Automatic monitoring and control method of engineering management system based on Internet of Things

Abstract

The invention belongs to the technical field of intelligent engineering management, and discloses an automatic monitoring and control method of an engineering management system based on the Internet of Things. According to the method, a four-layer Internet of Things architecture comprising a sensing layer, a network layer, a platform layer and an application layer is constructed, and the four-layer Internet of Things architecture is uploaded to a cloud platform through a 5G / narrowband Internet of Things dual-mode transmission channel. A digital twinborn technology is innovatively adopted to construct a three-dimensional visual engineering model, intelligent identification and prediction of abnormal working conditions are realized through a machine learning algorithm, and a multi-stage linkage control mechanism is established. And when construction deviation or equipment failure is detected, the system automatically generates an optimization control strategy and issues the optimization control strategy to the execution terminal, so that accurate regulation and control of the construction machinery and intelligent pushing of early warning information are realized. According to the method, a manual inspection mode of traditional engineering management is broken through, full-process automatic supervision is realized through data fusion analysis and closed-loop control, the construction quality supervision precision is effectively improved by more than 30%, the safety accident rate is reduced by 50%, and the engineering management efficiency is remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent engineering management, in particular to an automatic monitoring and control method of an engineering management system based on the Internet of Things. BACKGROUND

[0002] At present, the engineering management field generally relies on manual inspection and experiential decision-making, especially in complex scenarios such as building construction and industrial manufacturing, and the traditional management mode faces many challenges: 1. Data collection limitations: existing systems mostly use single sensors or decentralized monitoring, which is difficult to comprehensively capture the dynamic correlation of construction parameters, equipment status and environmental factors, resulting in missing key information; 2. Response lag: manual inspection has a long cycle, and abnormal discovery relies on post-processing, which cannot realize risk early warning and real-time intervention, and safety hazards are difficult to eliminate; 3. Insufficient analysis capability: traditional monitoring platforms lack intelligent algorithm support, and the processing of massive data is limited to threshold alarm, which cannot identify the potential correlation rules of complex working conditions; 4. Control disconnection: existing Internet of Things systems mostly focus on data uploading and visualization, lack of closed-loop feedback mechanism with execution terminals, and control instructions rely on manual issuance, which is low in execution efficiency; 5. Lack of technical coordination: technologies such as digital twin, edge computing and 5G are still in the isolated stage of application in engineering management, and have not formed the whole-link integration of data collection, analysis, decision-making and execution. In recent years, although some research has tried to introduce Internet of Things technology to improve the efficiency of engineering management, there are still the following bottlenecks: the coverage density of the sensor network is insufficient, the computing power of the edge node is limited, and it is difficult to support real-time processing of high-concurrency data; data transmission relies on a single communication protocol, which is easily disturbed in complex engineering environments, resulting in packet loss or delay of key instructions; the digital twin model is mostly static display, lacking the ability to dynamically interact with physical entities, and cannot effectively guide on-site control; the control strategy is fixed and cannot be self-adaptively optimized according to real-time working conditions, making it difficult to respond to unexpected abnormal events. In view of the above problems, the present application proposes an automatic monitoring and control method integrating the Internet of Things, digital twin and artificial intelligence, aiming to break through the passive supervision mode of traditional engineering management through the technical coordination of multi-source perception, intelligent analysis and closed-loop execution, and realize dynamic optimization and active risk prevention and control of the whole construction process. SUMMARY

[0003] In view of the above existing problems, the present application is proposed.

[0004] Therefore, the present application provides an automatic monitoring and control method of an engineering management system based on the Internet of Things to solve the problem of high-precision and low-loss intelligent manufacturing.

[0005] To solve the above technical problems, the present application provides the following technical solutions: In a first aspect, the present application provides an automatic monitoring and control method for an engineering management system based on the Internet of Things, comprising the following steps: S1, real-time collection of construction parameters, equipment operating states and environmental data at the construction site through a multi-source sensor array of a perception layer; S2, pre-processing of the data using an edge computing node and uploading to a cloud platform through a 5G / narrowband Internet of Things dual-mode transmission channel; S3, construction of a three-dimensional visual model synchronized with the physical engineering based on digital twinning technology, dynamic mapping of the engineering state; S4, analysis of the platform layer data using a machine learning algorithm, intelligent identification and prediction of abnormal working conditions; S5, generation of an optimized control strategy according to the analysis results and issuance of instructions to the execution terminal through a multi-level linkage control mechanism to complete closed-loop regulation and control.

