition of a pressure gauge calibration system; An automatic calibration system for multi-specification pressure gauges, comprising: Conveying and transferring module: used to transport pressure gauges of different specifications that are laid flat to be tested; Visual recognition module: used to read and recognize the information from the pressure gauges transmitted by the conveying and transfer module; Robot collaboration module: used for grasping, cleaning connectors, installing and disassembling, and sorting pressure gauges to be tested conveyed by the transport module; Automatic pressure gauge measurement module: It is used to monitor the occupancy of N pressure gauge testing stations, and to perform metrological performance testing on the pressure gauges placed at the pressure gauge testing stations according to the settings, and to take pictures and save the test results. It is also used to synchronously assign the metrological performance test data during the testing process to the corresponding pressure gauge calibration template in the laboratory information management system, generate the pressure gauge calibration original record required in the calibration procedure, and then synchronously map and generate the corresponding calibration certificate. Control Module: Based on the material information of multiple pressure gauges transmitted by the vision recognition module and the occupancy status of the pressure gauge testing stations transmitted by the automatic pressure gauge metering module, the module generates a dynamic optimal sequence of each pressure gauge to be tested and its testing station, determines the grasping order and verification path, and controls the robot collaboration module to grasp the pressure gauges according to the dynamic optimal sequence and mapping allocation scheme. The module also cleans the joints of the pressure gauges to be tested and places them on the corresponding testing station. Then, the automatic pressure gauge metering module is controlled to run, and after all testing items are completed, a testing completion signal is issued. Finally, an action command is sent to the robot collaboration module to execute the pressure gauge disassembly and sorting operations.

[0023] Furthermore, the automatic pressure gauge metering module includes: The testing station submodule includes N pressure gauge testing stations for placing pressure gauges to be tested; N is a positive integer. Pressure Standard Submodule: Set up at the pressure gauge testing station in the testing station submodule as a standard instrument for pressure gauge testing, used to calibrate the pressure gauge to be tested; Pressure gauge tapping test submodule: Set up on the pressure gauge test station in the test station submodule, for performing tapping tests on the pressure gauges under test as required by the procedure. Visual recognition monitoring submodule: used to identify pointer pressure gauge readings in real time during calibration, convert analog quantities into digital quantities, and photograph and store all detection information. It can also monitor the occupancy of the pressure gauge detection station in the detection station submodule in real time. Pressure controller: Used to control the pressure of the pressure gauge under test; Automatic pressure gauge calibration system: This system controls the automatic calibration of pressure gauges connected to the pressure gauge testing station, outputs the required pressure for the pressure gauge to be tested, controls the pressure standard submodule to calibrate the pressure gauge to be tested, and controls the pressure gauge tapping test submodule to perform tapping tests on the pressure gauge. Based on the visual recognition monitoring submodule, the system outputs the digital data of the pressure gauge test during the pressure gauge testing process, forming a synchronously stored file containing images, data, and timestamps.

[0024] Furthermore, the process for determining the range of the pressure standard corresponding to the pressure gauge testing station is as follows: Obtain historical data on pressure gauge testing over multiple periods, including pressure gauge range, accuracy class, and the number of historical pressure gauge tests within each range. The proportion of pressure gauges in each range is calculated, the maximum permissible error of pressure gauges in each range is calculated, and then the requirements for standard instruments in JJG52-2013 Verification Procedure for Elastic Element Type General Pressure Gauges, Pressure Vacuum Gauges and Vacuum Gauges are combined with the requirement that the absolute value of the maximum permissible error of the standard instrument should not be greater than 1 / 4 of the absolute value of the maximum permissible error of the pressure gauge under test, to obtain the optimal pressure standard instrument combination scheme for each testing station.

[0025] Furthermore: the optimal pressure standard combination scheme includes: For the first range segment where the proportion of pressure gauges being tested is greater than the preset threshold, at least two pressure gauge testing stations should be configured. For the second range where the proportion of pressure gauges being measured is less than the preset threshold, at least one pressure gauge testing station shall be configured. Furthermore, for the top K pressure gauge ranges in terms of quantity in historical testing data, all pressure gauge testing stations are configured to meet their testing requirements. Based on the proportion of testing data, different stations are assigned appropriate ranges to maximize the testing compatibility of various pressure gauge specifications and avoid wasting station resources.

[0026] Furthermore: A dynamic optimal sequence is generated for each pressure gauge to be tested and its corresponding testing station. The grabbing order and verification path are determined. This is achieved through a distributed constraint optimization algorithm, which optimally sorts the grabbing of all pressure gauges in the tray and simultaneously completes the dynamic optimal mapping between each pressure gauge and its corresponding testing station, determining the grabbing order and verification path. The specific process is as follows: Construct a dictionary object to map pressure gauge testing stations within the physical workshop, and assign physical constraint boundaries to each station, including the minimum pressure threshold, the maximum pressure threshold, the list of compatible connector specifications, and the initial lifecycle status. Based on the construction of a dictionary object, a pressure gauge detection database is connected, and historical calibration records of pressure gauges are taken as training samples. The feature dimensions include the coordinates of the robotic arm's gripping point, the end position of the pressure gauge, the ambient air pressure, and the ambient light. A dynamic time prediction model for pressure gauge detection is constructed for predicting the time consumption during the pressure gauge detection process. Establish an independent daemon thread to monitor the status and collect data at the pressure gauge detection station; Based on the status monitoring and data of the pressure gauge testing station and the time consumption prediction during the pressure gauge testing process, the matching relationship between the pressure gauge testing task and the pressure gauge testing station is established, constraints are set, and multi-dimensional constraint solving is performed to obtain the pressure gauge grabbing order and verification path.

