Urban construction project whole-process quality safety closed-loop supervision and tracing platform and method

Abstract

The invention discloses an urban construction project whole-process quality safety closed-loop supervision and tracing platform and method, and relates to the field of data tracing, and the method comprises the steps: constructing a project whole-life-cycle digital twin based on urban construction project data, and presetting a quality safety standard parameter set; implanting an Internet of Things traceable unit in the urban construction project key entity, and mapping the association relationship and the entity to a digital twinborn body; processing data acquired by the sensor network to generate a structured engineering state data set; an anomaly detection model is constructed based on CNN and LSTM, and when data deviates from a threshold value, early warning is triggered and a responsibility chain is traced; generating a processing instruction tree, and distributing the processing instruction tree to a responsibility subject; and the responsibility subject uploads a disposal process and a result, binds the disposal process and the result with an early warning event ID after Hash encryption, and stores the disposal process and the result in a traceability database. The method has the advantages that efficient quality safety supervision and tracing of the whole urban construction project process are achieved through the digital twinborn bodies, and the project safety and management efficiency are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of data traceability, particularly to a city construction project whole-process quality and safety closed-loop supervision and traceability platform and method. BACKGROUND

[0002] City construction project quality and safety problems have always been a key challenge in urban construction, especially in the case of long construction period and high project complexity, how to ensure the safety and quality of each link has become the focus of the industry. With the development of information technology and intelligent technology, traditional quality and safety management methods have been unable to meet the needs of modern city construction projects, and how to realize whole-process quality and safety supervision has become a new topic. Closed-loop supervision and traceability methods have emerged, aiming to integrate information from engineering design, construction, acceptance and other stages to realize whole-process monitoring and management and ensure the traceability of quality and safety at each link.

[0003] The current city construction project whole-process quality and safety closed-loop supervision and traceability methods on the market generally have defects such as information island, untimely data, and inaccurate supervision. Traditional methods usually rely on manual or partial monitoring means, lack of real-time and comprehensiveness, often leading to difficulty in fully grasping the real-time state of the project and the risk of lag. Many methods fail to fully integrate BIM, GIS, Internet of Things and artificial intelligence and other cutting-edge technologies, lacking comprehensive digital means to realize precise management of the project's whole life cycle. In addition, some existing traceability systems rely heavily on traditional manual recording and management, making it difficult to guarantee the accuracy and timeliness of the data and to achieve immediate traceability of the responsibility chain. In terms of early warning and abnormal handling, many methods still rely on simple monitoring based on rules, making it difficult to effectively identify abnormalities through deep learning and intelligent analysis, and unable to timely discover potential safety hazards, increasing the risk of accidents. SUMMARY

[0004] To improve existing platforms and methods, a city construction project whole-process quality and safety closed-loop supervision and traceability platform and method is provided, which realizes efficient quality and safety supervision and traceability of city construction projects throughout the whole process through digital twins, Internet of Things, edge computing and artificial intelligence technology, ensuring project transparency, real-time monitoring and responsibility traceability, and improving project safety and management efficiency.

[0005] To achieve the above purpose, the technical solution adopted by the present application is:

[0006] The city construction project whole-process quality and safety closed-loop supervision and traceability method comprises:

[0007] Based on city construction project BIM data, design drawings, construction specifications and geographic information system data, a digital twin of the project's whole life cycle is constructed, and a set of quality and safety standard parameters is preloaded in the digital twin;

[0008] The Internet of Things traceable unit is implanted in the key entity of the urban construction project, and the correlation and entity are mapped to the digital twin;

[0009] Based on the edge computing gateway, the construction data, environmental parameters and personnel position data collected by the sensor network in real time are integrated, and the structured engineering state data set is generated after spatio-temporal alignment processing;

[0010] Based on the convolutional neural network and the long short-term memory network, an anomaly detection model is constructed, when the real-time data deviates from the standard threshold, the warning is triggered and the responsibility chain of the associated traceable unit is traced back;

[0011] Based on the warning level and the responsibility chain tracing result, a disposal instruction tree is generated, which is distributed to the responsible subject through the smart contract in the blockchain network;

[0012] The responsible subject uploads the disposal process and result through the mobile terminal, which is bound to the warning event ID after being hashed and encrypted, and stored in the traceability database.