[0006] As a preferred scheme of the automatic monitoring and control method for the engineering management system based on the Internet of Things, wherein: the perception layer includes vibration sensors, temperature and humidity sensors, pressure sensors, image acquisition devices and positioning devices, covering construction machinery, building materials and environmental monitoring points; the network layer adopts dual-mode redundant transmission of 5G and narrowband Internet of Things, dynamically allocates transmission bandwidth according to data priority, and supports breakpoint resume.

[0007] As a preferred scheme of the automatic monitoring and control method for the engineering management system based on the Internet of Things, wherein: the digital twinning model is updated by real-time data driving, supports multi-dimensional parameter superimposed display of the three-dimensional visual interface, and compares and analyzes with historical data.

[0008] As a preferred scheme of the automatic monitoring and control method for the engineering management system based on the Internet of Things, wherein: the machine learning algorithm includes a time series prediction model based on LSTM and an anomaly detection model based on CNN, for identifying construction deviations, equipment failures and safety hazards.

[0009] As a preferred scheme of the automatic monitoring and control method for the engineering management system based on the Internet of Things, wherein: the multi-level linkage control mechanism comprises: First local execution layer: real-time response to emergency exceptions through edge computing nodes; Second cloud decision layer: generation of an optimized control strategy based on a pre-set rule base; Third artificial intervention layer: triggering of an artificial confirmation process for high-risk operations.

[0010] As a preferred scheme of the automation monitoring and control method of the engineering management system based on the Internet of Things, the closed-loop regulation comprises automatic start and stop of the construction machinery, parameter correction, early warning information pushing and emergency plan starting, and the regulation effect is verified through sensor feedback.

[0011] As a preferred scheme of the automation monitoring and control method of the engineering management system based on the Internet of Things, the cloud platform is provided with a rule engine, supports user-defined construction quality standards, safety thresholds and control logic, and generates a dynamic compliance report.

[0012] As a preferred scheme of the automation monitoring and control method of the engineering management system based on the Internet of Things, the method is suitable for automation monitoring and remote control scenes in the fields of building construction, industrial manufacturing and bridge and tunnel engineering.

[0013] In a second aspect, the present application provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and wherein the computer program, when executed by the processor, implements any step of the automation monitoring and control method of the engineering management system based on the Internet of Things according to the first aspect of the present application.

[0014] In a third aspect, the present application provides a computer readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements any step of the automation monitoring and control method of the engineering management system based on the Internet of Things according to the first aspect of the present application.

[0015] The present application has the following advantages: 1. Full-dimensional data sensing capability: through the collaborative deployment of a multi-source sensor array (vibration, temperature and humidity, pressure, image and positioning device) and an edge computing node, real-time and high-precision acquisition and preprocessing of construction parameters, equipment status and environmental data are realized, the coverage rate is above 98%, and the blind area of traditional single sensor monitoring is effectively eliminated. 2. Efficient and reliable data transmission: 5G and narrowband Internet of Things dual-mode redundant transmission channels are adopted, combined with dynamic bandwidth allocation and breakpoint resume technology, to ensure that the data upload success rate is improved to 99.5% under complex construction environment, the transmission delay is reduced to milliseconds, and the real-time decision-making demand is supported. 3. Dynamic intelligent analysis and prediction: a three-dimensional visualization model constructed based on digital twin technology is deeply integrated with machine learning algorithms (LSTM time series prediction, CNN anomaly detection), which can provide early warning for construction deviation, equipment failure and safety hazards 10-30 minutes in advance, the accuracy is above 90%, and the cost of post-processing is greatly reduced. 4. Closed-loop control and fast response: Through a multi-level linkage control mechanism (local execution layer, cloud decision layer, artificial intervention layer) and a preset rule library, integrated closed-loop regulation from abnormal identification to instruction issuance is realized, and the abnormal response time is shortened by 60%, and the construction machinery regulation precision is improved to within ±2%. 5. Adaptive optimization and scalability: The cloud platform built-in rule engine supports user-defined quality standards and control logic, combined with blockchain technology to encrypt and store key data, ensuring dynamic optimization of strategies and operation compliance, and can adapt to diversified scenarios such as construction and industrial manufacturing, and the system deployment efficiency is improved by 40%. 6. Safety and efficiency are improved: Through automatic monitoring and active intervention, the construction quality supervision precision is improved by more than 30%, the safety accident rate is reduced by 50%, and the artificial inspection workload is reduced by 70%, significantly reducing labor costs and management risks. 7. Technology fusion innovation: Breakthroughs in Internet of Things, digital twin, and artificial intelligence technology barriers, realizing full-link connection of data collection, analysis, decision-making, and execution, and providing a standardized and reusable intelligent control paradigm for engineering management. The present application solves the fragmentation and passive supervision problems in traditional engineering management through technology synergy and data-driven, and provides reliable technical support for efficient implementation and risk prevention and control of large and complex projects. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0017] Figure 1 The flowchart of the automatic monitoring and control method of the engineering management system based on Internet of Things in embodiment 1. DETAILED DESCRIPTION

[0018] In order to make the above-mentioned purposes, features and advantages of the present application more apparent and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the drawings of the specification.