[0027] Furthermore: the independent daemon thread includes: The device anomaly guardian thread is used to send read holding register requests to the PLC controller at fixed intervals to monitor hardware fault codes. The underlying feedback listening thread is used to continuously traverse the workstation configuration dictionary object, filter out workstations with the status of being occupied, and read the level of their completion feedback coil. The external interface polling thread is used to initiate polling requests to the LIMS system via the HTTP protocol at fixed intervals to obtain the pressure gauge tasks to be inspected.

[0028] Furthermore, the constraints include those for establishing a matching relationship between pressure gauge testing tasks and pressure gauge testing stations, encompassing the following three dimensions: Pressure constraint: The range attribute of the pressure gauge under test belongs to the closed interval formed by the minimum pressure threshold and the maximum pressure threshold of the workstation. Mechanical constraints: The connector type of the pressure gauge to be tested is available in the compatible connector specification list at the workstation; State constraint: The current lifecycle state of the workstation is idle; The execution priority order of the constraint solving is as follows: First priority: workstation status constraint, prioritize workstations in the waiting state to execute unloading and sorting tasks; Second priority: Pressure constraints and mechanical constraints are checked simultaneously when selecting idle workstations to load new tasks.

[0029] Furthermore: the conveying and transfer module includes: Standardized tray: Used to support and flatten multiple pressure gauges of different ranges to be tested; The trolley module is used to receive multiple standardized pallets and perform long-distance transfers within a predetermined area; the trolley module uses multiple AGV trolleys. Conveying mechanism: Used to sequentially place multiple trays to the detection area after the trolley module arrives at the designated position.

[0030] Furthermore, the system also includes a pre-processing module for mechanical cleaning and debris recovery of the pressure gauge connector threads and sealing surfaces; The pre-processing module before verification includes: Rotary brush head: Used to clean the sealing surface of the pressure gauge connector. Drive unit: used to provide power for rotating the rotating brush head; Negative pressure suction module: used to collect and clean the waste generated during the cleaning process of the rotating brush head.

[0031] An intelligent scheduling method for an automatic calibration system for multi-specification pressure instruments includes the following steps: The pressure gauges to be tested are carried by AGV carts in a gauge-pallet mode. The AGV carts carry multiple pallets and use laser navigation and ground positioning landmarks to deliver each pallet to the designated position on the shelf. The multiple pallets are then transferred to the testing area in sequence. The visual recognition module scans and identifies the unique identifiers of all pressure gauges in the tray within the detection area, obtains information on the pressure gauge size, range, accuracy class, connector type, and dial diameter, and transmits it to the control module. Based on the real-time identified pressure gauge information, the occupancy status of each pressure gauge testing station, and the range matching degree of the pressure controller, a distributed constraint optimization algorithm is used to generate the dynamic optimal sequence and mapping allocation scheme between each pressure gauge to be tested and the pressure gauge testing station, thereby completing the dynamic optimal mapping between the pressure gauge and the pressure gauge testing station and determining the grabbing order and verification path. The robot collaboration module, following the scheduling instructions and intelligent scheduling sequence, picks up pressure gauges from the tray to clean the connectors and uses a rotating brush head to mechanically clean the threads. After cleaning, the robot collaboration module moves the pressure gauge to the mapped and assigned pressure gauge testing station, achieving precise docking between the connector and the calibration interface, automatically completing locking and sealing, and performing airtightness pre-testing; The automatic pressure gauge metering module automatically executes the test process according to the JJG52-2013 verification procedure; it determines whether the test is qualified or unqualified according to the procedure, and simultaneously stores the verification results, process images, timestamps, and equipment numbers in the local database and cloud platform to achieve full traceability. The verification data is synchronized to the laboratory information management system in real time via an interface, automatically generating original records, verification results and certificates; The robot collaboration module removes the inspected pressure gauges from the pressure gauge testing station and automatically sorts them according to the pass or fail results; the AGV carts transfer the inspected pallets to the inspected area shelves, complete the warehousing and archiving, and realize continuous and uninterrupted automated inspection.

[0032] Example 1: An automatic calibration system for multi-specification pressure gauges, comprising: Conveying and transferring module: used to transport multiple pressure gauges of different specifications that are laid flat to be tested; realizes palletized transportation, batch long-distance transfer of AGV trolleys, and automatic transmission to the testing area; The conveying and transfer module includes: Standardized pallets: used to support and hold multiple pressure gauges of various specifications to be tested, placed flat on the surface. The trolley module is used to receive multiple standardized pallets and transport them over long distances within a predetermined area; the trolley module uses multiple AGV trolleys (automatic guided vehicles). Conveying Mechanism: When the AGV arrives at the designated position, it sequentially releases the transfer pallet to the transmission mechanism, which then transports it to the detection area. The conveying mechanism includes a conveyor belt. Visual Recognition Module: The visual recognition module is positioned above the pressure gauges to be tested. Through global scanning, it identifies the information of all pressure gauges in the tray conveyed to the testing area by the transport module. The visual recognition module obtains information on all instruments under test in the current tray by scanning the unique identifier of the pressure gauge (before the pressure gauge is sent to the automated testing laboratory, the pressure gauge has been assigned a unique identifier in the laboratory information management system, which includes information such as the pressure range, commissioning unit, serial number, manufacturer, pressure gauge diameter, and connectors). This information is then sent to the control module. The visual recognition module uses a visual camera. Robot collaboration module: used for grasping, cleaning connectors, installing and disassembling, and sorting pressure gauges to be tested; the robot collaboration module: adopts a six-axis collaborative robot, equipped with a vision positioning unit and a force control unit, to complete the automatic grasping, positioning, installation, disassembly, and sorting of pressure gauges; The automatic pressure gauge measurement module monitors the occupancy of the three pressure gauge testing stations and performs metrological performance tests on the pressure gauges placed at these stations according to predefined settings. It covers instruments with multiple ranges, dial diameters, and connector types, meeting the needs of mixed testing of various specifications. The module also photographs and stores the test results, synchronizes the metrological performance test data during the testing process, assigns it to the corresponding pressure gauge calibration template in the laboratory information management system, generates the original pressure gauge calibration record required by the calibration procedure, and then synchronously maps it to generate the corresponding calibration certificate based on the certificate generation module. The certificate is then issued after review by the calibration personnel.

