Digital twinning-based air conditioner pipeline intelligent manufacturing and fault detection method

Abstract

The invention discloses an air conditioner pipeline intelligent manufacturing and fault detection method based on digital twinning, relates to the technical field of data center infrastructure and fault detection, and aims to solve the problems of low construction efficiency, serious error accumulation, operation and maintenance logic faults and the like in a traditional process. The system is split into five functional modules through modular assembly design, a prefabricated straight pipe section and a combined section are combined with an equipment decoupling adjusting section to eliminate errors, and a standardized machining drawing is output; the quality is ensured through laser cutting, precise welding and galvanization processes, and high-precision field installation is achieved by combining paper number dual-mode guidance; a valve unique coding system is constructed based on BIM mapping, a two-dimensional code is generated to realize three-dimensional model calling and emergency guidance, a digital twin platform dynamically associates equipment control logic and a spatial position through a logic unit group, pushes hierarchical positioning guidance and links AR technology guidance operation when a fault occurs, and realizes full life cycle data tracing based on a block chain.

Description

Technical Field

[0001] The present invention relates to the field of data center infrastructure and fault detection technology, and in particular to a method for intelligent manufacturing and fault detection of air-conditioning ducts based on digital twins. Background Art

[0002] The next-generation data center air conditioning water cooling system consists of a cooling water system, a chilled water system, and a cooling capacity distribution system. A pipe network connects the cooling tower, chillers, pumps, and terminal equipment. Existing air conditioning piping designs often employ a two-dimensional linear layout. Construction requires on-site cutting of standard pipe fittings, welding, and assembly, followed by disassembly, return to the factory, galvanization, and subsequent reinstallation. This process presents challenges such as a lengthy processing chain, multiple disassembly and assembly that can introduce errors, and a difficult-to-control construction cycle. Furthermore, critical steps, such as galvanizing for corrosion protection, require cross-site operations, leading to quality risks and increased costs.

[0003] Existing technologies have significant limitations: First, traditional BIM models focus on geometric structures while ignoring the dynamic control logic between devices (such as the linkage rules between valves and pumps). Fault detection decisions rely on manual experience, which carries the risk of misoperation. Second, fault location requires manual comparison of drawings and on-site markings, which is inefficient and prone to errors due to model deviations. Third, valve fault detection data is scattered, making full lifecycle traceability impossible. For example, while Chinese invention patent CN115758553B proposes a modular assembly method, it focuses on optimizing the layout of equipment in computer rooms and does not address the error elimination, logic mapping, and fault detection coordination issues unique to air conditioning ducts, making it difficult to meet the precision and intelligence requirements of complex pipe network systems. Summary of the Invention

[0004] The present invention aims to solve the problems of low processing efficiency, serious error accumulation and logical faults in the equipment during the fault detection process of the existing data center air-conditioning piping system caused by traditional two-dimensional design and construction. It realizes high-precision prefabrication through modular assembly design combined with the equipment decoupling adjustment section and flange minimization rules. It adopts digital processing and installation closed-loop control of welding quality and galvanizing process standards, and constructs a valve coding system and digital twin platform based on BIM mapping to realize fault level positioning and intelligent guidance of emergency operations, thereby opening up the data flow of the entire chain of design-construction-fault detection, and overcoming the defects of existing modular technology in the air-conditioning piping scenario, such as lack of logical association and insufficient error control.

[0005] The present invention proposes a method for intelligent manufacturing and fault detection of air-conditioning ducts based on digital twins. The method includes: assigning a globally unique code to the valves and mapping it to the BIM model, defining the valves, ducts, and equipment with control logic associations as a logical unit group, and realizing the dynamic binding of the logical unit group and the three-dimensional visualization model based on the spatial topology table of the digital twin platform; detecting valve fault signals through the air-conditioning group control system and pushing the fault information to the digital twin platform; the digital twin platform analyzes the fault location code and generates a highlighted guide for the fault room at the floor level and a highlighted guide for the fault point at the room level; automatically associating the three-dimensional visualization model group of the fault point and the emergency operation guide, and synchronously pushing them to the fault detection personnel through the terminal.

