A pipe segment intelligent production line based on robot and AI control and a production method thereof

Abstract

The application discloses a kind of based on robot and AI control's pipe piece intelligent production line and production method, the production line includes AI general control unit, multi-robot system, data acquisition unit and maintenance kiln system, the AI general control unit is by industrial server and the AI algorithm software platform deployed in industrial server constitutes, the AI general control unit is generated to each batch pipe piece production, generates pipe piece initial production data list, according to production data list, generates the pipe piece mould type with electronic tag, and production data list is sent to each process station by production scheduling optimization model, collaborative path planning model, quality prediction model, adaptive adjustment model and maintenance optimization model and is launched to cross-process linkage constraint.This application can realize the automatic and intelligent collaborative production control of pipe piece whole process, ensure seamless connection between each process, avoid the excessive consumption of energy in traditional maintenance mode, effectively reduce the amount of steam.

Description

Technical Field

[0001] This invention belongs to the field of shield tunnel construction equipment technology, and in particular relates to an intelligent production line and production method for tunnel segments based on robot and AI control. Background Technology

[0002] With the continuous expansion of urban rail transit, underground utility tunnels, and other infrastructure construction, shield tunneling has been widely adopted due to its advantages such as safety, efficiency, and minimal impact on the ground environment. As the main permanent lining structure of shield tunnels, the quality of shield segments directly affects the overall safety and service life of the tunnel. Traditional segment production mainly relies on manual labor or semi-automated machinery, resulting in problems such as low automation, inconsistent production rhythm, poor quality stability, high dependence on skilled workers, and high energy consumption. Currently, the production models for tunnel boring machine (TBM) segments both domestically and internationally are mostly still in the mechanized or semi-automated stage. Typical production lines usually consist of multiple independent or semi-interlocking workstations, such as mold cleaning, rebar cage installation, concrete pouring, vibration, curing, and demolding. The connections between these processes mainly rely on gantry cranes and railcars for transport, resulting in poor automation continuity. Production rhythm is easily affected by human factors and equipment scheduling. In the concrete pouring and vibration stages, fixed or simple robotic arms are often used for extensive operations, making it difficult to make precise and adaptive adjustments based on the actual distribution of concrete within the mold. This easily leads to quality defects such as voids and honeycomb-like pitting inside the segments. The curing process usually relies on manual experience to control the temperature and humidity curves of the curing kiln, lacking real-time data feedback and closed-loop control, resulting in uneven segment strength development and high energy consumption. Furthermore, various data records during the production process (such as material proportions, equipment status, environmental parameters, and quality inspection results) are scattered, making it difficult to achieve full-process quality traceability and production optimization. While existing technologies have partially incorporated industrial robots to perform repetitive tasks, the lack of intelligent collaboration between robot units and between robots and other production equipment (such as curing kilns and testing equipment) results in poor overall system flexibility, hindering rapid response to changes in production orders or adjustments to process parameters. This discrete, experience-dependent production model has become a bottleneck restricting the improvement of tunnel segment quality, the reduction of production costs, and industrial upgrading, making it difficult to meet the demands of large-scale, fast-paced urban underground engineering construction. Summary of the Invention

[0003] The purpose of this invention is to provide an intelligent production line and method for tunnel segments based on robot and AI control. This invention enables full-process automation of tunnel segments and intelligent collaborative production control of multiple robot systems. It provides unified scheduling and path planning for the multiple robot systems, ensuring seamless connection between various processes, avoiding excessive energy consumption in traditional maintenance methods, and effectively reducing steam usage. To achieve the above objectives, the following technical effects are adopted: According to one aspect of the present invention, a smart production line for tube segments based on robot and AI control is provided, characterized in that: the smart production line for tube segments includes an AI central control unit, a multi-robot system, a data acquisition unit, and a curing kiln system; The AI ​​central control unit consists of an industrial server and an AI algorithm software platform deployed within the industrial server. The AI ​​algorithm software platform integrates a production scheduling optimization model, a collaborative path planning model, a quality prediction model, an adaptive adjustment model, and a maintenance optimization model. The multi-robot system includes at least a mold cleaning robot, a steel reinforcement cage welding robot, a steel reinforcement cage handling robot, a concrete pouring robot, a surface finishing robot, a demolding robot, and an AGV transfer robot arranged at each workstation of the segment production process along the track conveyor line. Each robot communicates with the AI ​​central control unit via an industrial Ethernet. The data acquisition unit is deployed along each workstation on the track conveyor line to collect production data of each workstation and upload it to the AI ​​central control unit. The curing kiln system includes a curing kiln body and a curing control unit. The curing kiln body is used to receive tube segments for steam curing. The curing control unit receives curing process parameters issued by the AI ​​central control unit and controls the curing process parameters within the curing kiln body to perform steam curing on the tube segments. For each batch of tunnel segments, the AI ​​central control unit generates an initial production data list for the tunnel segments, and obtains the type of tunnel segment mold with electronic tags based on the production data list. The production data list is then distributed to each workstation through the production scheduling optimization model, collaborative path planning model, quality prediction model, adaptive adjustment model, and maintenance optimization model for cross-process linkage constraints. The process stations arranged along the track conveyor line include steel bar processing station, mold preparation station, concrete pouring and vibration station, curing station, and segment demolding and operation station. RFID electronic tag readers are installed at the entrance of each process station arranged along the track conveyor line. The RFID electronic tag readers are communicatively connected to the AI ​​central control unit. The data acquisition unit and the multi-robot system are communicatively connected to the AI ​​central control unit via industrial Ethernet or wireless network.

[0004] In a further preferred embodiment of the above scheme, the data acquisition unit includes: Displacement and temperature sensors installed on the segment mold are used to monitor the mold closing accuracy and real-time temperature; a high-definition industrial camera is configured at the mold preparation station to take pictures during mold cleaning; and a laser scanner for detecting rebar parameters is deployed at the rebar processing station. Temperature and humidity sensors, material level sensors, and flow meters are used at the concrete pouring and vibration stations to monitor the concrete mix proportions and mixing quality. And a curing sensor array installed inside the curing kiln body at the curing station to collect curing environment parameters; In addition, force and vision sensors are integrated on each robot end effector to provide real-time feedback of force control and visual positioning information during various operations such as grasping, handling, and vibration.

[0005] In a further preferred embodiment of the above scheme, a curing kiln body is set at the curing station of the track conveyor line, and the track conveyor line passes through the inlet and outlet ends of the curing kiln body. A curing control unit connected to the AI ​​central control unit is set outside the curing kiln body at the curing station. The curing sensor array includes a temperature sensor, a humidity sensor, and a non-contact infrared sensor for monitoring the temperature of the tube segment body. The temperature sensor, humidity sensor, carbon dioxide concentration sensor, and non-contact infrared sensor are respectively connected to the curing control unit.