[0013] Preferably, the digital twin of the whole life cycle of the project is constructed based on the BIM data, design drawings, construction specifications and geographic information system data of the urban construction project, and the quality and safety standard parameter set is preloaded in the digital twin, which specifically includes:

[0014] The BIM model is spatially associated with the geographic information system data to check the consistency of the geographic position and the actual site height of the BIM model;

[0015] The BIM data, design drawings, construction specifications and geographic information system data of the urban construction project are centralized through the BIM platform to form a complete digital model;

[0016] Based on the complete digital model, a dynamic digital twin is constructed in combination with the life cycle of each stage of design, construction, operation and maintenance;

[0017] Based on the construction specifications and quality standards, the specific quality and safety standard parameter set is obtained and embedded in the digital twin.

[0018] Preferably, the Internet of Things traceable unit is implanted in the key entity of the urban construction project, and the correlation and entity are mapped to the digital twin, which specifically includes:

[0019] The key entity components that need to be traced in the project are obtained, including steel bars, concrete, walls, doors and windows, and a unique identification QR code is added to each key entity component;

[0020] The Internet of Things traceable unit uniquely associates the material batch number, equipment unique ID and personnel identity information of the corresponding entity;

[0021] All collected Internet of Things data is associated with corresponding entities, and the real-time state of all entities and the associated relationship is mapped into the digital twin in real time, and the digital twin updates the model in real time based on data.

[0022] Preferably, the edge computing gateway integrates construction data, environmental parameters and personnel location data collected by the sensor network in real time, and generates structured engineering state data set after spatio-temporal alignment processing, specifically including:

[0023] The edge computing gateway is deployed in the core area of the construction site, covering all key facilities, equipment and personnel positions.

[0024] Real-time construction data, environmental parameters and personnel location data are collected by multi-source sensors, and all collected real-time data are time-synchronized through the network time protocol.

[0025] The data of different sensors are fused and processed, and the data processed by spatio-temporal alignment and fusion are converted into structured engineering state data set, including construction state, environmental parameters, structural response, equipment operation and personnel location.

[0026] Preferably, the abnormal detection model is constructed based on convolutional neural network and long short-term memory network, and when the real-time data deviates from the standard threshold, the warning is triggered and the responsibility chain of the associated traceable unit is traced back, specifically including:

[0027] Based on the spatial distribution and response state of each sensor, the engineering state data are obtained, the spatial features in the engineering state data are extracted through convolution operation in the convolutional neural network, and the data dimension is reduced through the pooling layer.

[0028] The spatial features extracted by the convolutional neural network are combined with the time dimension of the engineering state data, and the data change trend and dependency between time points are learned through the LSTM network.

[0029] The LSTM adjusts the memory and forgetfulness of historical data through the gating mechanism, and obtains the time sequence abnormal data in the construction process.

[0030] Based on the data set processed by the convolutional neural network and the LSTM network, an abnormal detection model is constructed, and the model is trained.

[0031] The real-time input engineering state data are compared with the trained model data, and when the data deviates from the preset threshold of the quality safety standard parameter set, the warning is triggered, and detailed alarm information data are generated.

[0032] Based on the Internet of Things traceable unit, the specific responsible person and responsible subject are traced through the mapping relationship of the Internet of Things.

[0033] Preferably, the treatment instruction tree is generated based on the early warning level and the responsibility chain tracing result, and is distributed to the responsibility subject through a smart contract in the blockchain network, specifically comprising:

[0034] Based on the early warning level and the traced responsibility person and responsibility subject, an instruction tree containing different treatment suggestions is generated, and the instruction tree is divided into multiple levels, each level corresponding to different treatment measures, including shutdown rectification instructions, design change suggestions and process optimization schemes;

[0035] The distribution of instructions is automatically executed through a smart contract in the blockchain network, and the multi-layer treatment instruction tree is transmitted to the responsibility person and the responsibility subject, and each instruction is bound to the responsibility subject through a smart contract in the blockchain;

[0036] The smart contract records the execution of each instruction and feeds back the processing progress.