[0006] Preferably, the globally unique code generated for the valves by this method is marked on the physical valves in the form of two-dimensional codes. After the fault detection personnel scan the codes, the terminal automatically loads the associated information according to the scenario, loads the valve maintenance records during daily inspections, and loads the operation guide containing the linkage rules of the logical unit group during fault emergencies.

[0007] Preferably, this method eliminates errors and modularly splits the system based on the equipment decoupling adjustment section, fully prefabricated straight pipe section, and branch pipe adjustment section, and outputs the prefabrication processing drawing through the flange minimization rule.

[0008] Preferably, this method splits the ducts into multiple functional modules according to the building space and system attributes to which they belong and modularly splits the system, and outputs the component-level prefabrication processing drawing through the physical form classification of the prefabricated straight pipe section and the prefabricated combined section, where the prefabricated combined section includes ≥2 PCM types and 1 PCS type.

[0009] Preferably, according to the handling conditions and installation space limitations corresponding to the module types, this method splits the prefabricated straight pipe section and the combined section, and satisfies the flange minimization and flange spacing constraints: the maximum length of the prefabricated straight pipe section in the chilled water plant is 6 - 8m and the flange is ≥200mm away from the support and hanger; the flange of the prefabricated straight pipe section in the cooling water riser shaft is ≥400mm away from the structural beam and ≥200mm away from the rib plate; the length of the tee connection of the T-shaped combined section in the air-conditioning area of the machine room is ≤150mm; the long side of the horizontal main pipe in the roof cooling tower area is ≤4m.

[0010] Preferably, this method sets the equipment decoupling adjustment section at the equipment interface of the chilled water plant to the main pipe tee, the vertical and horizontal pipe joints, the horizontal pipe joints at the bottom / top of the cooling water riser shaft, the tee of the chilled water riser shaft and the horizontal main pipe joint, and the flange at the end of the branch pipe connecting to the equipment in the air-conditioning area of the machine room according to the error elimination requirements.

[0011] Preferably, this method processes the prefabricated parts through the cycle of blanking - welding - detection, dynamically selects the starting section according to the module type, and combines the paper and digital model guidance for installation.

[0012] Preferably, the method generates prefabricated parts through cutting-assembling-welding, and performs ultrasonic testing on all welds. If the test fails, the process returns to the welding process for cyclic processing. After the test passes, pressure testing and hot-dip galvanizing are performed in sequence, wherein the average thickness of the hot-dip galvanized layer is ≥90μm and the local minimum thickness is ≥80μm.

[0013] Preferably, the method realizes finished product protection by port blocking, the thickness of the blocking wood board is ≥ 1 / 20 of the pipe diameter, and anti-scratch protection measures are taken for the sensor and exhaust valve short pipe on the main pipe section.

[0014] Preferably, the method dynamically selects the starting section according to the module type, the refrigeration station is based on the distribution of the tee or adjustment section of the directly connected equipment, the riser shaft is fixedly started from the bottom section, the end area selects the middle section or the opposite starting point based on the position of the adjustment section, and the installation direction is guided by a paper number dual-mode containing a prefabricated part numbering diagram and a three-dimensional model.

[0015] The present invention effectively solves the problems of low construction efficiency and serious error accumulation in traditional air-conditioning piping systems by combining modular assembly design with digital processing and installation. It uses the equipment decoupling adjustment section and flange minimization rules to achieve high-precision prefabrication, significantly reducing the risks of on-site processing and rework. The valve unique coding system and digital twin platform based on BIM mapping open up the dynamic association between the logical relationship of equipment and spatial entities, support rapid fault location and intelligent emergency guidance, and break through the limitations of reliance on manual experience in traditional fault detection modes. At the same time, the durability and construction quality of the pipeline are ensured through the standardization of the galvanizing process and finished product protection measures. Compared with existing modular technologies, the present invention deeply integrates design, construction and fault detection data streams to provide an innovative solution for collaborative optimization of the entire life cycle of data center air-conditioning piping systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a flow chart of the method of the present invention;

[0017] Figure 2 is a flow chart of the modular assembly design method of the present invention;

[0018] Figure 3 It is a flow chart of the digital processing and installation method of the present invention. DETAILED DESCRIPTION

[0019] Example 1

[0020] according to Figure 1 As shown, the present invention proposes an intelligent manufacturing and fault detection method for air-conditioning ducts based on digital twins, covering logic-space dual-dimensional modeling, QR code interaction and fault level positioning, to achieve full life cycle management from daily inspections to emergency response.