[0006] In a further preferred embodiment of the above scheme, a dehumidification system is provided on the side wall of the curing kiln, and a steam delivery pipe and a humidification delivery pipe are provided inside the curing kiln at the curing station. A steam heater and a steam control valve are respectively installed on the steam delivery pipe inside the curing kiln, and a humidification control valve is installed on the humidification delivery pipe. The curing control unit includes a controller and a network communication module. The controller is connected to the AI ​​central control unit through the network communication module. The steam heater, steam control valve, humidification control valve, and dehumidification system are respectively connected to the curing control unit.

[0007] In a further preferred embodiment of the above scheme, the production scheduling optimization model is used to receive the production data list and generate the production queue of the process station according to the production plan list, and then dynamically adjust the production tasks according to the production queue and perform cooperative motion control on the robot. The collaborative path planning model is used to generate collision-free motion operation paths, operation status and operation time windows for each robot system at each work station, ensuring that each robot system at each work station operates collaboratively and on time and safely according to the prescribed path. The quality prediction model is used to predict, identify and compare the production data of each process station collected by the data acquisition unit with the corresponding pre-stored quality inspection data to identify whether the production data of each process station for producing and processing pipe segments meets the requirements. The adaptive adjustment model is used to adaptively adjust the cycle time and motion path of the robot system at each work station based on the real-time acquired operation status and operation time window of each robot system at each work station, combined with the scheduling trigger instructions generated by the production queue and sent to each robot system. The maintenance optimization model is used to call a preset maintenance parameter model according to the characteristics of each batch of segments, monitor the occupancy status of the static stop area and the kiln exit time, and obtain the distribution of maintenance parameters at different maintenance stages in the maintenance kiln in real time to compare and analyze whether there is a deviation in maintenance parameters. It automatically provides feedback and makes decisions on whether the maintenance parameters need to be dynamically optimized. According to the kiln exit time and the occupancy status of the static stop area, it sends the kiln exit command and kiln entry command of the segment mold to the production scheduling optimization model in advance. The production scheduling optimization model issues scheduling commands to schedule the AGV transfer robot to wait for the cured segments to be transferred to the stacking area or water curing tank for storage and secondary maintenance.

[0008] In a further preferred embodiment of the above scheme, the production scheduling optimization model also dynamically adjusts the production queue and production tasks based on factors such as the priority of the production plan list, the status of the segment mold, and the performance parameters of the robot system at each workstation.

[0009] In a further preferred embodiment of the above scheme, the production scheduling optimization model also reads the mold status, the occupancy status of the curing kiln, and the busy / idle status of the robot based on the production queue, and binds the electronic tag of the segment mold to the priority of the production plan list, and adjusts the production queues for allocating process positions; the production queues for process positions include a mold cleaning queue, a rebar placement queue, a pouring queue, a finishing queue, a kiln placement queue, a demolding queue, and a transfer queue, and each production queue carries an expiration date; In a further preferred embodiment of the above scheme, if dynamic optimization of curing parameters is required, the steam valve opening is adjusted in real time or the curing length is extended / shortened. Based on the current curing parameters, deep learning is performed to determine whether there are deviations in the curing parameters within the curing kiln, and real-time feedback and decision-making are made based on the current deviations.

[0010] According to another aspect of the present invention, the present invention provides an intelligent production method for tunnel segments based on robot operation and intelligent control, the intelligent production method for tunnel segments comprising the following steps: Step 1: Before production begins, the AI ​​central control unit generates a production plan list based on the order information; based on the production plan list, a production data list for each process station is preset; the AI ​​central control unit generates a production queue for each process station based on the production data list for each process station, and issues production tasks to each process station in sequence according to the production queue and dynamically issues running path instructions to each robot system. Step 2, Rebar cage fabrication: Obtain the rebar cage production task from the production data list, generate the fabrication parameters based on the rebar cage production task, automatically grab the pre-specified rebar material from the material area based on the fabrication parameters, automatically bind and weld the rebar processing station, and then transport the bound and welded rebar cage to the track conveyor line according to the production queue through the rebar cage handling robot and send it to the mold preparation station. Step 3, pre-processing and transfer of segment molds: Obtain the segment mold model according to the production data list, retrieve the corresponding segment mold to the mold preparation station according to the segment mold model, perform surface cleaning and release agent spraying pre-treatment instructions on the segment mold in sequence, and then transport the steel reinforcement skeleton sent from the steel reinforcement processing station into the pre-treated segment mold, and transfer it to the concrete pouring and vibration station according to the production queue via the track conveyor line. Step 4, Concrete Pouring and Vibration: Obtain the concrete pouring task from the production data list, and activate RFID or visual recognition to transmit the segment mold to the concrete pouring and vibration station and obtain the mold model. Upload the mold model to the AI ​​central control unit and retrieve the corresponding 3D model data according to the mold model. Calculate the cavity volume and structural features of the current segment mold. Based on the cavity volume and structural features and according to the production queue, execute the preset concrete receiving quantity and receive the concrete command. Pour the concrete into the segment mold at a uniform speed and in a quantitative manner. Vibrate during or after pouring. Dynamically adjust the vibration parameters based on the real-time collected concrete slump data. Step 5, Static Stop and Initial Curing: Obtain the static stop and initial curing tasks according to the production data list, and wait in real time for the completed casting, pouring and vibration of the segment molds to be delivered along the track conveyor line. Obtain the electronic tag of the segment mold and send it into the static stop area of ​​the curing kiln body according to the production queue for parking. Wait for the instruction to be sent into the curing kiln body to execute the initial curing task. After the initial curing is completed in the curing kiln body and the solidified segment model is formed, the segment model carried by the segment mold is moved out of the curing kiln body along the track conveyor line to the segment demolding and operation station. Step 6, Demolding and Secondary Curing: Obtain the concrete demolding and secondary curing tasks from the production data list. Scan the electronic tags of the segment molds delivered after curing in real time according to the production queue. Then execute the demolding operation command on the segment model inside the segment mold. Transfer the demolded segment model to the designated stacking area or water curing tank for storage and secondary maintenance. The empty mold after demolding is returned to the cleaning station to wait for the next process.

