Unmanned detection system and method for concrete test

Abstract

The invention relates to an unmanned detection system and method for a concrete test. The system comprises a robot cluster, an intelligent maintenance and strength detection laboratory and a test detection system, the robot cluster, the intelligent maintenance and strength detection laboratory and the test detection system are in communication connection through an industrial Internet of Things protocol; the industrial Internet of Things protocol comprises an OPC UA and an MQTT; the robot cluster comprises a concrete slump test and test block forming robot workstation, an AGV type sampling and feeding robot and a test block transfer robot. The beneficial effects of the invention are that the system can achieve the unmanned operation of the whole test process, and improves the detection efficiency.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of concrete test detection, and particularly relates to a concrete test unmanned detection system and method. BACKGROUND

[0002] With the continuous improvement of the quality requirements of the construction industry on concrete, the traditional concrete test detection gradually exposes the problems of insufficient detection efficiency and real-time performance. The existing concrete test relies on manual operation, such as slump test, test block molding, sampling, curing, etc. Not only is the efficiency low, but it is also easily affected by human factors, and cannot be fed back to the production system in real time, resulting in lagging quality control, such as the risk of not meeting the standard of slump or strength cannot be timely warned. In addition, the data is fragmented and the error is large, the test data is scattered in paper records or isolated systems, it is difficult to interconnect with production management system (MES), vehicle scheduling system, and the manual measurement error is significant, the slump error is ± 15mm, which does not meet the requirements of the quality management system. The currently built concrete intelligent mixing station also has the problems of single device function fragmentation, lack of multi-robot collaborative control, inability to cover the whole process, detection data stored in centralized database, risk of tampering, and inability to meet the engineering audit demand. SUMMARY

[0003] The purpose of the present application is to overcome the deficiencies in the prior art and provide a concrete test unmanned detection system and method.

[0004] In a first aspect, a concrete test unmanned detection system is provided, comprising:

[0005] a robot cluster, an intelligent curing and strength test laboratory, and a test detection system;

[0006] The robot cluster, the intelligent curing and strength test laboratory, and the test detection system are connected through an industrial Internet of Things protocol; the industrial Internet of Things protocol includes OPC UA and MQTT; the robot cluster includes a concrete slump test and test block molding robot workstation, an AGV type sampling and feeding robot, and a test block transfer robot.

[0007] As a preferred, it further includes a concrete generation MES system, a mixing building, an intelligent scheduling system of a mixing truck, and an unmanned mixing truck; the concrete generation MES system is connected with the mixing building and the intelligent scheduling system of the mixing truck through an industrial Internet of Things protocol, and the intelligent scheduling system of the mixing truck is connected with the unmanned mixing truck and the AGV type sampling and feeding robot through an industrial Internet of Things protocol.

[0008] As preferred, the concrete slump test and test block molding robot workstation are used to complete the concrete slump test and test block molding, the AGV sampling robot is used to perform concrete sampling, and the test block transfer robot is used to transfer the molded test block.

[0009] As preferred, the intelligent maintenance and strength detection laboratory is used for standardized maintenance and unmanned strength detection of test blocks.

[0010] As preferred, the test detection system manages data through digital twin visualization and blockchain storage technology.

[0011] In a second aspect, a concrete test unmanned detection method is provided, which is executed by the system of any one of the first aspect, comprising:

[0012] Step 1: The concrete generation MES system receives the pouring instruction and issues the production instruction to the intelligent scheduling system of the mixing building and the mixing truck. The intelligent scheduling system of the mixing truck sends the receiving instruction to the unmanned mixing truck, and the unmanned mixing truck goes to the discharge port of the mixing building to receive materials according to the receiving instruction.

[0013] Step 2: The AGV sampling robot receives the scheduling system instruction of the mixing truck and performs sampling, and then sends the sample to the feeding port of the concrete slump test and test block molding robot workstation;

[0014] Step 3: The concrete slump test and test block molding robot workstation performs concrete slump test on the sample and prepares test blocks;

[0015] Step 4: The test block is sent to the intelligent maintenance and strength detection laboratory by the test block transfer robot for maintenance and strength detection.