[0021] After the system is put into operation, globally unique codes are first assigned to all valves in the BIM digital twin platform. The coding rule is "VLV - module code - serial number" (for example, the code for the valve at the outlet of the chilled water station is VLV - FS - 005), and a mapping relationship is established with the component information table. The component information table includes valve model, opening and closing logic function, historical maintenance records, and associated equipment control rules (such as the interlock logic between the start and stop of the cooling water pump and the valve). Valves, pipes, and equipment with control logic associations are defined as logical unit groups (such as the flow control group composed of the chilled water circulation pump, outlet regulating valve, and pressure sensor). In the 3D visualization view of the digital twin platform, any change in the state of a component within the logical unit group (such as the adjustment of the valve opening) will trigger the dynamic response of the associated equipment model in real time (such as the synchronous change of the pump speed). Based on the spatial topology table, the platform dynamically binds the logical unit group to the building space. For example, when a valve in the air - conditioning branch pipe of a certain room in the machine room air - conditioning area fails, the valve and the connected pipes in the room - level 3D model are automatically highlighted.

[0022] A waterproof QR code label bound to the BIM component ID is pasted on the surface of each physical valve. The QR code is printed with anti - ultraviolet ink to ensure long - term readability in outdoor environments. During daily inspections, the fault detection personnel use an explosion - proof terminal to scan the QR code of the valve. The terminal automatically loads the 3D visualization model group of the valve, including the BIM models of the valve body, adjacent pipes, and associated equipment (such as pressure sensors), and pushes the maintenance history records, current opening state, and next maintenance date. If the valve is in a fault emergency scenario (such as a system alarm indicating leakage), after scanning the code, the terminal preferentially loads the emergency operation guide: the interface shows the operation sequence of isolating the valve step by step (such as first closing the upstream valve VLV - FS - 005, and then locking the downstream valve VLV - FS - 006), and associates and displays the 3D model of the operation influence range (such as the list of affected air - conditioning terminal equipment after isolation).

[0023] When the air-conditioning group control system detects a valve failure (such as jamming or leakage), it pushes the fault code, location code, and timestamp to the digital twin platform through the OPC UA protocol. After the platform analyzes the location code, it automatically generates a two-level fault guidance model. In the floor-level guidance model, the faulty room (such as Room A12 on the 3rd floor of the machine room air-conditioning area) is highlighted with a red flashing border in the BIM 3D view, and the number of affected devices and the fault level in the room are marked; in the room-level guidance model, the model of the fault point (such as valve VLV-CR-318) flashes red, and the associated pipes and devices (such as the connected chilled water branch pipe and the air-conditioning terminal) are semi-transparently highlighted. At the same time, an emergency operation panel pops up, showing the isolation plan upstream and downstream of the valve, the spare part inventory location, and the maintenance personnel scheduling status. After the fault detection personnel arrive at the scene, the terminal automatically switches to the augmented reality (AR) mode based on the Bluetooth beacon, identifies the valve entity through the camera, and superimposes and displays virtual guiding arrows to guide the wrench rotation direction and torque value (such as rotate counterclockwise by 90° to the closed position, with a torque limit of 80 N·m).

[0024] After the fault handling is completed, the fault detection personnel upload the disposal records (such as the gasket replacement model, bolt tightening torque value) through the terminal. The platform synchronously updates the maintenance history in the valve component information table and triggers the function verification test of the logic unit group (such as simulating the water pump flow rate to meet the standard when the valve is fully open). All fault detection data (including barcode scanning records, fault alarms, operation logs) are stored in the blockchain database, supporting traceability by time, equipment, or responsible person dimensions. For example, if a valve has three similar faults within one year, it is possible to trace back to the welding inspection records in the processing stage, the flange docking deviation during installation, and the historical maintenance measures through the component ID, assisting in root cause analysis.