[0011] In a further preferred embodiment of the above scheme, whenever the segment mold moves to the next process station, the RFID electronic tag reader at that process station reads the electronic tag of the segment mold, and reports the electronic tag of the segment mold and the current process station number to the AI ​​central control unit. The AI ​​central control unit updates the location information of the segment mold accordingly and verifies whether the segment mold holds the production queue corresponding to the current station. If the verification passes, the robot at the current process station is allowed to execute the production task at that station; if the verification fails, an alarm is triggered and the task execution is prevented. In summary, the present invention adopts the above technical solution, and the present invention has the following technical effects: (1) In terms of production efficiency, it achieves full-process automation and intelligent collaboration, enabling unified scheduling and path planning for multiple robot systems, ensuring seamless connection between various processes, and significantly reducing production interruptions and time waste caused by hoisting, waiting, and manual intervention in the traditional model. The robots can work 24 hours a day without interruption, the production rhythm is stable and controllable, and the overall production efficiency is qualitatively improved, which can meet the needs of large-scale, fast-paced urban underground engineering construction. (2) Regarding product quality stability, the production status can be monitored in real time through data acquisition units (such as vision sensors, force sensors, temperature and humidity sensors, etc.) distributed throughout the production line. For example, in the concrete pouring stage, the pouring speed and location are monitored in real time to ensure uniform concrete distribution; in the vibration stage, the vibration frequency and depth are adaptively adjusted based on sensor feedback to effectively avoid over-vibration or under-vibration, and significantly reduce the generation of internal defects such as air bubbles and voids in the segments. During the curing process, the core temperature and environmental parameters of the segments are monitored in real time, and the temperature and humidity curve of the curing kiln is dynamically adjusted to make the concrete strength development more uniform and efficient, ensuring that each segment can meet the design strength requirements; (3) It has significant advantages in reducing energy consumption and costs. Through intelligent maintenance and optimized control, it achieves precise energy supply, avoids excessive energy consumption in traditional maintenance methods, and effectively reduces the amount of steam or electricity used. Automated production reduces reliance on a large number of skilled workers; (4) It realizes the digitalization and traceability of the production process. The data of the whole process is recorded and stored in the system in real time, forming a complete production data chain. It provides strong data support for quality traceability, process analysis, cost accounting and optimization decision-making, promotes the transformation and upgrading of tunnel segment production towards digitalization, networking and intelligence, conforms to the development trend of intelligent construction, and provides a comprehensive solution for shield tunnel segment manufacturing industry that is efficient, high-quality, low-consumption and intelligent. Attached Figure Description

[0012] Figure 1 This is a schematic diagram of the overall structure of a smart production line for pipe segments based on robot and AI control according to the present invention; Figure 2 This is a schematic diagram of the internal layout of the curing kiln of the present invention; Figure 3 This is a schematic diagram of the control principle of a smart production line for pipe segments based on robot and AI control according to the present invention; In the attached diagram, the components are: AI central control unit 1, data acquisition unit 2, track conveyor line 3, segment mold 4, RFID electronic tag reader 5, curing kiln 10, and automatic lifting door 11. Detailed Implementation

[0013] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and preferred embodiments. However, it should be noted that many details listed in the specification are merely to provide the reader with a thorough understanding of one or more aspects of the present invention, and these aspects of the invention can be implemented even without these specific details.

[0014] Combination Figure 1 As shown, according to the present invention, a smart production line for tunnel segments based on robot and AI control includes an AI central control unit, a multi-robot system, a data acquisition unit, and a curing kiln system. The AI ​​central control unit consists of an industrial server and an AI algorithm software platform deployed within the industrial server. The AI ​​algorithm software platform integrates a production scheduling optimization model, a collaborative path planning model, a quality prediction model, an adaptive adjustment model, and a curing optimization model. The multi-robot system includes at least mold cleaning robots, rebar cage welding robots, rebar cage handling robots, concrete pouring robots, surface finishing robots, demolding robots, and AGV transfer robots arranged at each workstation of the tunnel segment production process along the track conveyor line 3. Each robot communicates with the AI ​​central control unit via an industrial Ethernet network. The data acquisition unit is deployed at each workstation along the track conveyor line 3 to collect production data from each workstation and upload it to the AI ​​central control unit. The curing kiln system includes a curing kiln body and a curing control unit. The curing kiln body receives tunnel segments for steam curing, and the curing control unit receives data from the AI ​​central control unit. The system generates maintenance process parameters and controls the parameters of each maintenance process within the maintenance kiln to perform steam maintenance on the segments. For each batch of segments produced, the AI ​​central control unit generates an initial production data list and distributes this data list to each workstation through a production scheduling optimization model, collaborative path planning model, quality prediction model, adaptive adjustment model, and maintenance optimization model for cross-process linkage constraints. Workstations arranged along the track conveyor line include rebar processing workstations, mold preparation workstations, concrete pouring and vibration workstations, maintenance workstations, and segment demolding and operation workstations. RFID electronic tag readers 5 are installed at the entrance of each workstation along the track conveyor line, and these readers are communicatively connected to the AI ​​central control unit. The data acquisition unit and the multi-robot system communicate with the AI ​​central control unit via industrial Ethernet or a wireless network. Based on the production data list, they hoist segment molds with electronic tags. Each segment mold type includes at least the mold model, its corresponding mold shape, and mold size. The mold type is then distributed to the mold preparation workstation, and the corresponding segment mold is retrieved by the robot.

[0015] An AGV (Automated Guided Vehicle) transfer robot for loading and transferring tunnel segments is installed at the segment demolding and operation station. The AGV transfer robot is electrically connected to the AI ​​central control unit. In this invention, each robot body of the multi-robot system is the main actuator of the production line. The robot system includes, but is not limited to, the following types of industrial robots: a six-axis heavy-duty robot for handling and installing rebar skeletons at the rebar processing station, equipped with a vision recognition system and adaptive gripper; a mold cleaning robot for cleaning tunnel segment molds at the mold preparation station; a concrete pouring and vibration robot dedicated to concrete pouring at the concrete pouring and vibration station, equipped with a high-precision flow control valve and three-dimensional path scanning function, and a collaborative robot for finishing and smoothing the surface of tunnel segments, with force control feedback to ensure work quality; and AGV transfer robots and / or track-type mobile robots for transferring finished tunnel segments at the segment demolding and operation station. These robots do not operate independently, but are dynamically assigned tasks and coordinated in motion by the AI ​​central control unit 1 according to the actual work cycle time and process status.

[0016] In this invention, combined with Figure 1 , Figure 2 and Figure 3As shown, more specifically, the data acquisition unit includes: displacement and temperature sensors installed on the segment mold for monitoring the mold closing accuracy and real-time temperature; temperature and humidity sensors, level sensors, and flow meters for materials at the concrete pouring and vibration stations for monitoring the concrete mix proportions and mixing quality; and a curing sensor array installed inside the curing kiln at the curing station for collecting curing environment parameters; the curing sensor array includes temperature sensors, humidity sensors, carbon dioxide concentration sensors, and a non-contact infrared sensor for monitoring the segment body temperature, the temperature sensors, humidity sensors, and non-contact infrared sensors being connected to the curing control unit; and force and vision sensors integrated on each robot end effector for real-time feedback of force control information and visual positioning information during the grasping, handling, and vibration operations; more specifically, in this invention, when a mold cleaning is configured at the mold preparation station... High-definition industrial cameras are used to detect the cleanliness and damage of the mold surface; laser scanners are deployed at the rebar processing station to verify the size and positional accuracy of the rebar skeleton processed by the rebar processing equipment; material level sensors, temperature and humidity sensors, temperature sensors, and flow meters are installed at the concrete pouring and vibration stations to monitor the state of concrete raw materials, concrete mix proportions, and mixing quality, and vibration sensors are configured during pouring and vibration to assess concrete density; an array of temperature and humidity sensors and a carbon dioxide concentration sensor are arranged inside the curing kiln 10 to accurately monitor the temperature, humidity, and carbon dioxide concentration of the curing environment; the track conveyor line 3 runs through both ends of the curing kiln 10, and the segments loaded in the segment mold 4 move along the track conveyor line 3 while being steam-heated and cured and accelerated in the curing kiln 10; data collected by the data acquisition unit 2 is uploaded in real time to the AI ​​central control unit 1 via the EtherCAT bus.