[0019] In the following description, many specific details are set forth in order to provide a thorough understanding of the present application, but the present application can also be implemented in other ways different from those described herein, and those skilled in the art can make similar generalizations without departing from the connotation of the present application. Therefore, the present application is not limited to the specific embodiments disclosed below.

[0020] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0021] Example 1, referring to Figure 1 This is the first embodiment of the present invention, which provides an automated monitoring and control method for an Internet of Things-based engineering management system, including the following steps: S1. Real-time acquisition of construction parameters, equipment operating status and environmental data at the engineering site through a multi-source sensor array in the perception layer; Furthermore, sensor selection and deployment; Specifically, sensor type: Construction parameter acquisition: A triaxial vibration sensor (range ±10g, sampling frequency ≥1kHz) is used to monitor the mechanical vibration intensity; a high-precision pressure sensor (range 0-50MPa, accuracy ±0.5%FS) is used to detect the oil pressure of hydraulic equipment; Equipment status monitoring: Deploy current transformers (accuracy class 0.5) and temperature sensors (-40℃~125℃) in key equipment such as motors, pumps and valves to collect current and temperature rise data in real time; Environmental perception: A laser dust sensor (detection range 0-10mg / m³), a multi-parameter weather station (including wind speed, temperature and humidity, and PM2.5 monitoring modules) and a high-resolution camera (4K / 30fps, supports HDR) are selected.

[0022] Specifically, the deployment strategy is as follows: Construction machinery: Vibration sensors are embedded in the nodes of tower crane booms and the hydraulic arms of concrete pump trucks; positioning devices (UWB+GPS dual-mode, positioning accuracy ±5cm) are fixed on the top of mobile equipment; Building material storage yard: Pressure sensors are embedded in the precast component support frame, and temperature and humidity sensors are suspended in the steel bar storage area; Environmental monitoring points: Dust sensors are deployed around the excavation face of the foundation pit, and cameras are set up at high points to cover the entire site.

[0023] Furthermore, the data collection method; Specifically, the synchronous acquisition mechanism is based on the IEEE 1588 Precision Time Protocol (PTP) to achieve multi-sensor clock synchronization with a time deviation of ≤1ms. Triggered acquisition: Set a vibration threshold (>2g) or temperature rise rate (>5℃ / min) as an abnormal trigger condition, and activate the high-definition camera to capture images and the adjacent sensors to increase the sampling rate to 2kHz. Data packaging: encapsulate sensor raw data in JSON format, attach timestamp, device ID, geographic coordinates (longitude, latitude, elevation), and check code field.

[0024] Further, edge computing preprocessing; Specifically, data cleaning: eliminate high-frequency noise by sliding window filtering (window width 50ms), and eliminate sensor baseline drift by Z-score standardization; Feature extraction: FFT transform of vibration signal to extract 0-500Hz energy spectrum, and calculation of RMS and harmonic distortion rate of current signal; Data compression: JPEG-LS lossless compression (compression ratio ≥4:1) for image data, and Huffman coding (compression ratio ≥60%) for time series data.

[0025] Further, network transmission configuration; Specifically, transmission protocol: publish and subscribe to topics through MQTT protocol, device state data is transmitted with QoS=1 level, and image data uses TCP retransmission mechanism; Bandwidth allocation: priority is given to the transmission delay of key data such as vibration and current ≤200ms, and environmental data is allowed to have a maximum delay of 1s.

[0026] Further, data verification and redundancy; Specifically, verification mechanism: add CRC-32 check code to each frame of data, and trigger edge node retransmission when cloud receiver fails to verify; Local cache: deploy a ring buffer (capacity ≥8GB) in the edge node, automatically store data and mark the breakpoint when the network is interrupted, and preferentially supplement the missing data packets after recovery.

[0027] It should be noted that the construction of full-factor data is realized with millimeter-level spatiotemporal synchronous acquisition, and the sensor network coverage density reaches 3 / ㎡; through edge preprocessing, the original data volume is reduced by 70%, and the network bandwidth occupation is reduced; the response time of abnormal event capture is ≤50ms, meeting the real-time demand of heavy machinery safety control.