[0033] Control Module: Based on the physical information of multiple pressure gauges transmitted by the vision recognition module, and the occupancy status of the pressure gauge testing stations, pressure controller range matching degree, and estimated testing time transmitted by the automatic pressure gauge metering module, the control module generates a dynamic optimal sequence and mapping allocation scheme for each pressure gauge to be tested and its testing stations through a distributed constraint optimization algorithm. The control module then controls the robot collaboration module to grasp the pressure gauges according to this dynamic optimal sequence and mapping allocation scheme, cleans the joints of the pressure gauges to be tested, and places them on the corresponding testing stations. The automatic pressure gauge metering module then runs and, after completing all testing items, issues a testing completion signal. Finally, the control module sends an action command to the robot collaboration module to execute pressure gauge disassembly and sorting operations. The visual recognition module includes a high-definition industrial camera and an adaptive light source, used for global scanning of the tray and identification of the unique identifier on the pressure gauge (this barcode is already included in the laboratory information management system, containing the specifications, range, manufacturer, connector type, and accuracy of the pressure gauge being inspected). The automatic pressure gauge module includes: The testing station submodule includes three pressure gauge testing stations for placing pressure gauges to be tested: a first pressure gauge testing station, a second pressure gauge testing station, and a third pressure gauge testing station. Each of these stations has its own independent pressure interface and locking mechanism, and also has a corresponding independent pressure gauge tapping test submodule, pressure standard submodule, pressure controller, and visual recognition monitoring submodule, supporting simultaneous parallel calibration of multiple instruments. The station comprises a first pressure gauge testing station, a second pressure gauge testing station, and a third pressure gauge testing station. The first pressure gauge testing station can automatically calibrate pressure gauges with a measuring range of 0.025MPa to 4MPa and an accuracy class of 1.0 or lower. The second pressure gauge testing station can automatically calibrate pressure gauges with a measuring range of 0.16MPa to 6MPa and an accuracy class of 1.0 or lower. The third pressure gauge testing station can automatically calibrate pressure gauges with a measuring range of 0.6MPa to 10MPa and an accuracy class of 1.0 or lower. Pressure Standard Submodule: Located at the pressure gauge testing station in the testing station submodule, it serves as a standard instrument for pressure gauge testing and is used to calibrate the pressure gauge to be tested; Pressure gauge tapping test submodule: Set up on the pressure gauge test station in the test station submodule, for performing tapping tests on the pressure gauges under test as required by the procedure. The visual recognition and monitoring submodule is used to identify the readings of the pointer pressure gauge in real time during calibration, convert analog quantities into digital quantities, and photograph and store all detection information. It can also monitor the occupancy of the pressure gauge detection station in the detection station submodule in real time. The visual recognition and monitoring submodule uses a visual camera. Pressure controller: Used to control the pressure of the pressure gauge under test; Automatic pressure gauge calibration system: This system controls the pressure controller of pressure gauges already connected at the pressure gauge testing station, outputting the required pressure for the pressure gauge to be tested. It controls the pressure standard submodule to calibrate the pressure gauge and the pressure gauge tapping test submodule to perform tapping tests on the pressure gauge. Based on the visual recognition monitoring submodule, the system outputs digital signals of the pressure gauge test data during the testing process, assigning them to the corresponding pressure gauge calibration template in the laboratory information management system. This generates the original calibration record of the pressure gauge required in the calibration procedure, and then synchronously maps and generates the corresponding calibration certificate. After review by the calibration personnel, the certificate is released and a file containing images, data, and timestamps is synchronously stored.

[0034] Before being sent to the automated testing laboratory, each pressure gauge is assigned a unique identifier (synchronized with the laboratory information management system, containing information such as the instrument name, range, accuracy class, manufacturer, scale division, and connector type of the instrument under test). The visual recognition module scans the unique identifier of the pressure gauge to obtain information on all the instruments under test in the current tray and sends it to the control module. The control module will generate the optimal testing sequence through a distributed constraint optimization algorithm based on the real-time identified instrument information, as well as the received occupancy status of each pressure gauge testing station, the range matching degree of the pressure controller, and the estimated testing time. This optimal sequence will be sorted among all the pressure gauges in the tray, and the dynamic optimal mapping between each pressure gauge and its corresponding testing station will be completed, determining the grabbing order and the verification path.