[0017] In this invention, combined with Figure 1 , Figure 2 and Figure 3As shown, a curing kiln body 10 is installed at the curing station of the track conveyor line 3. The track conveyor line 3 passes through the inlet and outlet ends of the curing kiln body 10. A curing control unit connected to the AI ​​central control unit is installed outside the curing kiln body 10 at the curing station. Electronic tags are installed on the outer wall of the segment mold. RFID electronic tag readers 6 are installed at the inlet and outlet positions of the curing kiln body at the curing station. The RFID electronic tag readers are connected to the curing control unit. The production process is recorded through electronic tags to digitally identify and trace the segment mold, achieving full-flow traceability. The process data is recorded and stored in the system in real time. A dehumidification system is installed on the side wall of the curing kiln at the curing station. A steam delivery pipe and a humidification delivery pipe are installed inside the curing kiln at the curing station. A steam heater and a steam control valve are installed on the steam delivery pipe inside the curing kiln, and a humidification control valve is installed on the humidification delivery pipe. The steam heater, steam control valve, humidification control valve, and dehumidification system are respectively connected to the curing control unit. The curing control unit includes a controller and a network communication module. The controller is a PLC controller. The controller is connected to the humidification control valve, the steam heater, the steam control valve, the humidification control valve, and the dehumidification system via the Modbus TCP protocol. The network communication module is one or more of a local area network communication module, a 5G communication module, or a Wi-Fi communication module. The controller is connected to the AI ​​central control unit through the network communication module. The curing kiln is equipped with a curing control unit directly controlled by an AI central control unit 1. This control unit manages steam, spray humidification, and ventilation / heat dissipation. When a segment mold with an electronic tag arrives at the curing kiln after passing through the previous process, the tag information on the mold is read, and key parameters such as the mold model, concrete mix proportion, and kiln entry time are sent to the AI ​​central control unit 1. The AI ​​central control unit 1 can then set curing parameters based on the mold model, concrete mix, curing age, and environmental parameters, dynamically adjusting the curing curve to achieve precise and intelligent curing. For example, the AI ​​central control unit 1 can set various optimal curing curves or parameters for different types and specifications of segments (such as standard blocks, adjacent blocks, and capping blocks). Each curve defines in detail the target temperature, target humidity, and duration required for different curing stages (such as the static rest zone, low-temperature zone, heating period, heat preservation period, cooling period, and cleaning zone).

[0018] In this invention, the curing kiln 10 is a tunnel-type steam curing kiln used for steam curing of tunnel segments. The tunnel-type steam curing kiln has multiple independently controlled temperature and humidity curing zones. Each curing zone includes a static curing zone, a low-temperature zone, a heating zone, a heat preservation zone, a cooling zone, and a cleaning zone. The heating zone has multiple heating sections, and during the heating process, different temperature ranges are gradually increased within each heating section for curing. The heat preservation zone has multiple heat preservation sections, and during the heat preservation process, multiple sections with the same temperature are used for heat preservation, allowing for simultaneous heat preservation of the tunnel segments in a first-in, first-out order, thus accelerating production speed and efficiency. The cooling zone has multiple cooling sections, and different temperature ranges are gradually decreased within each cooling section. The tunnel segment mold 4 is then used for casting, pouring, and vibration. The concrete segments are conveyed along track 3 into the static resting area for slow water absorption and sequential steam curing. After preheating in the low-temperature zone, they are sent to the heating zone for steam curing at a rate of 10-15°C per hour until the maximum temperature is reached. Then, they are sent to the cooling zone for gradual cooling. After cooling, the segments are rinsed with warm water at the same temperature as the cooling zone and then sent out of the curing kiln 10. The tunnel-type steam curing kiln is equipped with heat-insulating automatic lifting doors 11 at both ends and between each curing zone to isolate each zone and ensure that the temperature of the curing kiln meets the requirements. A dehumidification system (exhaust fan) is installed in each zone. Steam delivery pipes and humidification delivery pipes are installed in the low-temperature zone, heating zone, constant temperature zone, and cooling zone, respectively.