[0016] As preferred, step 3 comprises:

[0017] Step 3.1: Fill part of the concrete sample into the slump cylinder through the mechanical arm, vertically lift the cylinder, and install the binocular camera above the slump cylinder inside the workstation to take a concrete slump image;

[0018] Step 3.2: Identify the slump value and spread value simultaneously by fusing the YOLOv5 algorithm and the minimum bounding box algorithm;

[0019] Step 3.3: Inject the remaining concrete sample into the standard test block mold and perform test block vibration molding and demolding;

[0020] Step 3.4: Spray a two-dimensional code on the demolded test block, which corresponds to the concrete associated information, including concrete pouring order information, strength grade, production date, and engineering site.

[0021] As preferred, in step 3 and step 4, the test information is sent to a concrete production MES system; and the concrete production MES system adjusts the production instruction according to the test information.

[0022] As preferred, further comprising:

[0023] Step 5, the test key data is encrypted and saved to a block chain storage node by the test detection system.

[0024] The beneficial effects of the present application are:

[0025] 1. Realize the whole process of test unmanned, improve the detection efficiency: from the concrete sampling, test to the whole process of robot instead of manual, reduce more than 70% of labor cost, the detection efficiency is improved by 3 times.

[0026] 2. Improve the data accuracy: the slump detection error is less than or equal to 2mm, the strength prediction error is less than or equal to 3%, which is better than manual operation.

[0027] 3. Real-time quality closed loop: the test data is fed back to the MES system within 5 minutes, supporting dynamic optimization of proportioning (such as automatic adjustment of water-cement ratio), reducing the quality risk.

[0028] 4. Data traceability: the whole process data is chained, supporting the test process backtracking of any batch of concrete, through the robot operation and the block chain storage, ensuring that the detection conforms to the national standard and is traceable throughout the process. BRIEF DESCRIPTION OF DRAWINGS

[0029] Figure 1 is the overall flow chart of the concrete test provided by the present application;

[0030] Figure 2 is the data architecture diagram between the test platforms provided by the present application. DETAILED DESCRIPTION

[0031] The present application will be further described below in conjunction with the embodiments. The following description of the embodiments is only for the purpose of helping to understand the present application. It should be noted that for ordinary people in the technical field, some modifications can be made to the present application without departing from the principles of the present application, and these improvements and modifications also fall within the protection scope of the claims of the present application.

[0032] Embodiment 1:

[0033] The prior art has the following problems:

[0034] Manual detection efficiency is low: concrete slump test, test block molding and curing need multi-post manual cooperation, single test time-consuming is as long as 24 hours or more, which cannot meet the real-time demand of modern engineering for quality data, leading to production proportioning adjustment lag, which is easy to cause quality risk such as unqualified strength.

[0035] Data fragmentation and experience dependence: Test data is isolated from the concrete production management system (MES) and vehicle intelligent scheduling system, and the test results rely on paper records or scattered storage, making it difficult to achieve full-process traceability. At the same time, the slump determination and test block vibration operations are significantly influenced by personnel experience, with high data dispersion (such as manual measurement error of ±15 mm), which seriously affects the compliance of major project acceptance.

[0036] Although there are individual automated devices (such as mechanical arm test block handling and single parameter detection instrument), their functions are fragmented, lack multi-robot collaboration logic and data interconnection architecture, and cannot cover the "sampling-molding-curing-detection" whole process. Moreover, the level of intelligence is insufficient, making it difficult to cope with complex working conditions (such as dynamic environmental interference).

[0037] With the development of intelligent manufacturing and industrial robot technology, it is possible to use robots, AGVs, industrial Internet of Things (IIoT) and other technologies to realize the automation and intelligence of concrete test detection. By connecting the concrete test detection platform with the production management system (MES) of the mixing plant, the mixing truck scheduling system, and the unmanned mixing truck, the data interconnection and intercommunication can be realized, and the digital and intelligent management of the whole process of concrete sampling-molding-curing-detection can be realized, ensuring the timeliness and accuracy of the data.

[0038] To solve the problems of the prior art, Embodiment 1 of the present application provides a concrete test unmanned detection system, comprising:

[0039] a robot cluster, an intelligent curing and strength test laboratory, and a test detection system;

[0040] The robot cluster, the intelligent curing and strength test laboratory, and the test detection system are connected through an industrial Internet of Things protocol; the industrial Internet of Things protocol includes OPC UA (high reliability data) and MQTT (high-frequency real-time data).