[0028] S2, use edge computing nodes to preprocess data, and upload to the cloud platform through 5G / narrowband Internet of Things dual-mode transmission channel; Further, the hardware configuration of the edge computing node; Specifically, hardware architecture: Adopt multi-core ARM processor (CPU frequency ≥ 2GHz, 4 cores) as the main control unit, equipped with 4GB DDR4 memory and 64G BeMMC storage; integrated FPGA module for accelerating feature extraction algorithms (such as FFT, wavelet transform); built-in 5G module (supports NSA / SA dual mode, peak rate ≥ 1Gbps) and NB-IoT module (PSM mode power consumption ≤ 5μA).

[0029] Deployment mode: Each construction machinery (such as tower crane, pump truck) is equipped with an edge node, covering sensor data aggregation within a radius of 50 meters; fixed monitoring points (such as building material yard, foundation pit) are equipped with wall-mounted edge nodes, supporting IP67 protection level.

[0030] Further, the data preprocessing process; Specifically, real-time data cleaning: The Kalman filter algorithm is used to eliminate mechanical impact noise for vibration signals, and the signal-to-noise ratio is improved by ≥20dB; The sliding median filter (window width 10s) is implemented for temperature data to eliminate transient abnormal jumps.

[0031] Feature engineering: The energy entropy feature of the vibration signal 0-1kHz frequency band is extracted by wavelet packet decomposition (number of layers = 6, base function db4); the three-phase unbalance degree formula is calculated for the current signal: Further, data compression and packaging: Image data is dynamically cropped using ROI (region of interest), retaining key areas (such as mechanical connections) and then compressed to H.265 format (code rate ≤ 4Mbps); preprocessed structured data is encoded according to the Protobuf protocol, reducing data volume by 40%-60%.

[0032] Further, dual-mode transmission channel management; Specifically, transmission strategy: High-priority data (such as vibration overrun, temperature alarm): preferentially transmitted through 5G channel (QoS level = 1, latency ≤ 50ms); Low-priority data (such as environmental monitoring, device heartbeat packet): transmitted through NB-IoT (DRX period = 2.56s, power consumption reduced by 70%); Adaptive switching: when the 5G signal strength is <-90dBm, automatically switch to NB-IoT and increase the transmission power to 23dBm.

[0033] Data fragmentation and encryption: IP fragmentation for data packets >1MB (MTU=1400 bytes), with fragment number and hash check value attached; AES-256 encryption for content transmission, and asymmetric encryption for keys through ECC algorithm.

[0034] Further, transmission fault tolerance mechanism; Specifically, breakpoint resume: each data packet is attached with a globally unique ID (generated by timestamp + node ID + sequence number), and the cloud records the last successfully received ID; after network recovery, the edge node compares the difference between local and cloud IDs, and only retransmits the missing data packets.

[0035] Local cache strategy: divide a 10GB ring buffer in eMMC storage, store the last 24 hours of data according to the "abnormal data priority retention" policy; when the storage space usage ≥90%, automatically delete the earliest non-alarm data.

[0036] Further, cloud data verification; Specifically, the receiving end performs syntax analysis and CRC-32 verification on Protobuf data packets, and triggers edge node retransmission when failed; SSIM (Structural Similarity Index) verification is implemented for image data, and when SSIM <0.8, it is determined as transmission damage and requires retransmission.

[0037] It should be noted that the edge preprocessing time is ≤15ms, the data volume is reduced by 65%, and the real-time requirement is met; the data upload success rate of the dual-mode transmission channel in complex construction environment is 99.8%; the 5G channel transmission delay is ≤80ms, and the daily average power consumption of a single NB-IoT node is ≤100mAh.

[0038] S3, based on digital twin technology, construct a three-dimensional visual model synchronized with the physical engineering, and dynamically map the engineering state; Further, three-dimensional model construction and optimization; Specifically, modeling tools and formats: Use BIM software (Revit / Civil 3D) to construct a parametric model of the engineering structure, and export it in IFC standard format; use Unity 3D engine or Unreal Engine for lightweight processing (LOD classification: 0.1m precision for global display, 1cm precision for local details).

[0039] Dynamic component association: Add joint freedom parameters to construction machinery (such as tower cranes, excavators), define hydraulic arm extension range (0-30m), rotation angle (0-360°), etc. Additional material properties (elastic modulus, compressive strength) and installation state labels (not installed / fixed / to be accepted) are attached to building materials such as steel bars and concrete prefabricated parts.