[0037] Preferably, the responsibility subject uploads the treatment process and result through a mobile terminal, binds the hash-encrypted data to the early warning event ID, and stores it in the tracing database, specifically comprising:

[0038] The responsibility subject uploads each step, result verification data and supporting files in the treatment process through a mobile terminal;

[0039] The uploaded data is verified for data integrity and content validity, and a unique digital fingerprint is generated for the successfully uploaded data through a hash algorithm;

[0040] Each early warning event is bound to a unique identifier, and the generated unique digital fingerprint is bound to the early warning event identifier;

[0041] The hash-encrypted and bound data is uploaded and stored in the tracing database to form a complete and tamper-proof evidence chain.

[0042] Further, a whole-process quality and safety closed-loop monitoring and tracing platform for urban construction projects is proposed, comprising:

[0043] Digital twin module: the module is used to integrate various data and construct a digital twin of the whole life cycle of urban construction projects, and a set of quality and safety standard parameters is preset;

[0044] Internet of Things tracing module: the module implants an Internet of Things traceable unit in key entities, traces the material, equipment and personnel information of each key entity component through a two-dimensional code, and maps it to the digital twin;

[0045] Data acquisition module: the module acquires construction data, environmental parameters and personnel positions in real time through an edge computing gateway and a sensor network, and generates a structured engineering state data set;

[0046] Abnormality detection module: This module combines convolutional neural networks and LSTM networks to detect abnormalities based on real-time data compared with standard thresholds, trigger early warnings, and trace back to the chain of responsibility;

[0047] Instruction tree generation module: This module generates a treatment instruction tree based on the early warning level and the chain of responsibility trace results, and automatically distributes instructions to responsible subjects through smart contracts in the blockchain;

[0048] Traceability database module: This module uses hash encryption to bind with early warning events and stores treatment process data in the traceability database;

[0049] Processor: The processor is used to process the calculation process of each formula and the construction calculation process of each model.

[0050] Compared with the prior art, the advantages of the present application are:

[0051] The digital twin constructed based on BIM and geographic information system data provides a virtual mapping of the entire life cycle of the project, which can reflect the actual state of the project in real time and embed quality and safety standards, ensuring effective supervision at each stage from design, construction to operation and maintenance. Secondly, the application of the Internet of Things traceable unit enables accurate tracking of material, equipment and personnel information for each key entity, timely detection of potential problems and chain of responsibility tracing. Through real-time data collection by edge computing gateways and sensors, as well as abnormality detection by convolutional neural networks and LSTM models, the system can accurately identify safety hazards in the construction process and immediately trigger early warnings to prevent accidents. Finally, combined with blockchain technology, the use of smart contracts automatically distributes treatment instructions and forms an unalterable evidence chain, ensuring the execution and tracking of each responsible subject, and improving the transparency of information and the traceability of responsibility. BRIEF DESCRIPTION OF DRAWINGS

[0052] Figure 1 Method diagram for the present application;

[0053] Figure 2 Digital twin construction diagram for the present application;

[0054] Figure 3 Internet of Things traceable unit diagram for the present application;

[0055] Figure 4 Engineering state data set generation diagram for the present application;

[0056] Figure 5 Abnormality detection model construction diagram for the present application;

[0057] Figure 6 Treatment instruction tree generation diagram for the present application;

[0058] Figure 7 The traceability database schematic diagram proposed by the present application. DETAILED DESCRIPTION

[0059] The following description is used to disclose the present application to enable a person skilled in the art to implement the present application. The preferred embodiments in the following description are only as examples, and other obvious modifications can be thought of by a person skilled in the art.