[0041] The robot cluster includes a concrete slump test and test block molding robot workstation, an AGV sampling and feeding robot, and a test block transfer robot. The concrete slump test and test block molding robot workstation is used to complete the concrete slump test and test block molding, the AGV sampling and feeding robot is used to perform concrete sampling, and the test block transfer robot is used to transfer the molded test block.

[0042] Specifically, the concrete slump test and test block forming robot workstation: adopts machine vision (binocular camera + AI algorithm) to automatically identify slump and spread, replacing traditional manual measurement; the mechanical arm integrates a vibration compaction module, automatically completes test block forming according to the standard process (GB / T 50081), and eliminates human operation differences. The concrete slump test and test block forming robot workstation comprises a concrete automatic feeding station, a concrete slump test and test block forming workbench, a mobile insertion and tamping station, a six-axis robot system scheduling station, a test block stationary rack, a cleaning station, a turret type test block constant temperature curing bin, and a matching electrical system, pipeline (water), etc.; the six-axis robot system integrates a clamp and an insertion and tamping mechanism, which is used for transferring the slump bucket and the test block between stations, and performs grabbing, bucket lifting, smoothing and tamping operations on the slump bucket and the test block box through appropriate posture and motion trajectory, automatically completes the slump test and test block forming according to the standard process (GB / T 50081), and eliminates human operation differences. The system uses machine vision (binocular camera + AI algorithm) to automatically measure the slump and spread of concrete, replacing traditional manual measurement. The concrete slump test and test block forming workbench can generate strong magnetic attraction to attract the slump bucket and the test block box to the platform reference position. RFID technology and image recognition (two-dimensional code) technology are used to realize digital traceability of concrete test blocks, and the process control parameter library of the concrete slump test and test block forming is used to control the test and test block forming quality.

[0043] AGV type sampling and feeding robot: the AGV type sampling and feeding robot integrates a laser SLAM module, an ultrasonic obstacle avoidance sensor, a six-degree-of-freedom mechanical arm, a constant temperature sampling container, a temperature and humidity monitoring unit, and a 5G industrial communication module on a navigation chassis, realizes automatic sampling and accurate feeding in the concrete production process, the laser SLAM realizes dynamic positioning (accuracy ±5 cm) in combination with the pre-set high-precision map of the mixing station, the ultrasonic sensor detects the position deviation of the mixer truck discharge port in real time and corrects the mechanical arm trajectory, the six-degree-of-freedom mechanical arm carries an adaptive clamp jaw, performs layered sampling (depth ≥30 cm) according to the ASTM C172 standard, the constant temperature sampling container maintains a 20±2℃ environment through a semiconductor refrigerating fin, the temperature and humidity monitoring unit synchronously records the transportation process parameters (data uploaded to the MES system), the 5G communication module ensures millisecond-level instruction interaction with the scheduling system, and finally guarantees the representativeness of the concrete sample and the reliability of the test data.

[0044] The test block transfer robot is integrated with a laser radar (LiDAR), a depth vision camera, a mechanical arm force control clamp, an RFID reader, an inertial measurement unit (IMU), an environment temperature and humidity sensor, and a ROS controller on an omnidirectional wheel moving chassis to realize the full-process unmanned transfer of the concrete test block from the molding area to the curing room and then to the pressure testing machine. The laser radar is used to construct a high-precision environment map and realize real-time navigation and obstacle avoidance, the depth vision camera is used to identify the test block stacking position and surface state (such as damage detection), the mechanical arm force control clamp is used to adaptively grasp the test block (150mmx150mmx150mm standard test block) through torque feedback, the RFID reader is used to bind the test block production batch and detection data, the IMU is used to monitor the real-time stability of the robot posture (anti-toppling), the environment temperature and humidity sensor is used to ensure that the transfer process meets the test block curing specification (humidity≥95%), and the ROS controller is used to integrate the multi-sensor data fusion, path planning and instruction interaction with the MES system, so as to realize the precise and traceable operation of the whole test block transfer process.

[0045] The intelligent curing and strength test laboratory includes a constant temperature and humidity curing bin and a full-automatic test block pressure testing machine.