[0019] In this invention, combined with Figure 1 , Figure 2 and Figure 3As shown, the production scheduling optimization model receives the production data list and generates production queues for each workstation based on the production plan list. It then dynamically adjusts production tasks and controls the coordinated motion of robots based on these queues. The model also dynamically adjusts the production queues and tasks based on factors such as the priority of the production plan list, the status of the segment mold, and the performance parameters of the robot system at each workstation, thereby optimizing the standard cycle time window, minimum total completion time, and maximum utilization of the robot system for each workstation. Furthermore, the model reads the mold status, the occupancy status of the curing kiln, and the robot's busy / idle status based on the production queue, binds the electronic tag (ID) of the segment mold to the priority of the production plan list, and adjusts the production queues allocated to each workstation. The production queues at each workstation include a mold cleaning queue, a rebar placement queue, a pouring queue, a finishing queue, a kiln entry queue, a demolding queue, and a transfer queue. Each production queue carries an expiration date. The collaborative path planning model is used to obtain the current production queue and generate a collision-free motion operation path for each robot system at each workstation, as well as the operation status and operation time window under the current operation path. This ensures that each robot system at each workstation operates collaboratively on time and safely according to the prescribed path. When a robot system experiences an operation timeout due to an anomaly, the adaptive adjustment model calculates the time margin for subsequent processes. If the time margin is greater than the threshold, the original operation cycle is maintained. If the time margin is insufficient, the motion path of the affected robot system is replanned to reduce idle travel time, and the AGV transportation system is coordinated to adjust the mold transfer sequence. The quality prediction model is used to predict, identify, and compare production data collected by the data acquisition unit from each workstation with corresponding pre-stored quality inspection data. This identifies whether the production data of each workstation's segment production meets requirements, thereby obtaining key production process parameters and their optimal ranges that affect segment quality (e.g., strength, durability, density, appearance defects) and production efficiency (e.g., cycle time, energy consumption). For example, pre-stored quality inspection data and corresponding workstation generation data are used to predict and identify segment defects (e.g., cracks, bubbles). Based on the identification and prediction, the accuracy and generalization ability of the quality prediction model are continuously optimized to achieve a steady improvement in production efficiency and product quality. The pre-stored quality inspection data includes, but is not limited to: historical production data accumulated over a long period, such as environmental temperature and humidity, concrete mix proportions, vibration parameters, steam curing curves, and segment performance parameters during the production of each segment; quality inspection data throughout the entire production process, such as segment appearance images collected by the robot vision system, geometric dimension data obtained by laser scanning, and strength test results; therefore, vision sensors are deployed at key workstations such as mold cleaning, application of release agent, installation of reinforcing cage, concrete pouring, and segment demolding to collect data on the mold's closure status, the cleanliness of the mold cavity, the uniformity of the release agent application, whether the reinforcing cage is installed in place, the flatness of the concrete pouring surface, and the appearance of the segment after demolding. By observing defect images, the AI-powered central control unit's quality prediction model can assess the completion quality of each process in real time and predict production trends through image recognition. For example, it can predict potential equipment failures and perform early warning maintenance, or optimize process parameters at each stage based on historical data to improve quality consistency and reduce energy consumption. Typically, torque or vision sensors are integrated into the end effector of robots performing tasks such as grasping, handling, and vibration. For instance, when a robot is installing rebar cages, torque or vision sensors can provide real-time feedback of visual images / torque data during the installation process, ensuring accurate positioning and docking of the rebar cage within the segment mold and avoiding collision damage. During vibration robot operations, torque sensors can... The system monitors the contact force between the vibrator and the concrete to ensure consistent vibration results. It continuously acquires all equipment status, production process parameters, material information, and environmental data through a data acquisition unit. Based on real-time monitoring of anomalies in the joint angles, motor current, temperature, vibration, running speed, and current position of each robot, the system detects and understands the operational progress of each robot, predicting the completion progress and time window of the current task. When any status becomes abnormal or exceeds a safety threshold, the quality prediction model immediately triggers an alarm and reports it to the production scheduling optimization model for anomaly handling. Based on the anomaly, a scheduling strategy is formulated (such as stopping and waiting, reducing speed, or activating backup equipment).The quality prediction model issues an early warning signal, the production scheduling optimization model suspends the allocation of subsequent production queues for this batch, and readjusts the production queues and production tasks. At the same time, the quality prediction model outputs the recommended adjustment value of the process parameters and forwards it to the corresponding robot system to achieve real-time process parameter optimization. An adaptive adjustment model is used to adaptively adjust the cycle time and movement path of each robot system at each workstation based on real-time acquisition of the operational status and time window of each workstation, combined with scheduling trigger commands generated by the production queue (such as robot deceleration by 10% or adaptive gripping of fixtures), and distribute them to each robot system. This adapts to meet the needs of segment production progress adjustments. In the vibration stage, the vibration frequency and depth are adaptively adjusted based on sensor feedback to effectively avoid over-vibration or under-vibration, significantly reducing the generation of internal defects in the segments, such as bubbles and voids. During the curing process, real-time monitoring data from temperature and humidity sensors in each zone of the curing kiln is received. Based on the real-time monitoring of the core temperature, humidity, and environmental parameters of the segments, it is determined whether the specified requirements are met. If not, the steam regulating valve opening and exhaust fan speed are dynamically adjusted based on the operational time window, thereby dynamically and adaptively adjusting the curing kiln. The temperature and humidity curves ensure more uniform and efficient concrete strength development, guaranteeing that each segment meets design strength requirements. This results in extremely high product quality consistency and reliability, thereby reducing energy consumption and maintenance costs. AI-optimized control enables intelligent and adaptive control of the segment maintenance process, ensuring each segment completes maintenance in the optimal environment. This significantly improves product quality consistency and production efficiency, avoiding excessive energy consumption in traditional maintenance methods and effectively reducing steam or electricity usage. During rebar handling and installation, the adaptive adjustment model automatically adjusts the robot's perception system and adaptively grasps the clamps. The maintenance model dynamically generates a unique temperature, humidity, and time curing curve based on the rebar specifications, concrete mix ratio, and segment cross-sectional dimensions of the batch of segments. The energy consumption optimization strategy also feeds real-time energy consumption data back to the production scheduling optimization model for energy cost assessment in the next batch scheduling. The AI-controlled unit's curing optimization model is used to call preset curing parameter models based on the characteristics of each batch of tunnel segments (reinforcement specifications, concrete mix ratio, segment cross-sectional dimensions, strength, durability, density, etc.) and monitor the occupancy status of the static resting zone and the kiln exit time. Within 2-4 hours after pouring, the poured segment molds are sent to the static resting zone of curing kiln 10 for initial settling before entering the kiln for fully automated steam curing. During curing, the distribution of curing parameters at different curing stages within the kiln is acquired in real time and compared with the preset curing parameter model to analyze for any deviations. Automatic feedback and decision-making are then made regarding whether dynamic optimization of curing parameters is needed. If dynamic optimization is required, the steam valve opening is adjusted in real time, or the curing duration is extended / shortened. To this end, the curing optimization model uses deep learning based on the current curing parameters to determine if deviations exist within the kiln, and provides real-time feedback and decisions based on these deviations. This significantly improves production efficiency, stabilizes the consistency of segment quality, and maximizes the efficiency of production. This reduces the cost of manual intervention and energy consumption. The maintenance parameter model includes, but is not limited to, steam adjustment parameters (adjustment of the opening of the steam control valve), maintenance environment parameters (temperature, humidity, carbon dioxide, etc.), maintenance curve parameters, static stop space, static stop time, etc. By monitoring the temperature, humidity, and maintenance time in the maintenance kiln in real time, optimal maintenance is achieved, making the actual maintenance curve parameters close to the maintenance curve of the preset maintenance parameter model. In this invention, the maintenance optimization model also issues the kiln exit command and kiln entry command of the segment mold 4 to the production scheduling optimization model in advance based on the kiln exit time and the occupancy status of the static stop area. The production scheduling optimization model issues scheduling commands to schedule the AGV transfer robot to wait for the maintenance segments to be transferred to the stacking area or water curing tank for storage and secondary maintenance, and to wait for the segment mold 4 to be sent into the static stop area of ​​the maintenance kiln 10 after pouring, casting, and vibration. This completes the planning of the optimal transfer path and maintenance operation path. The whole process is seamlessly connected, avoiding the maintenance kiln 10 being idle or congested.

[0020] The AI-powered central control unit (CCU) schedules multiple robots to work collaboratively. By analyzing historical collaboration data, the CCU optimizes task allocation logic and action timing among robots, reducing waiting time and avoiding interference in robot movement paths. This improves overall operational smoothness and production efficiency. The CCU coordinates and controls production queues at each workstation on the production line, achieving a high degree of automation, intelligence, and flexibility in tunnel segment production. This significantly improves production efficiency, stabilizes product quality, and allows for production trend prediction. For example, it can predict potential equipment failures and perform early warning maintenance, as well as optimize process parameters at each stage to improve quality consistency and reduce energy consumption. All production data is recorded and forms a "digital twin" for each tunnel segment, enabling full lifecycle quality traceability. Through coordinated control, the production process can gradually adapt to dynamic factors such as raw material fluctuations and equipment performance changes, while reducing reliance on human experience and energy consumption during production, ultimately achieving a steady improvement in production efficiency and product quality.