[0025] In addition, the platform regularly generates fault detection optimization suggestions: based on the valve action frequency and wear data, dynamically adjust the preventive maintenance cycle (such as shortening the maintenance interval of frequently operated valves from 6 months to 4 months); optimize the path planning algorithm of the floor guidance model according to the fault location time statistics (such as preferentially navigating to the room closest to the staircase). All data is closed-loop linked through the digital twin platform with the BIM models in the design and construction links. For example, the location of frequently faulty valves is fed back to the modular assembly design process, promoting the addition of redundant flange clearances in the same type of adjustment section in the next project.

[0026] This embodiment realizes the intelligent upgrade of the air-conditioning water-cooled system fault detection through logic-space dynamic mapping, two-way QR code interaction, and fault level positioning, significantly reducing the risk of manual intervention and improving the fault response efficiency and system reliability.

[0027] In the valve group control scenario of the refrigeration station, the digital twin platform realizes fault self-healing response through dynamic logic unit groups. When the outlet valve of the main cooling water pump is detected to be stuck, the platform automatically activates the backup valve interlocking mechanism: the backup pipeline is highlighted in the three-dimensional model, and a voice prompt is pushed "Switch to the backup circuit, please confirm the pump status". After the fault detection personnel scan the QR code of the faulty valve, the terminal synchronously loads the historical operation records (such as the alarm of excessive opening and closing times in the past three months), and combines the AR interface to overlay virtual operation instructions, dynamically mark the wrench rotation angle and safe operation area (such as red warning to avoid live cabinets). If the valve maintenance involves the coordination of multiple devices (such as closing the upstream filter valve), the terminal automatically associates the QR codes of adjacent devices and generates cross-device operation sequence instructions.

[0028] For complex spatial positioning scenarios, the platform uses a hierarchical topology matching algorithm to improve the accuracy of fault point identification. When the cooling water riser well valve reports a fault: based on the building space grid code, the BIM model is aligned with the physical space coordinate system, and a floor heat map is rendered on the navigation interface (the faulty room is gradient red); the pipeline direction is identified through augmented reality technology, and a virtual indicator arrow is projected on the terminal interface (such as turning right along the corridor to the third column); after arriving at the faulty room, the terminal calls the Bluetooth beacon ranging function, highlights the faulty valve in the AR view, and displays a floating list of maintenance tools (such as No. 24 Allen wrench, sealant model). The closed-loop linkage design of fault detection data enables maintenance records to be automatically written back to the BIM model, triggering preventive maintenance reminders for valves of the same type (such as "spot checks are recommended for valves with the same number in adjacent areas"), forming a full-chain optimization of design-construction-fault detection.

[0029] Example 2

[0030] according to Figure 2 As shown, this invention focuses on the specific process of modular assembly design, deeply refining the entire process from system decomposition and error elimination strategies to the output of prefabrication processing drawings. The system is divided into five independent modules based on the building's spatial layout and functional characteristics: the refrigeration station, cooling water riser shaft, chilled water riser shaft, computer room air conditioning area, and rooftop cooling tower area. The pipes and fittings for each module are factory-prefabricated through the morphology of prefabricated straight pipe sections (PL) and prefabricated combined sections (PCM / PCS).