[0046] Specifically, the constant temperature and humidity curing bin is integrated with a PID temperature control module, an ultrasonic humidifier, a multi-channel RFID read-write system, a circulating air duct structure and an industrial-grade sealed box to realize the standardized curing of the concrete test block in the whole life cycle. The PID temperature control module dynamically adjusts the temperature based on the thermocouple feedback (20±1℃, in line with the ISO 1920-3 standard), the ultrasonic humidifier maintains the humidity≥95% (accuracy±2%), the RFID read-write system automatically scans the test block label and binds the curing time and batch information, the circulating air duct ensures the uniform distribution of temperature and humidity (fluctuation≤±0.5℃) through the centrifugal fan and the guide plate, the industrial-grade sealed box is designed with a polyurethane insulation layer and an airtight door to prevent external environmental interference, and the curing data is uploaded to the MES system in real time through the Modbus TCP protocol to ensure the traceability and compliance of the test block strength development.

[0047] The mechanical arm sends the test block into the testing machine. The full-automatic test block pressure testing machine realizes unmanned detection and intelligent analysis of the compressive strength of concrete by configuring a servo hydraulic loading system, a six-axis collaborative mechanical arm, a high-precision pressure sensor, an AI analysis module, and an industrial Internet of Things interface, etc. The servo hydraulic system accurately pressurizes at a rate of 0.5 MPa / s (in accordance with the ASTM C39 standard), the six-axis mechanical arm positions and grabs the test block and places it on the pressure bearing table (with a repeat positioning accuracy of ±0.1 mm), the pressure sensor collects the load-displacement curve in real time (with a sampling frequency of 1 kHz), the AI analysis module predicts the test block failure mode based on the convolutional neural network (CNN) and calculates the compressive strength (with an error of ≤2%), and the detection result is directly connected to the platform database through the OPC UA protocol, synchronously triggering quality alarm or proportioning optimization instructions, meeting the real-time and data tamper-proof requirements of engineering acceptance.

[0048] The test detection system has built-in data acquisition, analysis, and feedback modules, supports digital twin visualization and blockchain storage. The embodiment of the application displays the test progress, data distribution, and abnormal early warning (such as triggering an alarm when the slump is out of tolerance) in real time by developing a digital twin interface. In addition, key data is stored through blockchain technology to ensure that the detection results are tamper-proof and meet the engineering acceptance audit requirements.

[0049] Specifically, the test detection system realizes intelligent management of concrete test data through integration of multiple source heterogeneous data acquisition modules, real-time stream processing engines, AI analysis clusters, digital twin visualization platforms, and blockchain storage nodes, etc. The data acquisition module is compatible with OPC UA and MQTT protocols, and can real-time aggregate multi-dimensional data such as slump, spread, and pressure strength (sampling frequency ≥100Hz). The stream processing engine realizes abnormal value filtering and feature extraction (such as slump drop detection) based on the Flink framework. The AI analysis cluster predicts the strength development trend and generates a quality report (confidence ≥98%) through the LSTM model. The digital twin platform constructs a 3D visualization interface based on Three.js, dynamically maps the test progress, equipment status, and data distribution (supports VR interaction), and the blockchain node uses the Hyperledger Fabric framework to SHA-256 encrypt the key data (original record, operation log) and chain it to ensure the integrity and tamper resistance of the audit traceability. Finally, a "collection-analysis-decision-storage" closed loop is formed to meet the ISO / IEC 17025 laboratory accreditation and engineering acceptance specifications.

[0050] In addition, the system further comprises a concrete production MES system, a mixing building, an intelligent scheduling system of a mixing truck and an unmanned mixing truck; the concrete production MES system is in communication connection with the mixing building and the intelligent scheduling system of the mixing truck through an industrial Internet of Things protocol; and the intelligent scheduling system of the mixing truck is in communication connection with the unmanned mixing truck and an AGV type sampling and feeding robot through the industrial Internet of Things protocol.

[0051] The key communication details of the devices in the system are shown in Table 1.

[0052] Table 1

[0053]

[0054] Embodiment 2

[0055] On the basis of Embodiment 1, Embodiment 2 of the application provides a method for unmanned detection of concrete tests, and proposes a robot cluster collaborative operation mode for the whole process (sampling-molding-curing-detection) of concrete tests. Based on a distributed scheduling algorithm, multi-robot task allocation and path planning are realized. Through a hybrid architecture of OPC UA and MQTT protocols, the data barriers between production, transportation and detection systems are broken. Test data is fed back to the MES system in real time, and the concrete proportioning and production rhythm are dynamically optimized to realize data interconnection and closed loop, and to replace manual judgment by fusing machine vision (concrete slump and spreadability detection) and AI pressure curve analysis (test block strength prediction) to realize intelligent detection of concrete performance parameters.