[0060] The city construction project whole-process quality and safety closed-loop supervision traceability platform comprises:

[0061] The digital twin module is used for integrating various data and constructing the digital twin of the whole life cycle of the city construction project, and a quality and safety standard parameter set is preset;

[0062] The Internet of Things traceability module implants the Internet of Things traceable unit in the key entity, traces the material, equipment and personnel information of each key entity component through the two-dimensional code, and maps to the digital twin;

[0063] The data acquisition module acquires the construction data, environmental parameters and personnel positions in real time through the edge computing gateway and the sensor network, and generates the structured engineering state data set;

[0064] The anomaly detection module combines the convolutional neural network and the LSTM network, compares the real-time data with the standard threshold, performs anomaly detection and triggers the early warning, and traces to the responsibility chain;

[0065] The instruction tree generation module generates the disposal instruction tree according to the early warning level and the responsibility chain traceability result, and automatically distributes the instructions to the responsible subjects through the smart contract in the blockchain;

[0066] The traceability database module binds the disposal process data to the early warning event using the hash encryption, and stores the disposal process data into the traceability database;

[0067] The processor is used for processing the calculation process of each formula and the construction calculation process of each model.

[0068] Referring to Figure 1 As shown in the figure, the city construction project whole-process quality and safety closed-loop supervision traceability method comprises:

[0069] Step one: based on the city construction project BIM data, design drawings, construction specifications and geographic information system data, constructing the engineering whole life cycle digital twin, and presetting the quality and safety standard parameter set in the digital twin;

[0070] Step two: implanting the Internet of Things traceable unit in the key entity of the city construction project, mapping the correlation and entity to the digital twin;

[0071] Step three: Based on the edge computing gateway, integrate the construction data, environmental parameters and personnel location data collected by the sensor network in real time, and generate structured engineering state data set after spatio-temporal alignment processing;

[0072] Step four: Based on convolutional neural network and long short-term memory network, construct an anomaly detection model, when the real-time data deviates from the standard threshold, trigger an early warning and trace the responsibility chain of the associated traceable unit;

[0073] Step five: Based on the early warning level and the responsibility chain tracing result, generate a disposal instruction tree, which is distributed to the responsible subject through the smart contract in the blockchain network;

[0074] Step six: The responsible subject uploads the disposal process and result through the mobile terminal, which is bound to the early warning event ID after hash encryption and stored in the traceability database.

[0075] Referring to Figure 2 As shown, based on the BIM data of urban construction projects, design drawings, construction specifications and geographic information system data, a digital twin of the whole life cycle of the project is constructed, and a set of quality and safety standard parameters is preloaded in the digital twin, including:

[0076] The BIM model is spatially associated with the geographic information system data to check the consistency of the geographic location and actual site height of the BIM model;

[0077] The BIM data of urban construction projects, design drawings, construction specifications and geographic information system data are centralized through the BIM platform to form a complete digital model;

[0078] Based on the complete digital model, a dynamic digital twin is constructed in combination with the life cycle of each stage of design, construction, operation and maintenance;

[0079] Based on the construction specifications and quality standards, a set of specific quality and safety standard parameters is obtained and embedded in the digital twin.

[0080] Specifically, the geographic location coordinates in the BIM model are matched with the geographic coordinates in the GIS system to verify whether the BIM model accurately reflects the actual geographic location, and the height data in the BIM model is compared with the terrain height in the GIS data for consistency check;

[0081] The data is imported into the BIM platform, and the data is integrated through the platform to form a complete digital model. Based on the digital model, the relevant data of the design stage, construction stage and operation and maintenance stage is integrated into the model, including the design drawings and structural analysis data of the design stage, the construction progress, construction quality and site management data of the construction stage, and the equipment operation status, maintenance history and environmental monitoring data of the operation and maintenance stage. Through the integration of real-time monitoring data, the state of the digital twin is updated;

[0082] According to the relevant building construction standards and quality control requirements, the quality and safety standards in the construction process are collected, including the seismic standards, fire safety specifications and material strength requirements. The quality and safety standards and construction specifications are converted into a parameter set and embedded in the digital twin. Whether each stage meets the relevant standards is checked through the digital model.

