A Method and System for Load Status Assessment and Early Warning of Factory Machinery Based on Digital Twin

Abstract

This invention discloses a method and system for load status assessment and early warning of a factory construction machine based on digital twins, belonging to the field of intelligent construction technology. The method includes the following steps: constructing a digital twin model of the factory construction machine; collecting multi-source sensor data of the physical entity and driving the twin to update synchronously; performing adaptive fusion of the data based on confidence level to generate a comprehensive load status vector; calculating a comprehensive load status assessment index based on the vector; generating real-time thresholds through a dynamic early warning threshold model to achieve graded early warning; predicting the remaining service life of key components by combining historical data; and finally displaying the early warning and prediction results through a visual interface. This invention achieves real-time comprehensive assessment of load status, adaptive early warning, and service life prediction, significantly improving the safety and intelligent operation and maintenance level of the factory construction machine.

Description

Technical Field

[0001] This invention relates to the field of intelligent construction technology, specifically to a method and system for assessing and warning the load status of factory machinery based on digital twins. Background Technology

[0002] Mobile intelligent plant construction machines are large, mobile integrated equipment platforms used for the rapid construction of cast-in-place concrete structure plant buildings. They are characterized by complex structures, variable loads, and highly dynamic operating environments. Currently, the industry mainly relies on discrete single-point monitoring, periodic manual inspections, and experience-based judgment for load status monitoring and safety management of such heavy equipment.

[0003] However, existing technologies have the following shortcomings: single-point monitoring cannot perceive the overall structural response and lacks a comprehensive quantitative assessment of the structural system's safety; fixed threshold early warning cannot adapt to the dynamic load characteristics of different construction stages, and the structure often enters a non-ideal state when an alarm is triggered; existing methods cannot effectively accumulate and assess the fatigue damage of structures under cyclic loading, and cannot predict the remaining service life of key load-bearing components, resulting in a lack of foresight in operation and maintenance decisions.

[0004] Therefore, there is an urgent need for an intelligent monitoring and early warning technology that can achieve full-time perception, full-element fusion, full-state assessment and full-cycle prediction, in order to solve the core pain points of existing methods such as inaccurate assessment, unreliable early warning and unintelligent maintenance, and ensure the inherent safety and operational economy of the manufacturing machine under long-cycle and high-load operation. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to address the shortcomings of the prior art by providing a method and system for assessing and warning the load status of factory machinery based on digital twins.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] The method for assessing and providing early warning of plant machine load status based on digital twins includes the following steps:

[0008] Step S1: Construct a digital twin model of the manufacturing machine, wherein the digital twin model integrates the geometric information, physical attributes, and physical field behavior model of the manufacturing machine;

[0009] Step S2: Collect multi-source sensor data on the physical entity of the manufacturing machine, and map the multi-source sensor data to the digital twin model to drive the state of the digital twin model to be updated synchronously with the physical entity;

[0010] Step S3: Perform adaptive weighted fusion processing based on confidence assessment on the collected multi-source sensor data to generate a comprehensive load state vector;

[0011] Step S4: Calculate the comprehensive load state evaluation index based on the generated comprehensive load state vector;

[0012] Step S5: Generate the dynamic early warning threshold for the current moment based on the dynamic early warning threshold model, compare the calculated load state comprehensive evaluation index with the dynamic early warning threshold, and trigger graded early warnings based on the comparison results;

[0013] Step S6: Based on historical load data and the generated comprehensive load state vector, predict the remaining service life of key load-bearing components;

[0014] Step S7: Display the graded early warning information and the predicted remaining service life results in real time through a visual interface.

[0015] Furthermore, step S1 specifically includes the following steps:

[0016] Step S1.1: Extract geometric skeleton information based on the building information model of the factory machine;

[0017] Step S1.2: Assign material properties to the structural components in the geometric skeleton and define the connection relationships between the components. The material properties include elastic modulus, density and yield strength, and the connection relationships include hinged or fixed connections.

[0018] Step S1.3: Integrate the structural finite element analysis model, the hydraulic system dynamic model, and the platform kinematic model into the digital twin model.

[0019] Furthermore, step S2 specifically includes the following steps:

[0020] Step S2.1: Establish the mapping relationship between the real-time data stream collected by the physical sensors and the corresponding state parameters in the digital twin model;

[0021] Step S2.2: Based on the mapping relationship, input the real-time collected sensor data into the digital twin model to drive it to perform state calculation and update.

[0022] Furthermore, step S3 specifically includes the following steps:

[0023] Step S3.1: Collect strain and displacement data output by sensors deployed on the main steel platform, hydraulic support points and walking mechanism; at the same time, collect pressure, tilt angle and environmental data for independent system status monitoring and visualization.

[0024] Step S3.2: Perform normalization preprocessing on the collected raw strain, displacement, and pressure data;

[0025] Step S3.3: Real-time assessment of the confidence level of each strain and displacement data source. The assessment criteria include signal-to-noise ratio, deviation between the data and the values ​​calculated by the digital twin model, and sensor health status.