[0031] As the core area, the refrigeration station module has its pipelines connected to main equipment such as chillers, pumps, and plate heat exchangers. To address the issues of long equipment installation cycles and large interface deviations, a device decoupling adjustment section (RS) is set at the device interface to the main pipe tee. For example, in the pipeline of a series-connected pump, the adjustment section is located between the pump outlet flange and the main pipe tee, with a length 1.2 times the distance from the device interface flange to the tee center, and the installation deviation of the device is compensated by adjusting the clearance of the flange bolt holes. For the interface where the chiller is directly connected to the main pipe, a straight pipe adjustment section with a length 1.1 times the pipe diameter is set downstream of the main pipe tee, and the axis alignment is achieved by fine-tuning the flange. An L-shaped adjustment section is used at the butt joint of vertical and horizontal pipes, with both the horizontal and vertical arms having a length of 800 mm, and adjustable brackets are equipped to allow for an installation error compensation of ±5 mm in the three-dimensional space. The maximum length of the prefabricated straight pipe section is controlled within 8 m to meet the handling conditions of the main channel of the refrigeration station. When splitting, ensure the minimum number of flanges and a distance of ≥200 mm from the support and hanger. The combination of the tee and elbow in the multi-device parallel pipeline is a PCM-type prefabricated combined section, with the outer frame size strictly limited within 6.0 m × 1.0 m × 1.0 m, and the number of the same type of PCM is not less than 2 groups to improve the processing efficiency.

[0032] All the vertical risers in the cooling water riser shaft module are split into prefabricated straight pipe sections, and the single-section length is dynamically designed according to the floor height and the position of the structural beam. The flange is ≥400 mm away from the bottom surface of the structural beam to reserve an operating space for tightening the riser bolts, and at the same time, it is ≥200 mm away from the preset welding plate of the rib plate to avoid interference between the flange insulation layer and the rib plate. An adjustment section with a flange sleeve is set at the butt joint of the horizontal pipes at the bottom and top floors. The clearance between the inner wall of the sleeve and the riser is designed to be 6 mm, and the elevation matching of the horizontal pipe is achieved by fine-tuning the on-site welding. The refrigeration water riser shaft module is split into T-shaped prefabricated combined sections (PCS) at the connection of each floor tee and the horizontal main pipe. The length of the horizontal branch pipe of the tee is uniformly 200 mm, and the length of the vertical main pipe section is calculated according to the distance between the tees of the upper and lower floors to ensure that the distance between adjacent flanges is ≥1.5 times the pipe diameter. The flange of the prefabricated straight pipe section of the horizontal main pipe is ≥200 mm away from the support and hanger. If the pipe section contains a valve, the axis deviation between the valve flange and the pipe flange needs to be controlled within 1.5 mm.

[0033] An adjustment section is provided at the end of the branch pipe of the horizontal main pipe in the computer room air-conditioning area module. The length is 1.2 times the distance from the end flange to the interface of the air-conditioning equipment, and a rubber flexible joint is configured to absorb the displacement caused by equipment vibration. The length of the branch pipe of the T-shaped prefabricated combined section ≤ 150 mm. When assembling, the principle of "the main pipe in the middle and the branch pipes symmetrical" is adopted. The azimuth of the bolt holes on the flange of the branch pipe forms a 45° angle with the axis of the main pipe, which is convenient for adjusting the bolt direction during on-site installation. The horizontal main pipe and branch pipe of the roof cooling tower area module are used as the adjustment section as a whole. The length of the branch pipe reserves a on-site cutting allowance of ±50 mm according to the height of the cooling tower interface. The length of the branch pipe of the T-shaped prefabricated combined section is uniformly 200 mm. The long side of the prefabricated straight pipe section of the main pipe ≤ 4 m, and the flange is ≥ 200 mm away from the structural beam. And each straight pipe section contains at most 1 elbow or valve to reduce complexity.

[0034] In the stage of prefabrication drawing output, the prefabricated parts of each module generate standardized drawings according to the "one module · one table · two numbers · three drawings" system. The three-dimensional view details the spatial form and key dimensions of the prefabricated parts. For example, the included angle accuracy of the tee in the PCM type combined section of the chilled water station is 90° ± 0.5°, and the parallelism error of the flange surface ≤ 0.1 mm / m. The component list table lists the material specifications, weld inspection requirements and weight information. For example, the welds of the prefabricated straight pipe section of the cooling water riser shaft need 100% ultrasonic testing. The numbering rule strictly implements the "module code - type code - type serial number" system. For example, the code of the T-shaped combined section of the chilled water riser shaft is FWL-PCS-012 to ensure uniqueness and traceability. The front view of the three views for processing marks the total length and the flange spacing. The left view clarifies the height of the branch pipe, and the front view stipulates the uniform distribution parameters of the flange bolt holes. For example, the bolt holes of the PCS type combined section in the computer room air-conditioning area are evenly distributed in 8 holes, and the hole diameter error ≤ 0.5 mm.