[0083] Referring to Figure 3 The Internet of Things traceable unit is implanted in the key entities of the urban construction project, and the correlation and entities are mapped to the digital twin, specifically including:

[0084] The key entity components that need to be traced in the project are obtained, including steel bars, concrete, walls and doors and windows. A unique identification QR code is added to each key entity component;

[0085] The Internet of Things traceable unit uniquely associates the material batch number, equipment unique ID and personnel identity information of the corresponding entity;

[0086] All collected Internet of Things data is associated with the corresponding entity, and the real-time state and correlation of all entities are mapped to the digital twin in real time. The digital twin updates the model in real time based on the data.

[0087] Specifically, according to the engineering design and construction requirements, the key components that need to be traced are determined, such as steel bars, concrete, walls and doors and windows. These components usually involve quality and safety control, material traceability and construction process requirements;

[0088] A unique QR code is generated for each key component. Each QR code contains basic information about the component, such as component ID, material batch number, production date and installation location. The generated QR code label is pasted on the corresponding component to ensure that each component can be scanned and traced through the QR code;

[0089] The relevant component data is collected in real time through Internet of Things devices, and these data is associated with the QR code. Each time the component state changes, the construction progress, construction personnel information, etc. are collected in real time through the Internet of Things devices. Internet of Things sensors continuously collect data to reflect the health status of each component in real time. Whenever the state of a component changes, the system will provide real-time feedback;

[0090] The digital twin is a virtual model established based on BIM and Internet of Things data, which maps the information of all key components into the digital twin, including the state, position, and association with other components of each component, which are reflected in the twin in real time.

[0091] Referring to Figure 4 As shown, based on the edge computing gateway, the construction data, environmental parameters, and personnel location data collected by the integrated sensor network in real time are processed for spatio-temporal alignment to generate a structured engineering state data set, which specifically includes:

[0092] The edge computing gateway is deployed in the core area of the construction site, covering all key facilities, equipment, and personnel locations.

[0093] Real-time construction data, environmental parameters, and personnel location data are collected by multiple sensors, and all collected real-time data are time-synchronized through the Network Time Protocol.

[0094] The data from different sensors are fused and processed, and the data after spatio-temporal alignment and fusion are converted into structured engineering state data sets, including construction state, environmental parameters, structural response, equipment operation, and personnel location.

[0095] Specifically, the edge computing gateway is deployed in the core area of the construction site to ensure that the real-time data of all key facilities, equipment, and personnel can be captured and processed. The edge computing gateway should cover the entire construction area, especially for high-risk areas or important equipment for key monitoring, using the Network Time Protocol for time synchronization to ensure that the collection times of all sensors are accurate and consistent.

[0096] The data from different sensors are spatio-temporally aligned and fused by weighted average method or least squares method, and the fused and time-synchronized sensor data are converted into structured engineering state data sets, including: construction state: such as construction progress, construction quality, and construction safety data; environmental parameters: such as temperature, humidity, and air quality environmental monitoring data; structural response: such as building vibration, inclination, and deformation sensor data; equipment operation: such as equipment temperature, load, and power consumption data; personnel location: real-time monitoring of personnel location on the construction site to ensure personnel safety.

[0097] The original sensor data are preliminarily processed by the edge computing gateway and converted into a unified engineering state data set, which will be updated in real time. When the sensor data changes, the system will automatically update and feedback the data.

[0098] Referring to Figure 5As shown, an anomaly detection model is constructed based on a convolutional neural network and a long short-term memory network. When real-time data deviates from a standard threshold, a warning is triggered and the responsibility chain of the associated traceable unit is traced back. The responsibility chain specifically includes:

[0099] Based on the spatial distribution and response state of each sensor, engineering state data is obtained. Spatial features in the engineering state data are extracted through convolution operations in the convolutional neural network, and the data dimension is reduced through the pooling layer.

[0100] The spatial features extracted by the convolutional neural network are combined with the time dimension of the engineering state data, and the data change trend and dependency between time points are learned through the LSTM network.

[0101] The LSTM adjusts the memory and forgetfulness of historical data through a gating mechanism to obtain time sequence anomaly data in the construction process.

[0102] Based on the data set processed by the convolutional neural network and the LSTM network, an anomaly detection model is constructed for model training.

[0103] The real-time input engineering state data is compared with the trained model data. When it is detected that the data deviates from the preset threshold of the quality safety standard parameter set, a warning is triggered, and detailed alarm information data is generated.