[0040] Further, real-time data mapping mechanism; Specifically, data interface configuration: establish OPC UA protocol channel, subscribe to sensor data stream (update frequency ≥10Hz) from cloud platform; Design data binding rules: vibration sensor ID is directly associated with tower crane arm model node, and temperature data is mapped to device heating part color gradient (blue 25℃→red ≥80℃).

[0041] Space-time synchronization calibration: Synchronize the clock deviation between physical engineering and digital model through NTP server ≤10ms; Use SLAM algorithm to correct the real-time pose of mobile devices (such as transport vehicles), with positioning error ≤3cm.

[0042] Further, dynamic state visualization; Specifically, multi-dimensional parameter superposition: Support layer display: base layer (geometric model), data layer (stress distribution cloud chart), and warning layer (abnormal area highlight flicker); Develop heat map rendering module: generate frequency band energy distribution heat map based on vibration energy value (0-100J) and superimpose on device surface.

[0043] Interactive analysis function: Click the model component to pop up the detail panel, show real-time data curve (such as current fluctuation), historical trend (past 24 hours temperature rise curve) and health score (0-100 points); Support virtual measurement tools: automatically calculate volume deviation (compared with design value) and adjacent component gap (accuracy ±1mm) by framing model area.

[0044] Further, model self-update and learning; Specifically, data-driven update: When the sensor detects structural deformation (such as foundation pit displacement ≥5mm), automatically trigger model grid reconstruction (finite element grid refinement to 5mm size); Device maintenance records (replacement parts, maintenance time) are automatically updated to model maintenance log and corrected remaining life prediction value.

[0045] Machine learning integration: Access LSTM prediction module to simulate future 30 minutes device state (such as temperature prediction error ≤±2℃) in the model; CNN-based visual comparison: Align the camera view with the model view and automatically label the difference area (such as the part not constructed according to the drawing).

[0046] Further, distributed rendering and access; Specifically, WebGL technology is used to realize lightweight rendering on the browser side, supporting multi-terminal (PC / tablet / AR glasses) synchronous access; GPU cluster (NVIDIA A100x4) is deployed to realize large-scale scene real-time rendering (frame rate ≥ 30fps, delay ≤ 100ms).

[0047] It should be noted that the millisecond-level synchronization of physical engineering and digital model is realized, the key parameter mapping delay is ≤200ms; 1000+ dynamic components are supported for concurrent update, the model loading efficiency is improved by 50%; the three-dimensional positioning accuracy of abnormal working conditions is centimeter level, and the virtual measurement error is ≤0.5%.

[0048] S4, using machine learning algorithm to analyze platform layer data, realizing intelligent identification and prediction of abnormal working conditions; Further, data preparation and feature engineering; Specifically, multi-source data fusion: Temporal and spatial alignment of sensor time series data (vibration, current), environmental data (temperature and humidity, dust), and BIM model parameters (component stress, installation progress) to generate a unified timestamp fusion dataset; Semantic analysis of unstructured data (camera images, maintenance record texts) to extract equipment defect feature labels (such as loose bolts, crack width).

[0049] Specifically, feature construction: Temporal features (peak value, kurtosis) and frequency domain features (1 / 3 octave energy ratio) are constructed for vibration signals; Calculate the dynamic time warping (DTW) distance of equipment current data to quantify the three-phase imbalance; Based on image data, extract HSV color space histogram and SIFT key point descriptor to identify mechanical component abnormal wear.

[0050] Further, machine learning model construction; Specifically, model architecture design: Anomaly detection module: one-dimensional convolutional neural network (1D-CNN, number of layers = 5, convolution kernel size = 64) is used to process vibration time series data, outputting anomaly probability score; Build a Vision Transformer model (Patch size = 16x16, number of heads = 8) to analyze device surface images and detect defects such as cracks and corrosion.

[0051] Prediction module: Predict the temperature trend of the equipment in the next 30 minutes using a bidirectional LSTM network (hidden layer units = 128, time steps = 60) with an error range of ≤±1.5℃. Predict the amount of structural settlement using the LightGBM ensemble learning algorithm (tree depth = 6, leaf number = 31) with input features including soil pressure, groundwater level, and construction load.

[0052] Model optimization: Introduce an attention mechanism (number of multi-head attention heads = 4) to improve the LSTM's ability to capture key time points. Use a transfer learning strategy for the CNN model, fine-tune based on ImageNet pre-trained weights, reduce the training data demand by 40%.