[0035] To verify the feasibility of the modular design, a pre-assembly test is carried out in the factory. The prefabricated parts of the same module are temporarily fixed in the order of the numbers. A laser theodolite is used to detect the parallelism and coaxiality of the adjacent flanges. The parallelism deviation ≤ 1 mm / m, and the coaxiality deviation ≤ 2 mm. The chilled water station module simulates the installation of the adjustment section at the main equipment interface to verify whether the flange gap adjustment amount ≥ 10 mm to meet the on-site error compensation requirement of ±5 mm. The pre-assembly data is fed back to the BIM model in real time to dynamically optimize the disassembly plan of the subsequent modules. For example, adjust the length of the prefabricated straight pipe section of the cooling water riser shaft to avoid conflicts with the structural beam. Through closed-loop verification and model iteration, ensure that the first-pass qualification rate of on-site installation ≥ 98%, significantly reducing rework and construction period delays.

[0036] Through refined disassembly rules, multi-dimensional error compensation mechanisms and factory pre-assembly verification, this embodiment realizes a high degree of standardization and feasibility of the modular assembly design of air-conditioning pipes, providing an accurate data basis and quality guarantee for subsequent digital processing and fault detection.

[0037] Example 3

[0038] As shown in Figure 3 , this example elaborates in detail the whole process of digital processing and installation, from the factory production of prefabricated components to the closed-loop control of the whole chain of on-site modular assembly. In the prefabrication stage, according to the component-level drawings output by the modular assembly design, the factory uses a laser cutting machine to cut the pipes with high precision. The perpendicularity deviation of the cut is strictly controlled within 0.5°, and the bevel angle is set at 30°±2° according to the welding process requirements. Burrs and oxide layers are removed by mechanical grinding. In the alignment process, the pipes and flanges are fixed by a three-dimensional positioning fixture. The perpendicularity error between the flange end face and the pipe axis is ≤0.1mm / m. The flange bolt holes are evenly distributed with 8 or 12 holes, and the hole diameter tolerance is ≤0.3mm to ensure smooth insertion of bolts during on-site installation. Pulse current control is used for the argon arc welding backing, and there are no defects such as cracks and slag inclusions on the surface. After welding, all welds are immediately subjected to ultrasonic testing. The testing probe scans along the weld length in a zigzag path, and glycerol is selected as the coupling agent to ensure the stability of sound wave conduction. The unqualified welds are marked in red and returned to the welding station for gouging and re-welding until the re-inspection is qualified.

[0039] The prefabricated components that pass the inspection enter the pressure test stage. The test pressure is 1.5 times the design pressure, and the pressure holding time is ≥30 minutes. The test medium is clean water, and the temperature difference between the water temperature and the ambient temperature is controlled within 5°C to avoid dew condensation on the pipe surface interfering with the leakage judgment. After the pressure test is qualified, pickling treatment is carried out on the inner and outer surfaces of the pipe section: the concentration of the hydrochloric acid solution is 10% until the oxide scale completely falls off, and then it is rinsed with high-pressure clean water until neutral, and the surface moisture is dried with compressed air. In the hot-dip galvanizing process, the average thickness of the galvanized layer is ≥90μm, and the local minimum thickness is ≥80μm. After galvanizing, 10% of the pipe sections are randomly inspected with a magnetic thickness gauge. If the thickness of a single piece does not meet the standard, the whole batch is reworked and re-galvanized to ensure that the anti-corrosion performance meets the design requirements.

[0040] In the finished product protection stage, the ports of the galvanized prefabricated components are blocked with birch baffle plates, the thickness of which is ≥1 / 20 of the pipe diameter (for example, the baffle plate thickness of a DN150 pipe section is ≥7.5mm), and the edges are tightly fixed with stainless steel straps to prevent falling off during handling. The sensor stub pipes and exhaust valve interfaces on the main pipe section are wrapped with 5mm thick foam plastic sheaths to avoid scratching the galvanized layer during transportation. Waterproof QR code labels are pasted on the surface of the prefabricated components. The label information includes the component number (such as FS-PCM-005), processing date, and quality inspection personnel code. The tensile strength of the label is ≥10N, and the water immersion resistance time is ≥72 hours to ensure that the information is readable in the harsh on-site environment.