[0104] Based on the Internet of Things traceable unit, the specific person and subject responsible are traced through the mapping relationship of the Internet of Things.

[0105] Specifically, in the spatial dimension, the convolutional neural network processes the sensor data through a convolution kernel to extract patterns with local features, including locally weighted summation of data collected by multiple sensors using a convolution kernel to capture spatial features. The convolution operation formula is:

[0106]

[0107] Where F exout is the input sensor data matrix, K(m,n) is the convolution kernel, F exin (i+m,j+m) is an element in the output feature map, and m,n are the position indices of the convolution kernel.

[0108] Engineering state data not only has spatial features, but also has a time dimension. Through the LSTM network, combined with time series data, the model can learn the change trend and dependency of data over time. The LSTM adjusts the memory and forgetfulness of historical data through a gating mechanism to optimize the learning of time sequence data. It has three main gates: the forget gate determines the forgetting of the current input information; the input gate controls the storage of new information; and the output gate determines the final output information

[0109] An integrated anomaly detection model is constructed by combining the spatial features extracted by the convolutional neural network and the time series features learned by the LSTM network. The model can identify abnormal patterns in the construction process by learning the features in the normal construction state.

[0110] Compare the real-time data with the standard data output by the trained model to determine whether the quality and safety standard parameter set has deviated from the preset threshold. If the abnormal determination result is greater than the set threshold, trigger an early warning, and trace back to the specific construction component, equipment, and operator through the Internet of Things traceable unit.

[0111] Referring to Figure 6 The treatment instruction tree is generated based on the early warning level and the responsibility chain traceability result, and is distributed to the responsible subjects through the smart contract in the blockchain network, including:

[0112] Based on the early warning level and the traced responsibility person and responsible subject, an instruction tree containing different treatment suggestions is generated, and the instruction tree is divided into multiple levels, each level corresponding to different treatment measures, including stop work rectification instruction, design change suggestion and process optimization scheme;

[0113] The distribution of instructions is automatically executed through the smart contract in the blockchain network, and the multi-layer treatment instruction tree is transmitted to the responsibility person and the responsible subject. Each instruction is bound to the responsible subject through the smart contract in the blockchain;

[0114] The smart contract records the execution of each instruction and feeds back the processing progress.

[0115] Specifically, based on the early warning level of the event, each event is classified and traced to the responsible person and the responsible subject. The early warning level includes:

[0116] First-level early warning: highest risk, immediate forced measures are required;

[0117] Second-level early warning: higher risk, requires emergency treatment but can take slow measures;

[0118] Third-level early warning: moderate risk, treatment can be moderately delayed, but still needs to be taken;

[0119] Fourth-level early warning: low risk, only needs to be concerned and tracked;

[0120] According to the early warning level and the result of responsibility traceability, a multi-level instruction tree is generated, the root node of the tree represents the initial risk assessment, each layer corresponds to different treatment measures, the root node assesses the severity of the event, determines the early warning level, and notifies the relevant responsible person and responsible subject, the second layer further refines the treatment measures according to the early warning level, and the third layer analyzes and optimizes the third and fourth levels of early warning;

[0121] The distribution of instructions is automatically executed by a blockchain smart contract. Each instruction will be coded into the rules of the blockchain smart contract and bound to the relevant responsible person and subject. The contract will include the type of instruction, the responsible person, the subject, the time node, and the execution progress information. Each instruction has a unique identifier and can be traced. The contract automatically adjusts the instruction content according to the warning level and distributes it to the relevant responsible parties through the smart contract.

[0122] Referring to Figure 7 As shown, the subject uploads the disposal process and results through the mobile terminal, and the data is bound to the warning event ID after being hashed and encrypted, and is stored in the traceability database. The traceability database specifically includes:

[0123] The subject uploads each step in the disposal process, result verification data, and supporting files through the mobile terminal;

[0124] The uploaded data is verified for data integrity and content validity. A unique digital fingerprint is generated for the successfully uploaded data through a hash algorithm;

[0125] Each warning event is bound to a unique identifier, and the generated unique digital fingerprint is bound to the warning event identifier;

[0126] The hashed and bound data is uploaded and stored in the traceability database to form a complete and tamper-proof evidence chain.