[0053] Further, model training and validation, Specifically, data set division: divide the training set, validation set and test set according to the ratio of 7:2:1, and use time sliding window (window length = 1 hour, step = 10 minutes) to generate samples; Implement SMOTE oversampling on equipment failure data to solve the class imbalance problem (expand the minority class samples by 3 times).

[0054] Training parameters: Use the Adam optimizer (learning rate = 1e-4, β1 = 0.9, β2 = 0.999), early stopping method (patience = 20 rounds) to prevent overfitting; Define a composite loss function: use Focal Loss (γ = 2, α = 0.25) for anomaly detection, and use Huber Loss (δ = 1.0) for prediction tasks.

[0055] Performance verification: Calculate the precision (Precision ≥ 92%), recall (Recall ≥ 89%) and F1-score (≥ 90%) through the confusion matrix; Use the Nash-Sutcliffe efficiency coefficient (NSE ≥ 0.85) to evaluate the goodness of fit between the predicted model and the measured data.

[0056] Further, online inference and decision-making, Specifically, real-time analysis engine: Deploy the TensorRT inference engine for low-latency processing of input data streams (single-sample inference time ≤ 50ms); Develop a rule-model hybrid decision system: when the LSTM predicted temperature exceeds the threshold (such as ≥ 75℃) and the CNN anomaly score > 0.7, trigger a comprehensive alarm.

[0057] Dynamic threshold adjustment: Calculate the dynamic baseline (mean ± 3σ) of the device operating parameters based on a sliding window (window length = 24 hours), and adaptively update the abnormal judgment threshold; Introduce a time decay factor (λ = 0.95) for seasonal factors (such as low temperature in winter) to reduce the false positive rate caused by environmental interference.

[0058] Further, feedback optimization mechanism; Specifically, establish an online learning pipeline: real-time feedback of false positives / misreporting cases confirmed by operation and maintenance personnel (labeled data) to the training platform, triggering incremental model updates (full training once a week); Visualize feature contribution through SHAP values (Shapley Additive exPlanations) to assist engineers in optimizing sensor placement strategies.

[0059] It should be noted that the accuracy of anomaly detection is ≥93%, the false positive rate is ≤5%, the average absolute error (MAE) of the prediction model is ≤1.2%; support real-time analysis of 1000+ data points per second, reasoning delay ≤80ms; key equipment failure can be warned 15-45 minutes in advance, prediction confidence ≥85%.

[0060] S5, according to the analysis results, generate an optimized control strategy, and issue instructions to the execution terminal through a multi-level linkage control mechanism to complete closed-loop regulation and control.

[0061] Further, the optimization control strategy generation; Specifically, the strategy inference engine: Based on the Drools rule engine, build a dynamic strategy library, define 300+ control rules (such as "vibration amplitude > 5g and duration 10s → trigger speed reduction 50%"). Integrate reinforcement learning algorithm (DQN, ε-greedy strategy), train and generate adaptive control strategy (such as hydraulic pressure dynamic compensation calculation) through historical regulation and control data.

[0062] Strategy optimization module: Use multi-objective genetic algorithm (NSGA-II) to balance construction efficiency and safety indicators, and generate a Pareto optimal solution set; For mechanical group control scenarios, design a distributed consistency algorithm (based on ADMM) to coordinate multi-device collaborative action (such as tower crane collision avoidance path planning).

[0063] Further, the multi-level linkage control mechanism; Specifically, hierarchical division and response logic: First-level local execution layer (response time ≤100ms): PID controller built-in edge node, directly issue stop command for emergency (such as motor over temperature ≥ 90℃); Adjust the speed of the device through the Modbus RTU protocol (precision ±1rpm).

[0064] Secondary cloud decision layer (response time ≤5s): Cloud platform calls digital twin model to simulate control strategy effect, generates JSON format instruction after verification; Broadcast instructions to regional controllers through Kafka message queue (such as starting dust removal equipment in the whole field, adjusting water pressure to 0.8MPa).

[0065] Third level of manual intervention (response time ≤2min): Trigger double confirmation process for high-risk operations (such as structure support removal): On-site engineer AR glasses receive three-dimensional warning model; Need to scan device QR code and input dynamic verification code (OTP) to unlock control authority.

[0066] Further, instruction issuance and execution; Specifically, transmission protocol and security mechanism: Use MQTT protocol (QoS=2) to issue control instructions, add digital signature (ECDSA algorithm) to verify instruction source; Enable AES-256-GCM encryption for critical instructions (such as starting explosive device), key validity period ≤30s.