[0021] In this invention, combined with Figure 1 , Figure 2 and Figure 3 As shown, this invention also provides an intelligent production method for tunnel segments based on robot operation and intelligent control. This invention implements the intelligent production method for tunnel segments based on a robot and AI-controlled intelligent production line. The AI ​​central control unit 1 of this invention establishes real-time data connections with the subsystems (robot systems) of each workstation on the production line via industrial Ethernet. The AI ​​central control unit 1 is responsible for coordinating and controlling all production links. After the production line starts, the AI ​​central control unit 1 acts as the central server for the entire production line. Through communication protocols such as industrial Ethernet, 5G, or Wi-Fi, it establishes real-time data connections with all key equipment, including the concrete mixing plant, rebar processing equipment, mold cleaning robot, concrete pouring robot, vibration platform, curing kiln, tunnel segment demolding robot, and data acquisition unit. The massive amounts of data received by the AI ​​central control unit 1 are processed and analyzed in real time. For example, specifically regarding the control of the curing kiln, the kiln itself is equipped with a network of multiple high-precision sensors, including temperature sensors, humidity sensors, and non-contact infrared sensors for monitoring the temperature of the tube segments. These sensors are distributed to cover all areas of the curing kiln, ensuring the collection of representative real-time environmental data. The intelligent production method for the tube segments includes the following steps: Step 1: Before production begins, the AI ​​central control unit generates a production plan list based on order information. Based on this production plan list, a production data list for each workstation is preset. The AI ​​central control unit then generates production queues for each workstation based on these production data lists, and generates segment mold types with electronic tags based on the production data lists. The production scheduling optimization model receives the production data lists and generates production queues for each workstation based on the production plan list. It then sequentially issues production tasks to each workstation based on the production queues and dynamically sends running path instructions to each robot system. The production plan list mainly includes the segment model, quantity, and delivery date. The production data list includes the work cycle time, raw material parameters, preset initial values ​​of vibration parameters, and curing curve parameters for each stage. The production queues are primarily process sequence queues for segment processing. The production queues for each workstation include a mold cleaning queue, a rebar placement queue, a pouring queue, a finishing queue, a kiln entry queue, a demolding queue, and a transfer queue. Each production queue carries an expiration date. Step 2, Rebar skeleton fabrication: The AI ​​central control unit obtains the rebar skeleton production task based on the production data list and sends the production task to the robot system at the rebar processing station. Based on the rebar skeleton production task, the system generates production parameters and automatically grabs the rebar material of the predetermined specification from the material area based on the production parameters. The system performs automated binding and welding at the rebar processing station and then uses a rebar skeleton transport robot to transport the bound and welded rebar skeleton to the track conveyor line according to the production queue and deliver it to the mold preparation station. At the rebar processing station, a robot for binding the rebar skeleton begins its work. Based on the segment specifications and models issued by the AI ​​control unit 1, the robot automatically picks up the rebar of the predetermined specifications from the material area and performs precise skeleton forming and binding operations on a dedicated binding table. At this position, a six-axis heavy-duty robot identifies the position of the rebar through a vision sensor to ensure the accuracy of the binding points. The completed rebar skeleton is then transported to the next station by an automatic conveying system. Step 3: Segment Mold Pre-processing and Transfer. The AI ​​central control unit obtains the segment mold model from the production data list and sends it to the robot system at the mold preparation station. The robot system retrieves the corresponding segment mold based on the model and places it at the mold preparation station. An RFID electronic tag reader reads the electronic tag (ID) corresponding to each segment mold model entering the mold preparation station and uploads the electronic tag (ID) to the AI ​​central control unit. The segment mold is then subjected to surface cleaning and release agent spraying pre-treatment commands sequentially. The steel reinforcement cage from the steel reinforcement processing station is then transported into the pre-treated segment mold and conveyed via a track conveyor line to the concrete pouring and vibration station. At this point, mold cleaning and release agent spraying are performed. The mold cleaning robot begins operation. Equipped with cleaning tool heads and spraying devices, the robot first uses brushing and air blowing to clean the residue from the segment mold used in the previous cycle. Then, visual data captured by a high-definition industrial camera is uploaded to the AI ​​control unit 1 for visual inspection to confirm the cleanliness of the inner surface of the segment mold. After confirmation, the mold cleaning robot sprays the release agent evenly on the inner surface of the segment mold according to the spraying path and spraying amount set by the AI ​​control unit 1 to prepare for concrete pouring. After visual inspection and confirmation of the spraying by capturing the spraying data again, the robot transports the steel reinforcement skeleton from the steel reinforcement processing station into the segment mold. Step 4: Concrete Pouring and Vibration. The AI ​​central control unit obtains the concrete pouring task from the production data list and issues it to the robot system at the concrete pouring and vibration station. It then initiates RFID or visual recognition to transmit the information to the segment mold at the concrete pouring and vibration station and obtains the mold model. The mold model is uploaded to the AI ​​central control unit, which retrieves the corresponding 3D model data based on the mold model. The unit calculates the cavity volume and structural features of the current segment mold. Based on the cavity volume and structural features, and according to the production queue, it executes the preset concrete receiving quantity and receives the concrete command. The concrete is poured uniformly and quantitatively into the segment mold. Vibration is performed during or after pouring, and the vibration parameters are dynamically adjusted based on the real-time collected concrete slump data. The segment mold... The carried steel reinforcement cage is delivered to the concrete pouring and vibration station for the concrete pouring and vibration stage. The concrete pouring and vibration robot, which is dedicated to this process, receives instructions from the AI ​​control unit 1 and picks up the predetermined amount of concrete from the concrete mixing plant. The robot moves precisely above the prepared segment mold and performs uniform and quantitative concrete placement through the pouring arm at its end. During or after pouring, the vibration device integrated into the concrete pouring and vibration robot, such as a vibrator, penetrates into the concrete, allowing the segment mold to be efficiently vibrated on the vibration table to ensure uniform concrete density and eliminate air bubbles. The AI ​​control unit 1 dynamically adjusts the vibration parameters based on the real-time collected concrete slump data and analyzes the vibration data in real time to ensure that the density meets the standards. Step 5, Static Stop and Initial Curing: The AI ​​central control unit retrieves the static stop and initial curing tasks from the production data list and simultaneously assigns them to the robot system and curing kiln system at the curing station. It also waits in real-time for the completed casting and vibration of the segment mold 4, delivered along the track conveyor line 3. It acquires the electronic tag of the segment mold 4 and sends it to the static stop area of ​​the curing kiln body according to the production queue, awaiting the initial curing task instruction. After the initial curing is completed and the segment model solidifies within the curing kiln body, the segment mold 4 carrying the segment model is moved from the curing kiln body along the track conveyor line to the segment demolding and operation station. The segment mold that has completed casting and vibration is then transferred... The track-mounted mobile robot is delivered into the curing kiln for curing. At the entrance of the curing kiln, it reads the electronic tag on the current segment mold. The AI ​​central control unit 1 sends the curing process parameters to the controller of the curing control unit based on the concrete formula, material, age, and environmental parameters of the segment. The controller intelligently adjusts the temperature and humidity change curve of the curing kiln to achieve precise and intelligent curing and ensure the strength of the segment. Temperature sensors, humidity sensors, and non-contact infrared sensors inside the curing kiln transmit the real-time temperature and humidity data to the controller of the curing control unit. The controller adjusts the curing parameters based on the real-time temperature and humidity data and curing time to achieve optimal curing. Step 6: Demolding and Secondary Curing. The AI ​​central control unit obtains the concrete demolding and secondary curing tasks from the production data list and simultaneously issues them to the robot systems at the segment demolding and operation stations. According to the production queue, the robot scans the electronic tags of the segment mold 4 delivered after curing in real time, then executes the demolding operation command on the concrete inside the segment mold 4. It collects the process parameter data of the demolded segment. If the process parameter data is qualified, it allocates the next process production queue and notifies the robot collaborative path planning model to update the robot task queue. The demolded concrete is then transferred to a designated storage area or water curing tank for storage and secondary curing. The empty mold returns to the cleaning station, waiting to enter the next process. If the process parameter data is unqualified, a redo or scrap process is triggered and uploaded to the production scheduling optimization model to readjust the production queue. The force during the demolding operation is controlled between 40-100. kN, demolding speed 0.2-1.2mm / s; the empty mold is transferred to the mold preparation station to remove excess concrete from the surface, and then to the automatic rough finishing station where the finishing robot performs rough finishing operation to make the surface flatness error of the segment ≤2mm / m; after curing, the dedicated demolding robot at the segment demolding and operation station uses a vision positioning system to identify the mold locking device and automatically complete the mold opening action. Using the adaptive gripper of another gripping robot, the formed finished segment is taken out from the segment mold and placed on the AGV transfer vehicle. The finished segment is transferred by the AGV transfer vehicle to the stacking area or water curing tank, and the empty mold returns to the cleaning station to start the next cycle. In the entire production process, the AI ​​central control unit 1 continuously acquires production data in real time, optimizes the process parameters and collaborative strategies of each link, thereby continuously improving production efficiency, stabilizing product quality, and reducing unit energy consumption and manual intervention.