[0041] During the on-site installation phase, the construction team worked based on the modular installation starting section (SP) positioning diagram and directional guidance rules generated by the BIM model. The chiller station module began with the tee directly connected to the equipment, such as the chiller outlet main tee (numbered FS-SP-001). Installation extended along the adjusting section (RS) toward both ends. A laser level was used to calibrate the axis deviation of adjacent prefabricated flanges to ≤1mm / m. Bolt tightening torque was set to 120N·m±5% as per design requirements. The cooling water riser module began with the prefabricated straight pipe section (numbered CWL-PL-001) at the bottom floor and was installed upwards in sections. Verticality was verified using a total station after every three sections, with cumulative deviation ≤5mm / 30m. Any deviation was compensated for by the gap between the adjusting section flanges. The horizontal main pipe in the computer room air conditioning area began with the middle section (numbered CR-PL-012) and was installed symmetrically. The gap between the branch pipe flange and the air conditioning equipment interface was compensated by rubber flexible joints in the adjusting section, allowing for ±8mm radial displacement to absorb vibration caused by equipment operation.

[0042] During the construction process, the mobile terminal retrieves the digital assembly work order in real time, and synchronously displays the three-dimensional model of the prefabricated parts and the installation sequence number. For example, when installing the PCM-type combined section of the refrigeration station (number FS-PCM-008), the terminal highlights the azimuth angle of the docking flange of the adjacent prefabricated part FS-PL-009, and dynamically prompts that the bolt tightening sequence is "alternating diagonally and loading in three times". The paper assembly drawing provides the elevation of the prefabricated parts and the positioning information of the supports and hangers. For example, the center distance of the supports and hangers of the cooling water riser shaft PL-005 is 4.8m, the elevation error is ≤±3mm, and the preload force of the support and hanger bolts is calibrated to 80N·m. After the installation is completed, the flange gap of the adjustment section (RS) is tested with a feeler gauge, and the gasket thickness is adjusted so that the gap is evenly distributed within the range of 1 to 2mm to ensure that there are no gaps in the subsequent insulation layer construction.

[0043] After each module is installed, the construction data is uploaded to the digital twin platform via a mobile terminal, including flange docking deviation, bolt torque value, and adjustment section compensation. The platform automatically compares the theoretical data of the BIM model and generates an installation quality report. For example, after the installation of the chilled water riser well module, the system detected that the elevation deviation of the three-way branch pipe numbered FWL-PCS-015 was +4mm, which automatically triggered an early warning and pushed an adjustment plan: add 2mm gasket compensation when installing the next section of prefabricated parts. The processing inspection data, galvanizing records, and installation parameters of all prefabricated parts are stored in the blockchain database, supporting full life cycle traceability. If a section of the pipeline leaks during the fault detection period, the component number can be traced back to the welder number, inspection record, and installation torque value during processing to accurately locate the responsible link.

[0044] In this embodiment, through digital process control, multi-dimensional quality inspection, and dual-mode coordination guidance of paper count, high-precision processing and efficient installation of air-conditioning duct prefabricated parts are achieved, providing a reliable data basis and quality guarantee for intelligent fault detection after the system is put into operation.

Claims

1. A method for intelligent manufacturing and fault detection of air-conditioning ducts based on digital twins, characterized in that, The method comprises: Assign globally unique codes to valves and map them to BIM models. Define valves, pipelines, and equipment with control logic associations as logical unit groups. Dynamically bind logical unit groups to 3D visualization models based on the spatial topology table of the digital twin platform. Detect valve fault signals through the air conditioning group control system and push the fault information to the digital twin platform; The digital twin platform parses the fault location code and generates floor-level fault room highlight guides and room-level fault point highlight guides; it automatically associates the fault point 3D visualization model group and emergency operation instructions, and simultaneously pushes them to fault detection personnel through the terminal.