[0035] The dynamic optimal mapping and allocation scheme for generating the pressure gauges to be tested and the pressure gauge testing stations is achieved through a distributed constraint optimization algorithm. This algorithm optimally sorts all the pressure gauges in the tray and simultaneously completes the dynamic optimal mapping between each pressure gauge and its corresponding testing station, determining the grabbing order and the verification path. The specific process is as follows: S11: System initialization and hardware / software resource mapping logic During the instantiation phase (i.e., in the init constructor), the system first performs the binding of the hardware and software communication protocols and the initialization of the state machine. The system instantiates a ModbusTcpClient object and binds it to the factory's preset IP address and port number. If the connection fails, a fatal error is thrown, terminating the system startup. The system constructs a dictionary object named STATION_CONFIG in memory to map the three test stations within the physical workshop. Each station is assigned strict physical constraints, including the minimum pressure threshold (min_p) for the pressure gauge testing station, the maximum pressure threshold (max_p) for the pressure gauge testing station, a list of compatible connector specifications (sockets, including 'M20' and 'M14') for the pressure gauge testing station, and an initial lifecycle state (state set to 'IDLE'). S12: Machine learning model building logic based on real data; Before taking over physical equipment, the system calls the _init_ml_model module to complete the training of the dynamic time consumption prediction model; The system establishes a connection with the factory's historical database through pymysql and executes SQL statements to extract the most recent 1000 historical verification records; The extracted feature dimensions include the coordinates of the robotic arm's gripping point (start_pos), the end position of the pressure gauge (end_pos), the ambient air pressure (env_press), and the ambient light (env_light). Using actual time consumption (actual_time) as the label, a training set feature matrix train_X and a label array train_y are constructed, respectively. To ensure high system availability, this process is encapsulated in a try-except exception handling block. If a database connection timeout occurs or fewer than 10 historical data records are recorded, the system will automatically trigger the exception handling mechanism, injecting preset factory default feature vectors into train_X and train_y. Subsequently, the system instantiates and fits a Random Forest Regressor containing 100 decision trees for subsequent real-time time consumption prediction. S13 performs asynchronous concurrent listening and data buffering logic; The system's main program launches multiple independent daemon threads to achieve non-blocking status monitoring and data acquisition. Each thread runs in an infinite loop controlled by the `self.is_running` flag. The device fault daemon logic (thread_device_daemon) sends a request to the PLC to read the holding register (address 2000) every 1 second. If the read fault code is greater than 0, a hardware alarm is detected. This read process is strictly encapsulated in an exception handling block. When encountering electromagnetic interference common in industrial environments that causes connection loss, the system executes a pass to silently ignore the issue, preventing thread crashes.

[0036] The underlying feedback listening logic (thread_plc_listener) continuously iterates through STATION_CONFIG, filtering out workstations whose current state is 'BUSY'. For each selected workstation, the system calculates the corresponding completion feedback coil address based on the base address (1000) plus the workstation number (sid), and reads its voltage level. If the reading is high (True), it indicates that the physical action is complete, and the system immediately updates the workstation's state to 'WAIT_UNLOAD' and writes a low level (False) to the coil to complete the signal reset.

[0037] External interface polling logic (thread_vision_lims): This thread sends polling requests with a timeout limit (timeout=3) to the LIMS system's API interface via HTTP every 2 seconds. If the response status code is 200, the system will parse the returned JSON data stream and push the obtained instrument objects to be inspected one by one into the global thread-safe queue (vision_to_scheduler_queue).

[0038] S14 Production Scheduling Control and Constraint Solving Logic The core control flow of the system is executed by the `thread_task_execution` thread. Within each scheduling cycle, the system first transfers all data from the global queue to the thread's internal queue to avoid multi-threaded lock contention, and then performs strict priority checks. First priority: workstation status constraint, prioritize workstations in the waiting state to execute unloading and sorting tasks; The system first uses list comprehension to retrieve all candidate workstations with a status of 'WAIT_UNLOAD'.

[0039] If the candidate set is not empty, the system extracts the first candidate workstation as the target workstation (target_site).

[0040] The system writes the target workstation number to the PLC target address register and writes a high level to the unload trigger coil.

[0041] The system calls the wait_for_robot blocking function (this function polls the completion coil at address 1100 at a high frequency until a high level is detected, then resets and returns, ensuring that the physical action is absolutely completed).

[0042] After the action is confirmed, the system resets the state of the target workstation to 'IDLE', updates the current robotic arm position arm_pos to 0, sets the flag action_taken to True, interrupts the current loop using the continue statement, and directly enters the next scheduling cycle.

[0043] Second priority: Pressure constraints and mechanical constraints are checked simultaneously when selecting idle workstations to load new tasks.

[0044] If there is no unloading task and the internal queue is not empty, the system executes nested traversal logic: the outer layer traverses the queue of instruments to be inspected, and the inner layer traverses STATION_CONFIG to find the workstation with the status 'IDLE'.

[0045] Constraint Solving: In the inner traversal, the system performs strict Boolean logic judgments—the range attribute of the instrument under test should be greater than or equal to min_p and less than or equal to max_p, and its connector attribute (dia) must exist in the sockets list of the workstation.

[0046] Target locking: Once the judgment is successful, the system immediately records the target workstation number (target_site) and the matching gauge index (matched_gauge_idx), and then exits the inner and outer loops in sequence using break statements.