[0127] Specifically, the successfully uploaded data is hashed and encrypted to generate a unique digital fingerprint. The hash algorithm is SHA-256, and the formula is:

[0128] H(D) = extSHA-256(D)

[0129] The generated hash value is a fixed-length string, and for the same input, the hash value is unique, which ensures the uniqueness and tamper resistance of the data. A unique identifier is generated for each warning event, and the hash fingerprint of the data is bound to the identifier of the warning event to ensure the unique association between the data and the warning event;

[0130] The bound data is uploaded and stored in the traceability database. The database will save the relevant information and encrypted data of each warning event to form a complete evidence chain. Each uploaded data generates a unique fingerprint through a hash algorithm, and is associated with the event through the binding identifier. The uploaded data forms a tamper-proof record in the traceability database.

[0131] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0132] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

[0133] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A closed-loop supervision and traceability method for the quality and safety of the entire urban construction project process, characterized by: include: Based on urban construction project BIM data, design drawings, construction specifications and geographic information system data, a digital twin of the project's entire life cycle is constructed, and a set of quality and safety standard parameters is pre-set in the digital twin; Embed IoT traceability units in key entities of urban construction projects, and map relationships and entities into digital twins; Based on the edge computing gateway, the construction data, environmental parameters and personnel location data collected in real time by the sensor network are integrated and processed in time and space to generate a structured engineering status dataset; An anomaly detection model is built based on convolutional neural networks and long short-term memory networks. When real-time data deviates from the standard threshold, an early warning is triggered and the responsibility chain of the associated traceable units is traced. Generate a disposal instruction tree based on the warning level and responsibility chain tracing results, and distribute it to the responsible parties through smart contracts in the blockchain network; The responsible party uploads the disposal process and results through the mobile terminal, which are bound to the warning event ID after hash encryption and stored in the traceability database.

2. The closed-loop supervision and tracing method for the whole process quality and safety of urban construction projects according to claim 1 is characterized in that: The construction of a digital twin of the entire project life cycle based on urban construction project BIM data, design drawings, construction specifications and geographic information system data, and the pre-setting of quality and safety standard parameter sets in the digital twin specifically include: Spatially associate the BIM model with GIS data to check the consistency between the BIM model's geographic location and the actual location; Centralize urban construction project BIM data, design drawings, construction specifications, and geographic information system data through the BIM platform to form a complete digital model; Based on a complete digital model, a dynamic digital twin is constructed by combining the lifecycle of each stage of design, construction, operation and maintenance; Based on construction specifications and quality standards, obtain specific quality and safety standard parameter sets and embed them into the digital twin.

3. The closed-loop supervision and tracing method for the whole process quality and safety of urban construction projects according to claim 1 is characterized in that: The aforementioned steps of embedding IoT traceability units in key entities of urban construction projects and mapping relationships and entities to digital twins specifically include: Obtain key physical components that need to be traced in the project, including steel bars, concrete, walls, doors and windows, and add a unique QR code to each key physical component; The IoT traceability unit is uniquely associated with the material batch number, equipment unique ID, and personnel identity information of the corresponding entity; All collected IoT data are associated with the corresponding entities, and the real-time status and relationship of all entities are mapped to the digital twin in real time. The digital twin updates the model in real time based on the data.

4. The closed-loop supervision and tracing method for the whole process quality and safety of urban construction projects according to claim 1 is characterized in that: The edge computing gateway-based, integrated sensor network-collected construction data, environmental parameters, and personnel location data are used to generate a structured engineering status dataset after spatiotemporal alignment processing. Specifically, the following steps are involved: The edge computing gateway is deployed in the core area of ​​the construction site, covering all key facilities, equipment and personnel locations; Real-time collection of construction data, environmental parameters, and personnel location data through multi-source sensors, and synchronization of all collected real-time data through the network time protocol; The data from different sensors are fused and processed, and the data that have been aligned and fused in time and space are converted into a structured engineering status dataset, including construction status, environmental parameters, structural response, equipment operation and personnel location.