[0056] Specifically, the method comprises the following steps.

[0057] Step 1: The concrete production MES system receives a pouring instruction and issues a production instruction to the mixing building and the intelligent scheduling system of the mixing truck; the intelligent scheduling system of the mixing truck sends a material receiving instruction to the unmanned mixing truck, and the unmanned mixing truck goes to the discharge port of the mixing building to receive materials according to the material receiving instruction.

[0058] Specifically, the project engineering department publishes a pouring instruction to the concrete production MES system through a handheld terminal or a PC terminal according to the construction progress of the site for a certain pouring site.

[0059] Then, the MES system issues a production instruction to the mixing building and the mixing truck scheduling system, which contains information such as pouring instruction, production volume, mix proportion, strength grade, pouring site, etc. The scheduling system generates train information and path planning according to the real-time idle state of the mixing truck, sends a material receiving instruction to the unmanned mixing truck of the corresponding train, and the unmanned mixing truck receives the instruction and goes to the discharge port of the mixing building to receive materials through the planned path.

[0060] Wherein, the mixing building is also called concrete mixing building, which is an engineering mechanical equipment for concrete production in construction engineering. It is mainly composed of five modules of raw material storage (cement, aggregate, admixture), metering system, material conveying, mixing host and control system. Its function is to automatically weigh and mix raw materials into concrete according to the specified proportion, and convey it to the mixer truck through the discharge port. In this application, the mixing building is interconnected with MES through OPC UA protocol, which is the execution terminal of production instruction, and receives the proportioning instruction in real time and uploads the production status (such as mixing current, discharge amount).

[0061] Wherein, the raw material storage module is used to store cement, sand, stone and other raw materials, which are conveyed to the metering system by belt. The material metering system is used to accurately weigh the aggregate, cement, water and admixture (error ≤ ± 1%). The mixing host is used to mix the materials into homogeneous concrete. The control system is used to receive the proportioning instruction of MES and adjust the feeding proportion synchronously.

[0062] Step 2, the AGV type sampling and feeding robot receives the instruction of the mixer truck scheduling system and executes sampling, and then transports the sample to the feeding port of the concrete slump test and test block forming robot workstation.

[0063] Specifically, the AGV type sampling and feeding robot executes the following steps:

[0064] 1) Receive the mixer truck scheduling system instruction (through MQTT topic ` / agv / sample / task`), and obtain the target mixer truck number and driving path position information.

[0065] 2) Navigate to the mixer truck discharge port based on laser SLAM, and the mechanical arm executes multi-point sampling (3 times of longitudinal sampling, depth ≥ 30 cm).

[0066] 3) Put the sample into a constant temperature (20±2℃) and constant humidity (≥95%) container, and transport it to the feeding port of the concrete slump test and test block forming robot workstation by the AGV type sampling and feeding robot.

[0067] Step 3, the concrete slump test and test block forming robot workstation performs concrete slump test on the sample and prepares test blocks.

[0068] Step 3 includes:

[0069] Step 3.1, fill part of the concrete sample into the slump cylinder through the mechanical arm, then vertically lift the cylinder, install it inside the workstation, and the binocular camera installed above the slump cylinder takes the concrete slump image.

[0070] Step 3.2, identify the slump value and spread value synchronously by fusing YOLOv5 algorithm and minimum bounding box algorithm.

[0071] Specifically, step 3.2 includes:

[0072] Step 3.2.1, YOLOv5 locates the target area, detects the slump cone and crops the region of interest (ROI), reducing the image processing range;

[0073] Step 3.2.2, Minimum Bounding Box Algorithm for Accurate Measurement: Edge detection and contour analysis are performed within the target area, and Gaussian filtering + edge detection is performed on the ROI image to extract the concrete contour.

[0074] Step 3.2.3, Calculate the minimum bounding rectangle height (slump) and maximum bounding circle diameter (spread) of the contour.