[0047] Action sequence issuance: The system uses the `pop` method to remove the instrument from the buffer queue and maps its header attributes to underlying machine code (M14 maps to 14, M20 maps to 20). Subsequently, the system executes a continuous stream of cooperative instructions: Write a high level to the coil to trigger pre-cleaning, and call wait_for_robot to wait for loop closure; write the target workstation number and connector machine code to the register, write a high level to the coil to trigger loading, and call wait_for_robot again to wait for loop closure; after loading confirmation, write a high level to the start coil (300 + target workstation number) of the corresponding test socket to start the pressure test.

[0048] Status Update: The system calls the aforementioned pressure gauge detection dynamic timeout to predict the timeout, updates the target pressure gauge detection station status to 'BUSY', and updates the robotic arm position arm_pos to the target station number, then enters the next scheduling cycle. If no scheduling action is performed in the current cycle, the system triggers a 0.5-second sleep to reduce CPU power consumption.

[0049] This method also includes disaster recovery intervention and secure exit logic; During the main thread's daemonization process, the system is maintained by an infinite loop wrapped in a try-except structure. When an operator presses the stop button due to an emergency (triggering a KeyboardInterrupt exception), the system immediately sets the global lifecycle flag `self.is_running` to `False`. This assignment is broadcast to all child threads in the infinite loop (while `self.is_running`), allowing them to exit naturally after completing their current atomic operation. Subsequently, the main thread calls `self.plc.close()` to proactively release the Modbus TCP port connection, thereby safely severing underlying hardware communication and shutting down the system.

[0050] Furthermore: The robot collaboration module is equipped with a pre-processing module for mechanical cleaning of the pressure gauge connector threads and sealing surfaces and waste recycling. The pre-processing module before verification includes: a replaceable rotating brush head, a brush head drive unit, and a negative pressure suction module; Rotary brush head: Used to clean the sealing surface of the pressure gauge connector. Drive unit: used to provide power for rotating the rotating brush head; Negative pressure suction module: used to recycle waste generated during the cleaning process of the rotating brush head.

[0051] The rotating brush head is replaceable: it adopts a detachable structure and is equipped with a special brush head (with moderate hardness, which will not damage the thread profile and the smoothness of the sealing surface) to clean the sealing surface of the joint end face and remove surface oil, oxide scale and small debris. Brush head drive unit: driven by a stepper motor with adjustable speed. The speed can be preset according to the cleaning requirements of the connector. Low speed is used when cleaning threads, and high speed is used when cleaning sealing surfaces to improve cleaning efficiency.

[0052] The negative pressure suction module quickly and thoroughly recovers metal shavings, dust, and oil mixtures generated during the cleaning process.

[0053] After the robot inserts the pressure gauge connector into the pre-processing module, it sends a "to be cleaned" signal to the control module. The control module then sends a control signal to start the cleaning process. After cleaning is completed, the control module sends a "cleaning complete" signal, and the robot automatically transports the pressure gauge to the testing station. At the same time, the pre-processing module automatically resets, waiting for the next pressure gauge connector cleaning task, achieving a seamless connection between the cleaning and testing processes without any human intervention.

[0054] The control module includes a PLC, an industrial switch, and an industrial computer. The PLC is responsible for executing the underlying logic control, driving the AGV, conveyor belt, robot collaboration module, and pressure gauge automatic metering module to complete sequential actions and interlock protection, receiving sensor signals such as position and vision of each component, and outputting execution instructions to ensure the safe and reliable operation of the hardware. Industrial switches provide high-speed and stable industrial communication links, enabling data interaction and command transmission between vision cameras, pressure controllers, industrial computers, PLCs, and robot collaboration modules; The industrial control computer carries control system software, database and scheduling algorithms, and is responsible for data processing, task allocation, equipment status monitoring, human-machine interaction and data storage.

[0055] Furthermore, the back of the pressure gauge is provided with a QR code assigned by the laboratory information management system. The QR code includes information such as the pressure gauge's scale division, range, accuracy class, manufacturing unit, size, and connector type.

[0056] The various hardware units are interconnected through communication methods such as industrial Ethernet, ModbusTCP, and OPCUA to form a unified and collaborative automated verification hardware platform.

[0057] Figure 2 This is a diagram of the communication and multi-threaded architecture of the control module of the verification system; The control software module adopts a multi-threaded, asynchronous communication, and event listening mechanism to coordinate the joint operation of various modules. This system also adopts an integrated control architecture with multi-threaded parallel scheduling, asynchronous communication linkage, and event-driven triggering. Through the coordinated cooperation between the industrial control computer, PLC, and various execution modules, it achieves efficient, stable, and reliable operation of the entire process of automated verification.

[0058] The GUI main thread serves as the core of system interaction, providing a human-computer interaction interface to perform functions such as task entry, order management, process monitoring, data query, log display, and parameter configuration. This ensures that operators can intuitively grasp the system's operating status and perform manual intervention and mode switching when necessary.

[0059] As the underlying hardware signal processing unit, the PLC monitoring thread monitors and collects the status of each pressure gauge detection station, robot grasping and installation completion signals, AGV trolley positioning signals, and joint cleaning completion signals in real time. It feeds back the hardware execution results to the control software in real time, forming a closed-loop event mechanism of "hardware action - software response" to ensure that each process is executed in an orderly manner according to the verification logic.