[0022] In steps 1 to 6 above, whenever the segment mold is transferred to the next process station, the RFID electronic tag reader of that process station reads the electronic tag (ID) of the segment mold and reports the electronic tag (ID) of the segment mold and the number of the current process station to the production scheduling optimization model of the AI ​​central control unit. The production scheduling optimization model updates the location information of the segment mold accordingly and verifies whether the segment mold holds the production queue corresponding to the current station. If the verification passes, the robot of the current process station is allowed to execute the production task of that station. If the verification fails, the previous process station triggers an alarm and prevents the task from being executed. In this invention, after production is completed, the AI ​​central control unit aggregates all batches of production process data, quality inspection results, and actual energy consumption data, and periodically performs incremental training on the production scheduling optimization model, collaborative path planning model, quality prediction model, adaptive adjustment model, and maintenance optimization model to update and optimize model parameters, thereby optimizing the time estimation parameters in each model and improving scheduling accuracy. The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A smart production line for tunnel segments based on robot and AI control, characterized in that: The intelligent production line for tube segments includes an AI central control unit, a multi-robot system, a data acquisition unit, and a curing kiln system; The AI ​​central control unit consists of an industrial server and an AI algorithm software platform deployed within the industrial server. The AI ​​algorithm software platform integrates a production scheduling optimization model, a collaborative path planning model, a quality prediction model, an adaptive adjustment model, and a maintenance optimization model. The multi-robot system includes at least a mold cleaning robot, a steel reinforcement cage welding robot, a steel reinforcement cage handling robot, a concrete pouring robot, a surface finishing robot, a demolding robot, and an AGV transfer robot arranged at each workstation of the segment production process along the track conveyor line. Each robot communicates with the AI ​​central control unit via an industrial Ethernet. The data acquisition unit is deployed along each workstation on the track conveyor line to collect production data of each workstation and upload it to the AI ​​central control unit. The curing kiln system includes a curing kiln body and a curing control unit. The curing kiln body is used to receive tube segments for steam curing. The curing control unit receives curing process parameters issued by the AI ​​central control unit and controls the curing process parameters within the curing kiln body to perform steam curing on the tube segments. For each batch of tunnel segments, the AI ​​central control unit generates an initial production data list for the tunnel segments and generates a tunnel segment mold type with electronic tags based on the production data list. The production data list is then distributed to each workstation through a production scheduling optimization model, a collaborative path planning model, a quality prediction model, an adaptive adjustment model, and a maintenance optimization model for cross-process linkage constraints. The process stations arranged along the track conveyor line include steel bar processing station, mold preparation station, concrete pouring and vibration station, curing station, and segment demolding and operation station. RFID electronic tag readers are installed at the entrance of each process station arranged along the track conveyor line, and the RFID electronic tag readers are communicatively connected to the AI ​​central control unit. The data acquisition unit and the multi-robot system are communicatively connected to the AI ​​central control unit through industrial Ethernet or wireless network.

2. The intelligent production line for tunnel segments based on robot and AI control according to claim 1, characterized in that: The data acquisition unit package: Displacement and temperature sensors installed on the segment mold are used to monitor the mold closing accuracy and real-time temperature; a high-definition industrial camera is configured at the mold preparation station to take pictures during mold cleaning; and a laser scanner for detecting rebar parameters is deployed at the rebar processing station. Temperature and humidity sensors, material level sensors, and flow meters are used at the concrete pouring and vibration stations to monitor the concrete mix proportions and mixing quality. And a curing sensor array installed inside the curing kiln body at the curing station to collect curing environment parameters; In addition, force and vision sensors are integrated on each robot end effector to provide real-time feedback of force control and visual positioning information during various operations such as grasping, handling, and vibration.

3. The intelligent production line for tunnel segments based on robot and AI control according to claim 2, characterized in that: A curing kiln body is set up at the curing station of the track conveyor line. The track conveyor line passes through the inlet and outlet ends of the curing kiln body. A curing control unit connected to the AI ​​central control unit is set up outside the curing kiln body at the curing station. The curing sensor array includes a temperature sensor, a humidity sensor, and a non-contact infrared sensor for monitoring the temperature of the segment body. The temperature sensor, humidity sensor, carbon dioxide concentration sensor, and non-contact infrared sensor are respectively connected to the curing control unit.