[0075] The concrete slump image has challenges such as adhesion, edge blur, and light interference. Traditional manual measurement relies on operator visual scaling, which is easily affected by viewing angle errors, and single visual algorithm cannot balance efficiency and accuracy. Fusion of YOLOv5 and minimum bounding box algorithm can improve measurement accuracy and robustness, overcoming the false detection problem of single visual algorithm in dynamic light or dusty environment, ensuring synchronous and accurate output of spread (maximum bounding circle diameter) and slump (rectangle height). The specific advantages are as follows:

[0076] Fast positioning of YOLOv5: quickly identify the position of the slump cone (millisecond-level response) in complex backgrounds (such as dynamic light, dusty environment, and workstation instrument interference), crop the region of interest (ROI), greatly reduce the image processing range, reduce data processing volume, and improve detection efficiency

[0077] Sub-pixel accuracy of minimum bounding box: Gaussian filtering + edge detection is performed on the concrete contour within the ROI, and the minimum bounding rectangle / circle algorithm is used to eliminate manual visual errors (±2mm accuracy) to accurately measure the minimum bounding rectangle height (slump) and maximum bounding circle diameter (spread) of the concrete contour.

[0078] The fusion of both not only leverages the fast positioning capabilities of YOLOv5 algorithm, but also utilizes the minimum bounding box algorithm for precise measurement. Compared to single algorithm, the fused algorithm can more efficiently and accurately identify the slump value and spread value simultaneously, with an accuracy of ±2mm, which is superior to traditional manual measurement (slump error up to ±15mm), reducing human error and improving the reliability of concrete performance parameter detection.

[0079] Step 3.3, inject the remaining concrete sample into the standard test block mold and perform test block vibration molding and demolding.

[0080] For example, the remaining concrete is automatically injected into a standard test block mold (150mmx150mmx150mm), and a vibration table is vibrated at a frequency of 50Hz for 30 seconds. 4) After the test block is vibrated and formed, the mechanical arm sends the test block mold to the work station into the turret type constant temperature curing bin (16-20℃) for 48 hours, and then demolds.

[0081] Step 3.4, after demolding the test block, a two-dimensional code is sprayed, which corresponds to the concrete associated information, including the concrete pouring order information, the strength grade, the production date and the engineering site.

[0082] Step 4, the test block is sent into the intelligent curing and strength detection laboratory for curing and strength detection by the test block transfer robot.

[0083] Specifically, the test block transfer robot performs the following steps:

[0084] 1) The mechanical arm grabs the formed test block and scans the RFID tag (records the production batch and time).

[0085] 2) The test block is sent into the constant temperature and humidity curing bin (temperature 20±2℃, humidity≥95%) in the intelligent curing laboratory, and cured to the specified age (7d / 28d).

[0086] The intelligent curing laboratory performs the following steps:

[0087] 1) After curing, the mechanical arm places the test block in the full-automatic pressure testing machine.

[0088] 2) The testing machine is pressurized at a rate of 0.5MPa / s, and the pressure-displacement curve is collected in real time, and the final strength is predicted by the LSTM model (error≤3%).

[0089] 3) The detection result is written into the MES database through OPC UA, and the matching ratio adjustment instruction (such as dynamic optimization of water-cement ratio) is triggered.

[0090] It should be noted that the method provided in this embodiment provides a corresponding method for the system of Embodiment 1, so in this embodiment, the same or similar parts as Embodiment 1 can be mutually referenced, and will not be repeated in this application.

[0091] Embodiment 3:

[0092] On the basis of Embodiment 2, the concrete test unmanned detection method provided in Embodiment 4 of the present application is more specific, and before implementing the method, system deployment and initialization are required, including:

[0093] (1) Hardware networking

[0094] The test detection area is demarcated in the stirring station, the concrete slump test and the test block forming robot, the AGV type sampling and feeding robot, and the intelligent laboratory are deployed, and are connected to the edge server through an industrial Ethernet.

[0095] The unmanned stirring truck and the stirring building equipment access the system through a 5G network, and the MES system interacts with the edge server through an OPC UA protocol.

[0096] (2) Software configuration

[0097] A digital twin platform is deployed in the edge server, and a device information model (such as a robot working state and a test parameter) is defined.

[0098] An MQTT topic (for example, ` / agv / sample / status` for AGV type sampling and feeding robot state reporting and ` / mes / ratio_adjust` for mix ratio adjustment instruction issuing) is configured.