[0060] The device guardian thread is used to monitor the online status of key equipment such as vision cameras, multi-range pressure controllers, AGVs, robot collaboration modules, and pre-processing modules before calibration. It collects communication status, operating parameters, and fault information in real time. When communication interruption, abnormal pressure, positioning deviation, or equipment overload is detected, the system will immediately and automatically alarm and perform protective shutdown, fault location, and backup unit switching according to preset strategies to prevent the fault from escalating and improve the system's continuous operation capability and safety.

[0061] Based on the optimal verification sequence output by the intelligent scheduling module, the task execution thread enables synchronous verification, independent pressure control, time-sharing reading, and parallel judgment of pressure gauges of various specifications. This ensures that each pressure gauge testing station operates without interference and at full capacity, significantly improving the overall verification throughput and resource utilization.

[0062] Based on the aforementioned multi-threaded architecture, the system uses an industrial control computer to uniformly handle visual recognition, intelligent scheduling and decision-making, and efficiently coordinates the front-end actuators with the back-end laboratory information management system to complete the entire process of digital closed-loop control from information recognition, task scheduling, automated verification, data storage and certificate generation.

[0063] The samples to be tested are sent to the laboratory (each with a unique identifier and the pressure gauges to be tested are placed in a tray). The AGV (Automated Guided Vehicle) transport device receives the trays in batches for feeding. The trays are then transported to the testing area for visual recognition, intelligent scheduling, robot gripping, pressure gauge connector cleaning, robot gauge installation, automatic verification and judgment, data uploading, robot unloading and sorting of the gauges, and finally, the AGV returns to the warehouse.

[0064] A method for an automatic calibration system for multi-specification pressure gauges includes the following steps: S1: Collect available historical calibration data to build a database of pressure gauge specifications, ranges, quantities, and calibration cycles. Based on this, design the selection and combination of pressure standards with different ranges, as well as the number and distribution of workstations in the automatic pressure gauge metering module. S2: The AGV trolley in the application conveying and transfer module transfers multiple pallets to the designated position, releases the transfer pallets to the transmission mechanism in sequence, and then the transmission mechanism transports them to the pressure gauge detection area (below visual recognition). S3: The visual recognition module scans the unique identifier of the pressure gauge in the laboratory information management system to obtain the preset information of all the instruments under inspection in the current tray in the laboratory information management system, and sends it to the control module. The control module will generate the optimal inspection sequence based on the instrument information of all the pressure gauges under inspection in this group, as well as the occupancy status of each pressure gauge inspection station, the range matching degree of the pressure controller, and the expected inspection time, etc., through the distributed constraint optimization algorithm. It will perform optimal sorting of all the pressure gauges in the tray, and at the same time complete the dynamic optimal mapping of each pressure gauge to the corresponding pressure gauge inspection station, and determine the grabbing order and inspection path. S4: Robot collaboration module, which receives optimal mapping sequence information and automatically completes tasks such as grasping, joint cleaning, instrument installation, and sealing and locking. S5: The automatic pressure gauge metering module works in conjunction with multiple modules to achieve automatic pressure control, pressure holding, reading, tapping, pressure resistance and reverse stroke verification according to the procedure. The visual recognition monitoring submodule takes pictures and records each step. The pressure gauge verification system software automatically identifies technical indicators such as verification point readings, calculation errors, tapping displacement, and hysteresis error, and determines the pass / fail status based on the verification results. It also synchronizes the data to the laboratory information management system and generates a certificate. S6: After the entire verification process is completed, the pressure gauge automatic metering module sends information to the control module, the control module issues instructions, the robot collaboration module disassembles the instrument, and the conveying and transfer module delivers it to the inspected area for sorting and storage.

[0065] This method synchronously stores images, data, and timestamps throughout the entire verification process, enabling full traceability.

[0066] Figure 3 This is a flowchart of the testing process; An intelligent scheduling method for an automatic calibration system for multi-specification pressure instruments, wherein after the automatic calibration system is powered on, it performs fully automated calibration according to the following process: S1: Perform system initialization and device self-test. After the control module starts, it automatically connects all hardware devices such as vision cameras, pressure controllers, robot collaboration modules, PLCs, and AGVs to complete communication connections and status verification; it manages the identification devices, pressure controllers, and workstation status in a containerized manner, supporting hot-swapping and fault self-healing; at the same time, it starts PLC listening threads, device guardian threads, and event listening threads to monitor the operating status of each unit in real time.

[0067] S2. Instruments to be inspected are stored and transferred by AGV. The pressure gauges to be tested are carried by AGV carts in a "gauge-pallet" mode. The AGV carts carry multiple pallets and use laser navigation and ground positioning landmarks to deliver each pallet to the designated position on the shelf and to transfer multiple pallets to the testing area in sequence.

[0068] S3. Visual Recognition and Information Acquisition; The visual recognition module scans and identifies the unique identifiers of all pressure gauges in the tray within the detection area, obtaining information such as pressure gauge size, range, accuracy class, connector type, and dial diameter, and then uploads this information to the control module.

[0069] S4. Intelligent Scheduling and Optimal Workstation Allocation Based on the real-time identified instrument information, the occupancy status of each pressure gauge testing station, and the range matching degree of the pressure controller, the control module generates a dynamic optimal sequence and mapping allocation scheme for each pressure gauge to be tested and the pressure gauge testing station through a distributed constraint optimization algorithm, thereby completing the dynamic optimal mapping of "pressure gauge - pressure gauge testing station" and determining the grabbing order and verification path. S5. Robotic gripping and automatic connector cleaning The robot collaboration module, following the scheduling instructions and intelligent scheduling sequence, picks up pressure gauges from the tray to clean the connectors. It uses a rotating brush head to mechanically clean the threads, and a negative pressure suction module simultaneously collects debris, ensuring that the connectors are free of impurities and oil stains, thus guaranteeing reliable sealing in the future.