4. A smart production line for tunnel segments based on robot and AI control according to any one of claims 1 to 3, characterized in that: A dehumidification system is installed on the side wall of the curing kiln. A steam delivery pipe and a humidification delivery pipe are installed inside the curing kiln at the curing station. A steam heater and a steam control valve are installed on the steam delivery pipe inside the curing kiln, and a humidification control valve is installed on the humidification delivery pipe. The curing control unit includes a controller and a network communication module. The controller is connected to the AI ​​central control unit through the network communication module. The steam heater, steam control valve, humidification control valve, and dehumidification system are respectively connected to the curing control unit.

5. The intelligent production line for tunnel segments based on robot operation and intelligent control according to claim 1, characterized in that: The production scheduling optimization model is used to receive the production data list and generate the production queue of the process station according to the production plan list, and then dynamically adjust the production tasks according to the production queue and perform cooperative motion control of the robot. The collaborative path planning model is used to generate collision-free motion operation paths, operation status and operation time windows for each robot system at each work station, ensuring that each robot system at each work station operates collaboratively and on time and safely according to the prescribed path. The quality prediction model is used to predict, identify and compare the production data of each process station collected by the data acquisition unit with the corresponding pre-stored quality inspection data to identify whether the production data of each process station for producing and processing pipe segments meets the requirements. The adaptive adjustment model is used to adaptively adjust the cycle time and motion path of the robot system at each work station based on the real-time acquired operation status and operation time window of each robot system at each work station, combined with the scheduling trigger instructions generated by the production queue and sent to each robot system. The maintenance optimization model is used to call a preset maintenance parameter model according to the characteristics of each batch of segments, monitor the occupancy status of the static stop area and the kiln exit time, and obtain the distribution of maintenance parameters at different maintenance stages in the maintenance kiln in real time to compare and analyze whether there is a deviation in maintenance parameters. It automatically provides feedback and makes decisions on whether the maintenance parameters need to be dynamically optimized. According to the kiln exit time and the occupancy status of the static stop area, it sends the kiln exit command and kiln entry command of the segment mold to the production scheduling optimization model in advance. The production scheduling optimization model issues scheduling commands to schedule the AGV transfer robot to wait for the cured segments to be transferred to the stacking area or water curing tank for storage and secondary maintenance.

6. The intelligent production line for tunnel segments based on robot operation and intelligent control according to claim 5, characterized in that: The production scheduling optimization model also dynamically adjusts the production queue and production tasks based on the priority of the production plan list, the status of the segment mold, and the performance parameters of the robot system at each work station.

7. A segment intelligent production line based on robot operation and intelligent control according to claim 5 or 6, characterized in that: The production scheduling optimization model also reads the mold status, curing kiln occupancy status, and robot busy / idle status based on the production queue, binds the electronic tag of the segment mold to the priority of the production plan list, and adjusts the production queues of the allocated process stations; the production queues of the process stations include the mold cleaning queue, the rebar placement queue, the pouring queue, the finishing queue, the kiln entry queue, the demolding queue, and the transfer queue, and each production queue carries an expiration date.

8. The intelligent production line for tunnel segments based on robot operation and intelligent control according to claim 5, characterized in that: If dynamic optimization of curing parameters is required, the steam valve opening is adjusted in real time or the curing length is extended / shortened. Based on the current curing parameters, deep learning is used to determine whether there are deviations in the curing parameters within the curing kiln, and real-time feedback and decision-making are made based on the current deviations.

9. A method for intelligent production of tunnel segments based on robot operation and intelligent control, characterized in that: A method for intelligent production of tunnel segments based on robot and AI control, as described in any one of claims 1-8, comprises the following steps: Step 1: Before production begins, the AI ​​central control unit generates a production plan list based on the order information; based on the production plan list, a production data list for each process station is preset; the AI ​​central control unit generates a production queue for each process station based on the production data list for each process station, and issues production tasks to each process station in sequence according to the production queue and dynamically issues running path instructions to each robot system. Step 2, Rebar cage fabrication: Obtain the rebar cage production task from the production data list, generate the fabrication parameters based on the rebar cage production task, automatically grab the pre-specified rebar material from the material area based on the fabrication parameters, automatically bind and weld the rebar processing station, and then transport the bound and welded rebar cage to the track conveyor line according to the production queue through the rebar cage handling robot and send it to the mold preparation station. Step 3, pre-processing and transfer of segment molds: Obtain the segment mold model according to the production data list, retrieve the corresponding segment mold to the mold preparation station according to the segment mold model, perform surface cleaning and release agent spraying pre-treatment instructions on the segment mold in sequence, and then transport the steel reinforcement skeleton sent from the steel reinforcement processing station into the pre-treated segment mold, and transfer it to the concrete pouring and vibration station according to the production queue via the track conveyor line. Step 4, Concrete Pouring and Vibration: Obtain the concrete pouring task from the production data list, and activate RFID or visual recognition to transmit the segment mold to the concrete pouring and vibration station and obtain the mold model. Upload the mold model to the AI ​​central control unit and retrieve the corresponding 3D model data according to the mold model. Calculate the cavity volume and structural features of the current segment mold. Based on the cavity volume and structural features and according to the production queue, execute the preset concrete receiving quantity and receive the concrete command. Pour the concrete into the segment mold at a uniform speed and in a quantitative manner. Vibrate during or after pouring. Dynamically adjust the vibration parameters based on the real-time collected concrete slump data. Step 5, Static Stop and Initial Curing: Obtain the static stop and initial curing tasks according to the production data list, and wait in real time for the completed casting, pouring and vibration of the segment molds to be delivered along the track conveyor line. Obtain the electronic tag of the segment mold and send it into the static stop area of ​​the curing kiln body according to the production queue for parking. Wait for the instruction to be sent into the curing kiln body to execute the initial curing task. After the initial curing is completed in the curing kiln body and the solidified segment model is formed, the segment model carried by the segment mold is moved out of the curing kiln body along the track conveyor line to the segment demolding and operation station. Step 6, Demolding and Secondary Curing: Obtain the concrete demolding and secondary curing tasks from the production data list. Scan the electronic tags of the segment molds delivered after curing in real time according to the production queue. Then execute the demolding operation command on the segment model inside the segment mold. Transfer the demolded segment model to the designated stacking area or water curing tank for storage and secondary maintenance. The empty mold after demolding is returned to the cleaning station to wait for the next process.

10. The intelligent production method for tunnel segments based on robot operation and intelligent control according to claim 8, characterized in that: Whenever the segment mold moves to the next process station, the RFID electronic tag reader of that process station reads the electronic tag of the segment mold and reports the electronic tag of the segment mold and the number of the current process station to the AI ​​central control unit. The AI ​​central control unit updates the location information of the segment mold accordingly and verifies whether the segment mold holds the production queue corresponding to the current station. If the verification is successful, the robot of the current process station is allowed to execute the production task of that station. If the verification fails, an alarm is triggered and the task execution is prevented.

Images