[0099] Then, the concrete test unmanned detection method performed by the system includes the following steps:

[0100] Step 1, the concrete generation MES system receives a pouring instruction, and issues a production instruction to the intelligent scheduling system of the stirring building and the stirring truck; the intelligent scheduling system of the stirring truck sends a receiving instruction to the unmanned stirring truck, and the unmanned stirring truck goes to the discharge port of the stirring building to receive materials according to the receiving instruction;

[0101] Step 2, the AGV type sampling and feeding robot receives the stirring truck scheduling system instruction, performs sampling, and then transports the sample to the feeding port of the concrete slump test and test block forming robot workstation;

[0102] Step 3, the concrete slump test and test block forming robot workstation performs a concrete slump test on the sample and prepares a test block;

[0103] Step 4, the test block is sent to the intelligent curing and strength detection laboratory for curing and strength detection by a test block transfer robot.

[0104] In steps 3 and 4, test information is sent to the concrete generation MES system; and the concrete generation MES system adjusts the production instruction according to the test information.

[0105] Specifically, high-frequency data such as AGV state and slump value are published to the edge server through MQTT. The strength detection result and the production mix ratio are synchronized with the MES system through OPC UA.

[0106] Step 5, the test key data is encrypted and saved to the block chain storage node by the test detection system.

[0107] Specifically, the test key data (strength value, operation timestamp) is encrypted and chained (based on Hyperledger Fabric) for engineering acceptance audit calling.

[0108] It should be noted that the application of the blockchain is not only to save data. In the present application, the blockchain utilizes its decentralized, tamper-proof and traceable characteristics to not only achieve data saving, but also to ensure that the engineering acceptance party and the supervision unit can verify the authenticity of the data through the permission chain mechanism of Hyperledger Fabric. For example, when an auditor checks the 28-day strength data of a test block, the encrypted record on the chain can be called to match parameters such as curing temperature and humidity curve and press loading rate, forming an irrefutable quality evidence chain - which cannot be achieved by a centralized database.

[0109] On the one hand, the integrity and tamper resistance of the detection data are ensured, so that the data cannot be modified at will, and the strict requirements of engineering acceptance audit on data authenticity and reliability are met; on the other hand, based on the traceability of the blockchain, the test process of any batch of concrete can be traced back, which is convenient for supervision and examination of the entire test process. At the same time, the blockchain storage node interacts with other systems for data exchange, such as uploading data such as curing timestamp, temperature and humidity history record to the blockchain by the intelligent curing warehouse, ensuring the consistency and continuity of the whole process data, and providing strong support for the whole life cycle management of concrete quality.

[0110] In addition, the present application embodiment also designs the following exception handling mechanism:

[0111] 1. If the slump is out of tolerance (such as ±10mm of the set value), the system automatically freezes the batch of concrete outbound, and notifies the vehicle scheduling system to adjust the transportation plan.

[0112] 2. Sampling exception: AGV mechanical arm gripper pressure sensor detects sampling failure (such as container not in place), automatically retries 3 times and reports to MES.

[0113] 3. During the transfer of the test block, if the environmental temperature and humidity sensor detects that the transfer environment humidity is lower than 95%, which does not meet the test block curing specification, the system can automatically start the humidifying equipment to increase the humidity;

[0114] 4. When the IMU of the test block transfer robot detects that the robot posture is unstable and may fall, the system will immediately control the robot to stop moving and adjust the posture to ensure the safety of the test block.

[0115] 5. Robot failure: mechanical arm torque exceeds limit or stops moving; immediately switch to standby robot to take over the task and upload logs to the blockchain.

[0116] 6、Test block damage: The transfer robot identifies test block cracks through a depth vision camera, automatically isolates defective test blocks, and triggers a sample replacement process.

[0117] 7、In the intelligent maintenance and strength test laboratory, when the temperature and humidity of the constant temperature and humidity curing chamber exceed the standard range (temperature not in 20±1℃, humidity not in ≥95%±2%), the system will automatically adjust the PID temperature control module and the ultrasonic humidifier to restore it to the standard state.

[0118] 8、If the pressure sensor of the full-automatic test block pressure testing machine collects abnormal data, or the AI analysis module predicts that the test block strength differs greatly from the expected value, the system will pause the detection, troubleshoot and calibrate the equipment to ensure the accuracy of the test results.

[0119] 9、Blockchain storage failure: Local cache data that has not been chained will be automatically retransmitted after network recovery to ensure audit integrity.

[0120] It should be noted that the parts in this embodiment that are the same as or similar to those in Embodiment 2 can be mutually referenced, and will not be described in detail in this application.