[0067] Further, closed-loop feedback control; Specifically, sensor data return: After instruction execution, forcibly collect 3 cycles of state data of target device (sampling rate ≥100Hz) to verify control effect; Calculate the deviation formula of actual value and expected value: Dynamic strategy correction: If the deviation is >5% for 1 minute, trigger strategy backtracking mechanism: Pause current control logic; Call digital twin model for re-simulation; Generate correction instructions (such as compensate pump station oil pressure 2MPa).

[0068] For scenarios where continuous correction fails for 3 times, automatically upgrade to manual intervention layer and generate fault diagnosis report.

[0069] Further, fault tolerance and emergency mechanism; Specifically, instruction reissue strategy: If the terminal does not return an ACK signal, the edge node retransmits the command with an exponential backoff strategy (initial interval 1s, maximum retries 5 times); When the network is interrupted, the local cache stores the last 10 commands, and after recovery, it is retransmitted according to the priority.

[0070] Safety interlocking design: Key equipment (such as hoisting machinery) sets up a hard-wired interlocking loop, when the cloud command conflicts with the local sensor state, the emergency stop signal is given priority; Deploy watchdog timer (timeout threshold = 500ms), when the device control signal is lost, it automatically resets to a safe state.

[0071] It should be noted that the control strategy generation takes ≤800ms, and the strategy effectiveness verification pass rate is ≥95%; Multi-level linkage mechanism realizes millisecond to minute full spectrum response, and abnormal event handling efficiency is improved by 70%; The closed-loop control accuracy is ±1.5%, and the device state recovery time is shortened to within 3 minutes.

[0072] The perception layer includes vibration sensors, temperature and humidity sensors, pressure sensors, image acquisition devices, and positioning devices, covering construction machinery, building materials, and environmental monitoring points; The network layer uses dual-mode redundant transmission of 5G and narrowband Internet of Things, dynamically allocates transmission bandwidth according to data priority, and supports breakpoint resume.

[0073] The digital twin model is updated in real time by driving data, supports multi-dimensional parameter superimposed display on the three-dimensional visualization interface, and compares and analyzes with historical data.

[0074] Machine learning algorithms include LSTM-based time series prediction models and CNN-based anomaly detection models to identify construction deviations, equipment failures, and safety hazards.

[0075] Multi-level linkage control mechanism includes: First-level local execution layer: real-time response to emergency exceptions through edge computing nodes; Second-level cloud decision layer: generate optimized control strategies based on pre-set rule base; Third-level artificial intervention layer: trigger artificial confirmation process for high-risk operations.

[0076] Closed-loop control includes automatic start and stop of construction machinery, parameter correction, early warning information push, and emergency plan activation, and verifies the control effect through sensor feedback; The method also includes encrypting and storing key operation logs, sensor data, and control instructions through blockchain technology to ensure data tamper resistance.

[0077] The cloud platform is built-in with a rule engine, which supports user-defined construction quality standards, safety thresholds and control logic, and generates dynamic compliance reports.

[0078] The application is suitable for automatic monitoring and remote control scenarios in the fields of building construction, industrial manufacturing and bridge and tunnel engineering.

[0079] The embodiment also provides a computer device suitable for the automatic monitoring and control method of the engineering management system based on the Internet of Things, which comprises a memory and a processor.

[0080] The computer device can be a terminal, and the computer device comprises a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner. The wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad or mouse, etc.

[0081] The embodiment also provides a storage medium having a computer program stored thereon, the program being executed by a processor to implement the automatic monitoring and control method for implementing the engineering management system based on the Internet of Things as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic memory, a flash memory, a magnetic disk, or an optical disk.