[0070] S6. Precision clamping and sealing lock After cleaning, the robot collaboration module moves the pressure gauge to the mapped pressure gauge testing station. Through hand-eye vision and force-position hybrid control, it achieves precise docking between the connector and the calibration interface, automatically completes locking and sealing, and performs airtightness pre-testing.

[0071] S7. Automatically perform verification according to procedures. The automatic pressure gauge metering module, in accordance with the JJG52-2013 verification procedure, automatically executes processes such as pressure boosting, pressure holding, tapping, reading, retracement, and leakage testing. The visual recognition and monitoring submodule simultaneously acquires dial images and uploads them to the automatic pressure gauge verification system for all stages, including value recognition, error calculation, tapping displacement calculation, retracement error calculation, and verification result determination. Based on the procedure, it determines whether the gauge is qualified or unqualified and simultaneously stores the verification results, process images, timestamps, and equipment serial numbers in a local database and a cloud platform, achieving full traceability.

[0072] S8. Data Upload and Automatic Certificate Generation The calibration data is synchronized to the laboratory information management system in real time via an interface, automatically generating original records, calibration results and certificates, which are then released after being reviewed by the calibration personnel, completing the digital closed loop.

[0073] S9. Unloading, sorting, and AGV return to warehouse. The robot collaboration module removes the inspected instruments from the pressure gauge testing station and automatically sorts them according to the pass / fail results; the AGV trolley transfers the inspected pallets to the inspected area shelf, completes the warehousing and archiving, and the system waits for the next batch of tasks, realizing continuous and uninterrupted automated verification.

[0074] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An automatic calibration system for multi-specification pressure gauges, characterized in that, include: Conveying and transferring module: used to transport pressure gauges of different specifications that are laid flat to be tested; Visual recognition module: used to read and recognize the information from the pressure gauges transmitted by the conveying and transfer module; Robot collaboration module: used for grasping, cleaning connectors, installing and disassembling, and sorting pressure gauges to be tested conveyed by the transport module; Automatic pressure gauge metering module: used to monitor the occupancy status of N pressure gauge testing stations; The control module, based on the material information of multiple pressure gauges transmitted by the vision recognition module and the occupancy status of the pressure gauge testing stations transmitted by the automatic pressure gauge metering module, generates a dynamic optimal sequence of each pressure gauge to be tested and its testing station. It determines the grasping order and verification path, controls the robot collaboration module to grasp the pressure gauges according to the dynamic optimal sequence and mapping scheme, cleans the joints of the pressure gauges to be tested, places the pressure gauges on the corresponding testing station, and then controls the automatic pressure gauge metering module to run. Based on the completion signal, it sends action commands to the robot collaboration module to execute the pressure gauge disassembly and sorting operations. The process of generating the dynamic optimal sequence of each pressure gauge to be tested and the pressure gauge testing station, and determining the grabbing order and verification path is as follows: Construct a dictionary object to map the pressure gauge testing stations within the physical workshop, and assign physical constraint boundaries to each station; Based on constructing a dictionary object, connecting to the pressure gauge testing database, obtaining historical pressure gauge calibration records as training samples, and constructing a dynamic time prediction model for pressure gauge testing; Establish an independent daemon thread to monitor the status and collect data at the pressure gauge detection station; Based on the status monitoring and data of the pressure gauge testing station and the time consumption prediction during the pressure gauge testing process, the matching relationship between the pressure gauge testing task and the pressure gauge testing station is established, constraints are set, and multi-dimensional constraint solving is performed to obtain the pressure gauge grabbing order and verification path.

2. The automatic calibration system for multi-specification pressure instruments according to claim 1, characterized in that, The automatic pressure gauge module includes: Testing station submodule: includes N pressure gauge testing stations, used to place the pressure gauges to be tested; Pressure Standard Submodule: Set up at the pressure gauge testing station in the testing station submodule as a standard instrument for pressure gauge testing, used to calibrate the pressure gauge to be tested; Pressure gauge tapping test submodule: Set up on the pressure gauge test station in the test station submodule, for performing tapping tests on the pressure gauges under test as required by the procedure. Visual recognition monitoring submodule: used to identify pointer pressure gauge readings in real time during calibration, convert analog quantities into digital quantities, and photograph and store all detection information. It can also monitor the occupancy of the pressure gauge detection station in the detection station submodule in real time. Pressure controller: Used to control the pressure of the pressure gauge under test; Automatic pressure gauge calibration system: This system controls the automatic calibration of pressure gauges connected to the pressure gauge testing station, outputs the required pressure for the pressure gauge to be tested, controls the pressure standard submodule to calibrate the pressure gauge to be tested, and controls the pressure gauge tapping test submodule to perform tapping tests on the pressure gauge. Based on the visual recognition monitoring submodule, the system outputs the digital data of the pressure gauge test during the pressure gauge testing process, forming a synchronously stored file containing images, data, and timestamps.