[0121] In summary, the present application realizes an unmanned detection system through industrial Internet of Things protocols (OPC UA, MQTT) and interconnection with other systems, enabling efficient data interaction between systems and laying the foundation for the unmanned and digital management of the entire process of concrete testing. Moreover, the robot cluster cooperates in work, and the multi-robot system (slump test, AGV sampling, test block transfer) realizes task allocation and collision avoidance through a distributed scheduling algorithm, covering the entire process of “sampling-molding-curing-detection”, realizing the entire process of unmanned operation from sampling, molding to curing and detection. In addition, the intelligent maintenance and strength test laboratory standardizes the curing and unmanned strength detection of test blocks, and the test detection system uses digital twin visualization and blockchain storage technology to realize intelligent management of data. The data interconnection architecture uses a hybrid protocol of OPC UA (high reliability data) and MQTT (high-frequency real-time data) to break down the data barriers of MES, scheduling systems, and unmanned mixers (see Figure 2 ), realizing a “detection-feedback-optimization” closed loop (such as real-time stopping of concrete discharge when the slump exceeds the tolerance). These technologies work together to solve many problems in traditional concrete test detection, improve detection efficiency and data accuracy, and achieve data traceability.

Claims

1. A concrete test unmanned detection system, characterized in that, The system comprises: a robot cluster, an intelligent maintenance and strength test laboratory, and a test detection system; the robot cluster, the intelligent maintenance and strength test laboratory, and the test detection system are connected through an industrial Internet of Things protocol; the industrial Internet of Things protocol comprises OPC UA and MQTT; the robot cluster comprises a concrete slump test and test block molding robot workstation, an AGV sampling and feeding robot, and a test block transfer robot.

2. The concrete testing unmanned detection system according to claim 1, wherein, The system further comprises a concrete generation MES system, a mixing plant, an intelligent mixer scheduling system, and an unmanned mixer; the concrete generation MES system is connected to the mixing plant and the intelligent mixer scheduling system through an industrial Internet of Things protocol, and the intelligent mixer scheduling system is connected to the unmanned mixer and the AGV sampling and feeding robot through an industrial Internet of Things protocol.

3. The concrete testing unmanned detection system of claim 2, wherein, The concrete slump test and test block molding robot workstation is used to complete concrete slump detection and test block molding, the AGV sampling and feeding robot is used to perform concrete sampling, and the test block transfer robot is used to transfer the molded test blocks.

4. The concrete testing unmanned detection system of claim 3, wherein, The intelligent maintenance and strength test laboratory is used to standardize the maintenance of test blocks and to detect the strength of test blocks without human intervention.

5. The concrete testing unmanned detection system of claim 4, wherein, The test detection system manages data through digital twinning visualization and blockchain storage technology.

6. A method for unmanned detection of a concrete test, characterized in that The system is executed by the system of claim 1, comprising: Step 1: The concrete generation MES system receives pouring instructions and issues production instructions to the mixing plant and the intelligent mixer scheduling system; the intelligent mixer scheduling system sends a material receiving instruction to the unmanned mixer, and the unmanned mixer goes to the unloading port of the mixing plant to receive materials according to the instruction; Step 2: The AGV sampling and feeding robot receives the instruction from the mixer scheduling system, performs sampling, and then sends the sample to the feeding port of the concrete slump test and test block molding robot workstation; Step 3: The concrete slump test and test block molding robot workstation performs concrete slump test on the sample and prepares test blocks; Step 4: The test block transfer robot sends the test blocks to the intelligent maintenance and strength test laboratory for maintenance and strength detection.

7. The method of claim 6, wherein the method further comprises: Step 3 comprises: Step 3.1: Fill part of the concrete sample into the slump cylinder through the mechanical arm, vertically lift the cylinder, and install the double-camera above the slump cylinder inside the workstation to take a slump image of the concrete; Step 3.2: Identify the slump value and the spread value simultaneously by fusing the YOLOv5 algorithm and the minimum bounding box algorithm; Step 3.3: Inject the remaining concrete sample into the standard test block mold and perform test block vibration molding and demolding; Step 3.4: Spray a two-dimensional code on the demolded test block, which corresponds to the concrete associated information, including concrete pouring instruction information, strength grade, production date, and engineering site.

8. The method of claim 7, wherein, In steps 3 and 4, the test information is sent to the concrete generation MES system; the concrete generation MES system adjusts the production instructions according to the test information.

9. The method of claim 8, wherein, Further comprising: Step 5: The test detection system encrypts and saves the test key data to the blockchain storage node.

Images