2. The method for manufacturing and fault detection of air-conditioning ducts based on digital twin according to claim 1, characterized in that, The method is to mark the globally unique code generated by the valve in the form of a QR code on the physical valve. After the fault detection personnel scan the code, the terminal automatically loads related information according to the scenario, loads valve maintenance records during daily inspections, and loads operation instructions containing logical unit group linkage rules during fault emergencies.

3. The method for manufacturing and fault detection of air-conditioning pipelines based on digital twin according to claim 1, characterized in that, The method eliminates errors and modularizes the system based on the equipment decoupling adjustment section, fully prefabricated straight pipe section and branch pipe adjustment section, and outputs the prefabrication processing drawing through the flange minimization rule.

4. A method for manufacturing and fault detection of air-conditioning ducts based on digital twins according to claim 1 or 3, characterized in that The method splits the pipeline into multiple functional modules and modularizes the system according to the building space and system properties. The component-level prefabrication processing drawings are output based on the physical morphology classification of prefabricated straight pipe sections and prefabricated combination sections, where the prefabricated combination sections include ≥2 PCM types and 1 PCS type.

5. A method for manufacturing and fault detection of air-conditioning ducts based on digital twins according to claim 1 or 3, characterized in that, The method splits prefabricated straight pipe sections and combined sections according to the transportation conditions and installation space restrictions corresponding to the module types, and meets the constraints of flange minimization and flange spacing: the maximum length of the prefabricated straight pipe section of the refrigeration station is 6 to 8 meters, and the flange distance to the support bracket is ≥200 mm; the flange distance to the prefabricated straight pipe section of the cooling water riser shaft is ≥400 mm from the structural beam and ≥200 mm from the rib plate; the length of the T-shaped combined section of the three-way pipe in the computer room air-conditioning area is ≤150 mm; and the long side of the horizontal main pipe in the roof cooling tower area is ≤4 m.

6. A method for intelligent manufacturing and fault detection of air-conditioning ducts based on digital twins according to claim 1 or 3, characterized in that, The method sets the equipment decoupling adjustment section at the tee of the refrigeration station equipment interface to the main pipe, the vertical and horizontal pipe joints, the bottom / top horizontal pipe joints of the cooling water riser well, the joints between the tee of the chilled water riser well and the horizontal main pipe, and the terminal flange of the branch pipe connected to the equipment in the computer room air conditioning area according to the error elimination requirements.

7. A method for intelligent manufacturing and fault detection of air-conditioning ducts based on digital twin according to claim 1, characterized in that, The method processes prefabricated parts through a blanking-welding-inspection cycle, dynamically selects a starting section according to the module type, and combines paper-number dual-mode guided installation.

8. A method for manufacturing and fault detection of air-conditioning ducts based on digital twins according to claim 1 or 7, characterized in that, The method generates prefabricated parts through blanking, assembly and welding, and performs ultrasonic testing on all welds. If the test fails, the process returns to the welding process for cyclic processing. After the test passes, pressure testing and hot-dip galvanizing are performed in sequence, wherein the average thickness of the hot-dip galvanizing layer is ≥90μm and the local minimum thickness is ≥80μm.

9. A method for manufacturing and fault detection of air-conditioning ducts based on digital twin according to claim 1 or 7, characterized in that, The method protects the finished product by blocking the port, the thickness of the blocking wood board is ≥ 1 / 20 of the pipe diameter, and anti-scratch protection measures are taken for the sensor and the exhaust valve short pipe on the main pipe section.

10. A method for manufacturing and fault detection of air-conditioning ducts based on digital twin according to claim 1 or 7, characterized in that, The method dynamically selects the starting section according to the module type. The refrigeration station is based on the direct-connected equipment three-way valve or the regulating section distribution. The riser shaft is fixed to start from the bottom section. The end zone selects the middle section or the opposite starting point based on the position of the regulating section, and guides the installation direction through the paper-digital dual-mode including the prefabricated part number drawing and the 3D model.

Images