3. The automatic calibration system for multi-specification pressure instruments according to claim 2, characterized in that: The process for determining the range of the pressure standard corresponding to the pressure gauge testing station is as follows: Obtain historical data on pressure gauge testing over multiple periods, including pressure gauge range, accuracy class, and the number of historical pressure gauge tests within each range. The proportion of pressure gauges in each range is calculated, the maximum permissible error of pressure gauges in each range is calculated, and then the requirements for standard instruments in JJG52-2013 Verification Procedure for Elastic Element Type General Pressure Gauges, Pressure Vacuum Gauges and Vacuum Gauges are combined with the requirement that the absolute value of the maximum permissible error of the standard instrument should not be greater than 1 / 4 of the absolute value of the maximum permissible error of the pressure gauge under test, to obtain the optimal combination scheme of pressure standard instruments for each pressure gauge testing station.

4. The automatic calibration system for multi-specification pressure instruments according to claim 3, characterized in that: The optimal pressure standard combination scheme includes: For the first range segment where the proportion of pressure gauges being tested is greater than the preset threshold, at least two pressure gauge testing stations should be configured. For the second range where the proportion of pressure gauges being measured is less than the preset threshold, at least one pressure gauge testing station shall be configured. Furthermore, for the K ranges that rank highest in terms of quantity in historical testing data, all pressure gauge testing stations are configured to meet their testing requirements.

5. The automatic calibration system for multi-specification pressure instruments according to claim 1, characterized in that: The independent daemon thread includes: The device anomaly guardian thread is used to send read holding register requests to the PLC controller at fixed intervals to monitor hardware fault codes. The underlying feedback listening thread is used to continuously traverse the workstation configuration dictionary object, filter out workstations with the status of being occupied, and read the level of their completion feedback coil. The external interface polling thread is used to initiate polling requests to the LIMS system via HTTP protocol at fixed intervals to obtain the pressure gauge tasks to be checked.

6. The automatic calibration system for multi-specification pressure instruments according to claim 1, characterized in that: The constraints include those for establishing a matching relationship between pressure gauge testing tasks and pressure gauge testing stations, and include the following three dimensions: Pressure constraint: The range attribute of the pressure gauge under test belongs to the closed interval formed by the minimum pressure threshold and the maximum pressure threshold of the workstation. Mechanical constraints: The connector type of the pressure gauge to be tested is available in the compatible connector specification list at the workstation; State constraint: The current lifecycle state of the workstation is idle; The execution priority order of the constraint solving is as follows: First priority: workstation status constraint, prioritize workstations in the waiting state to execute unloading and sorting tasks; Second priority: Pressure constraints and mechanical constraints are checked simultaneously when selecting idle workstations to load new tasks.

7. The automatic calibration system for multi-specification pressure instruments according to claim 1, characterized in that: The conveying and transfer module includes: Standardized tray: Used to support and flatten multiple pressure gauges of different ranges to be tested; The AGV module is used to receive multiple standardized pallets and transfer them within a predetermined area; the AGV module uses multiple AGVs. Conveying mechanism: Used to sequentially place multiple trays to the detection area after the trolley module arrives at the designated position.

8. The automatic calibration system for multi-specification pressure instruments according to claim 1, characterized in that... It also includes a pre-processing module for mechanical cleaning and debris recovery of the pressure gauge connector threads and sealing surfaces; The pre-processing module before verification includes: Rotary brush head: Used to clean the sealing surface of the pressure gauge connector. Drive unit: used to provide power for rotating the rotating brush head; Negative pressure suction module: used to collect and clean the waste generated during the cleaning process of the rotating brush head.

9. The intelligent scheduling method for an automatic calibration system for multi-specification pressure instruments according to any one of claims 1-8, characterized in that: Includes the following steps: The pressure gauges to be tested are carried by AGV carts in a gauge-pallet mode. The AGV carts carry multiple pallets and use laser navigation and ground positioning landmarks to deliver each pallet to the designated position on the shelf. The multiple pallets are then transferred to the testing area in sequence. The visual recognition module scans and identifies the unique identifiers of all pressure gauges in the tray within the detection area, obtains information on the pressure gauge size, range, accuracy class, connector type, and dial diameter, and transmits it to the control module. Based on the real-time identified pressure gauge information, the occupancy status of each pressure gauge testing station, and the range matching degree of the pressure controller, a distributed constraint optimization algorithm is used to generate the dynamic optimal sequence and mapping allocation scheme between each pressure gauge to be tested and the pressure gauge testing station, thereby completing the dynamic optimal mapping between the pressure gauge and the pressure gauge testing station and determining the grabbing order and verification path. The robot collaboration module, following the scheduling instructions and intelligent scheduling sequence, picks up pressure gauges from the tray to clean the connectors and uses a rotating brush head to mechanically clean the threads. After cleaning, the robot collaboration module moves the pressure gauge to the mapped and assigned pressure gauge testing station, achieving precise docking between the connector and the calibration interface, automatically completing locking and sealing, and performing airtightness pre-testing; The automatic pressure gauge metering module automatically executes the test process according to the JJG52-2013 verification procedure; it determines whether the test is qualified or unqualified according to the procedure, and simultaneously stores the verification results, process images, timestamps, and equipment numbers in the local database and cloud platform to achieve full traceability. The verification data is synchronized to the laboratory information management system in real time via an interface, automatically generating original records, verification results and certificates; The robot collaboration module removes the inspected pressure gauges from the pressure gauge testing station and automatically sorts them according to the pass or fail results; the AGV carts transfer the inspected pallets to the inspected area shelves, complete the warehousing and archiving, and realize continuous and uninterrupted automated inspection.

Images