5. The closed-loop supervision and tracing method for the whole process quality and safety of urban construction projects according to claim 1 is characterized in that: The anomaly detection model based on the convolutional neural network and the long short-term memory network is constructed. When the real-time data deviates from the standard threshold, an early warning is triggered and the responsibility chain of the associated traceable units is traced. Specifically, the following are included: The engineering status data is obtained based on the spatial distribution and response status of each sensor. The spatial features in the engineering status data are extracted through the convolution operation in the convolutional neural network, and the data dimension is reduced through the pooling layer. Combine the spatial features extracted by the convolutional neural network with the time dimension of the engineering status data, and use the LSTM network to learn the data change trends and dependencies between time points; LSTM uses a gating mechanism to adjust the memory and forgetting of historical data and obtain time series anomaly data during the construction process; Based on the data set processed by convolutional neural network and LSTM network, an anomaly detection model is built and model training is performed; Compare the real-time input engineering status data with the trained model data. When it is detected that the data deviates from the preset threshold of the quality and safety standard parameter set, an early warning is triggered and detailed alarm information data is generated; Based on the traceability unit of the Internet of Things, the specific responsible persons and responsible entities can be traced through the mapping relationship of the Internet of Things.

6. The closed-loop supervision and tracing method for the whole process quality and safety of urban construction projects according to claim 1 is characterized in that: The generation of a disposal instruction tree based on the warning level and the responsibility chain tracing results, and the distribution to the responsible parties through the smart contract in the blockchain network, specifically include: Based on the warning level and the traceable responsible persons and entities, an instruction tree containing different disposal suggestions is generated. The instruction tree is divided into multiple levels, and each level corresponds to different disposal measures, including suspension and rectification instructions, design change suggestions and process optimization plans; The distribution of instructions is automatically executed through smart contracts in the blockchain network, and a multi-layer disposal instruction tree is delivered to the responsible persons and responsible entities. Each instruction is bound to the responsible entity through the smart contract in the blockchain; The smart contract records the execution of each instruction and provides feedback on the processing progress.

7. The closed-loop supervision and tracing method for the whole process quality and safety of urban construction projects according to claim 1 is characterized in that: The responsible party uploads the handling process and results through the mobile terminal, which are bound to the warning event ID after hash encryption and stored in the traceability database. Specifically, it includes: The responsible party uploads every step of the disposal process, result verification data and supporting documents through a mobile terminal; Verify the integrity and content validity of the uploaded data, and generate a unique digital fingerprint for the successfully uploaded data through a hash algorithm; Bind each warning event with a unique identifier, and bind the generated unique digital fingerprint to the warning event identifier; The hashed and bound data is uploaded and stored in the traceability database, forming a complete and tamper-proof chain of evidence.

8. A closed-loop supervision and tracing platform for the quality and safety of the entire urban construction project process, used to implement the closed-loop supervision and tracing method for the quality and safety of the entire urban construction project process as described in any one of claims 1 to 7, characterized in that: include: Digital twin module: This module is used to integrate various data and build a digital twin of the entire life cycle of urban construction projects, while also presetting a set of quality and safety standard parameters; IoT traceability module: This module embeds IoT traceability units in key entities, traces the materials, equipment, and personnel information of each key entity component through QR codes, and maps them to the digital twin; Data acquisition module: This module collects construction data, environmental parameters, and personnel locations in real time through edge computing gateways and sensor networks to generate a structured engineering status dataset; Anomaly Detection Module: This module combines convolutional neural networks and LSTM networks to detect anomalies and trigger warnings based on real-time data compared with standard thresholds, tracing back to the chain of responsibility. Instruction tree generation module: This module generates a disposal instruction tree based on the warning level and responsibility chain tracing results, and automatically distributes instructions to the responsible parties through smart contracts in the blockchain; Traceability database module: This module uses hash encryption to bind with warning events and stores the disposal process data in the traceability database; Processor: The processor is used to process the calculation process of each formula and the construction calculation process of each model.

Images