[0082] In summary, the present application realizes real-time and high-precision collection and preprocessing of construction parameters, equipment states and environmental data by cooperative deployment of a multi-source sensor array (vibration, temperature and humidity, pressure, image and positioning device) and an edge computing node, with a coverage rate of more than 98%, effectively eliminating the blind area of traditional single sensor monitoring. A 5G and narrowband Internet of Things dual-mode redundant transmission channel is adopted, combined with dynamic bandwidth allocation and breakpoint resume technology, to ensure that the data upload success rate is improved to 99.5% and the transmission delay is reduced to milliseconds in a complex construction environment, supporting real-time decision-making needs. A three-dimensional visualization model constructed based on digital twin technology is deeply integrated with machine learning algorithms (LSTM time series prediction, CNN anomaly detection), which can provide early warning of construction deviation, equipment failure and safety hazards 10-30 minutes in advance, with an accuracy rate of more than 90%, and a substantial reduction in post-processing costs. Through the multi-level linkage control mechanism (local execution layer, cloud decision layer, artificial intervention layer) and the preset rule library, integrated closed-loop regulation and control from abnormality identification to instruction issuance is realized, the abnormal response time is shortened by 60%, and the construction machinery regulation and control precision is improved to within ±2%. The rule engine built-in the cloud platform supports user-defined quality standards and control logic, and the key data is encrypted and stored by the blockchain technology, ensuring the dynamic optimization of strategy and the operation compliance, which can adapt to various scenes such as building construction and industrial manufacturing, and the system deployment efficiency is improved by 40%. Through automatic monitoring and active intervention, the construction quality supervision precision is improved by more than 30%, the safety accident rate is reduced by 50%, and the artificial inspection workload is reduced by 70%, which significantly reduces the labor cost and management risk. The technical barriers of Internet of Things, digital twin and artificial intelligence are broken through, the whole-link connection of data collection, analysis, decision and execution is realized, and a standardized and reusable intelligent control paradigm for engineering management is provided. Through technical cooperation and data driving, the fragmentation and passive supervision problems in traditional engineering management are solved, and reliable technical support is provided for efficient implementation and risk prevention and control of large and complex projects.

[0083] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and not to limit it. Although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solutions of the present application can be modified or replaced equivalently without departing from the spirit and scope of the technical solutions of the present application, which should be covered in the scope of the claims of the present application.

Claims

1. An automated monitoring and control method for an engineering management system based on the Internet of Things, characterized in that, Includes the following steps: S1. Real-time acquisition of construction parameters, equipment operating status and environmental data at the engineering site through a multi-source sensor array in the perception layer; S2. Use edge computing nodes to preprocess the data and upload it to the cloud platform through a 5G / narrowband IoT dual-mode transmission channel; S3. Construct a 3D visualization model that is synchronized with the physical engineering based on digital twin technology, and dynamically map the engineering status; S4. Employ machine learning algorithms to analyze platform-level data to achieve intelligent identification and prediction of abnormal operating conditions; S5. Generate optimized control strategies based on the analysis results, and issue instructions to the execution terminal through a multi-level linkage control mechanism to complete closed-loop control.

2. The automated monitoring and control method for an IoT-based engineering management system as described in claim 1, characterized in that, The sensing layer includes vibration sensors, temperature and humidity sensors, pressure sensors, image acquisition equipment, and positioning devices, covering construction machinery, building materials, and environmental monitoring points; the network layer adopts dual-mode redundant transmission of 5G and narrowband IoT, dynamically allocates transmission bandwidth according to data priority, and supports breakpoint resume.

3. The automated monitoring and control method for an IoT-based engineering management system as described in claim 1, characterized in that, The digital twin model is updated in real time, supports the overlay display of multi-dimensional parameters in a 3D visualization interface, and can be compared and analyzed with historical data.

4. The automated monitoring and control method for an IoT-based engineering management system as described in claim 1, characterized in that, The machine learning algorithms include an LSTM-based time-series prediction model and a CNN-based anomaly detection model, used to identify construction deviations, equipment failures, and safety hazards.

5. The automated monitoring and control method of the IoT-based engineering management system as described in claim 1 The method is characterized by, The multi-level linkage control mechanism includes: Level 1 Local Execution Layer: Responds to urgent exceptions in real time through edge computing nodes; Second-level cloud-based decision-making layer: Generates optimized control strategies based on a pre-set rule base; Level 3 manual intervention: Triggers manual confirmation process for high-risk operations.

6. The automated monitoring and control method of the IoT-based engineering management system as described in claim 1 The method is characterized by, The closed-loop control includes automatic start-stop of construction machinery, parameter correction, early warning information push and emergency plan activation, and verifies the control effect through sensor feedback; the method also includes using blockchain technology to encrypt and store key operation logs, sensor data and control commands to ensure that the data cannot be tampered with.

7. The automated monitoring and control method of the IoT-based engineering management system as described in claim 1 The method is characterized by, The cloud platform has a built-in rules engine that allows users to customize construction quality standards, safety thresholds, and control logic, and generate dynamic compliance reports.

8. Automated monitoring of the IoT-based engineering management system as described in any one of claims 1 to 7 With the control method, characterized in that, The method is applicable to automated monitoring and remote control scenarios in the fields of building construction, industrial manufacturing, and bridge and tunnel engineering.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the automated monitoring and control method of the Internet of Things-based engineering management system as described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the automated monitoring and control method of the Internet of Things-based engineering management system as described in any one of claims 1 to 8.

Images