[0026] Step S3.4: Dynamically allocate fusion weights according to the confidence level of each data source, perform weighted fusion calculation, and generate the comprehensive load state vector.

[0027] Furthermore, in step S4, the comprehensive evaluation index of the load state is a weighted sum of the stress safety factor, deformation safety factor, shear safety factor, and fatigue damage accumulation factor; the stress safety factor, deformation safety factor, and shear safety factor are the ratios of the real-time measured values ​​of the corresponding physical quantities to the design allowable values.

[0028] Furthermore, in step S5, the dynamic early warning threshold model is constructed based on a benchmark threshold, construction cycle, cumulative equipment operating time, and design life parameters. The output value fluctuates periodically with the construction stage and decreases linearly with the increase of equipment usage time. The specific formula is as follows:

[0029]

[0030] in, This represents the dynamic warning threshold at time t. Indicates the baseline warning threshold. This represents the influence coefficient of periodic loads. This represents a typical complete construction cycle. Indicates the aging and degradation coefficient of the equipment. This indicates the cumulative operating time of the equipment since it was put into use. This indicates the total design life of the equipment.

[0031] Further, in step S5, the tiered early warning includes:

[0032] The first-level warning is triggered when the comprehensive load status evaluation index exceeds the dynamic warning threshold, and the system provides a status prompt.

[0033] The second-level warning is triggered when the comprehensive evaluation index of the load state exceeds the first preset limit, and the system suggests adjusting or suspending the current operation.

[0034] The third-level warning is triggered when the comprehensive evaluation index of the load status exceeds the second preset limit, and the system forcibly locks the relevant equipment.

[0035] Furthermore, in step S6, the remaining service life is predicted using an exponential decay model, which multiplies the total designed life of the equipment by a decay factor; the decay factor is calculated based on a negative exponential function of the natural constant.

[0036] A digital twin-based system for assessing and warning the load status of manufacturing machinery is provided to implement a digital twin-based method for assessing and warning the load status of manufacturing machinery, including:

[0037] Sensor sensing modules are distributed across the stress-bearing parts of the manufacturing machine to collect strain, displacement, pressure, tilt angle, and environmental data in real time.

[0038] The edge computing and communication module, deployed on the main body of the manufacturing machine, is used for preprocessing, local fusion, and uploading of sensor data via wireless network;

[0039] The cloud-based digital twin engine module is used to build, maintain, and run digital twin models, and to receive data to drive model synchronization.

[0040] The core module of intelligent assessment and decision-making is used to calculate the comprehensive assessment index, manage dynamic early warning thresholds, trigger graded early warnings, and predict the remaining lifespan based on the fused load data.

[0041] The visualization module is used to display system status and early warning information.

[0042] Furthermore, the core module for intelligent evaluation and decision-making specifically includes:

[0043] The load state comprehensive evaluation index calculation unit is used to receive the comprehensive load state vector and calculate the load state comprehensive evaluation index.

[0044] The dynamic early warning threshold management unit is used to dynamically calculate and update the dynamic early warning threshold based on the current construction cycle time, the cumulative equipment running time, and preset parameters.

[0045] A multi-level early warning triggering unit is used to compare the comprehensive load status evaluation index with the dynamic early warning threshold in real time, and trigger the corresponding early warning level and related handling suggestions.

[0046] The remaining service life prediction unit is used to calculate and output the predicted value of the remaining service life of key load-bearing components based on historical load data and an exponential decay model.

[0047] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0048] 1. This invention achieves a complete mapping of physical entities in virtual space by constructing a digital twin model. It effectively filters noise and abnormal data by using adaptive weighted fusion to generate a reliable comprehensive load state vector. Based on this, it proposes a comprehensive load state evaluation index, which integrates multi-dimensional safety information such as stress, deformation, shear, and fatigue damage. It comprehensively and scientifically reflects the overall safety margin of the structure through intuitive quantitative values, thus changing the limitations of traditional single-point and one-sided evaluation.

[0049] 2. The dynamic early warning threshold model proposed in this invention can automatically adjust the threshold benchmark according to the construction cycle, accommodating normal periodic load fluctuations, and linearly tighten the safety margin based on the cumulative operating time of the equipment, effectively reducing false alarms and missed alarms caused by changes in the construction stage and equipment aging. Combined with multi-level early warning triggering logic, it can upgrade safety risk management from passive response to proactive prevention, fundamentally improving the accuracy and timeliness of early warnings.

[0050] 3. This invention establishes a remaining service life prediction model. By integrating the historical equivalent stress time history, it scientifically quantifies the accumulation process of fatigue damage and predicts the remaining service life of key load-bearing components under future load spectrum based on this. This transforms equipment maintenance from the traditional periodic inspection or post-failure repair mode to predictive maintenance based on real-time status. It allows for advance planning of repair or replacement plans, maximizing equipment utilization efficiency while ensuring safety, and significantly reducing the total life cycle maintenance cost. Attached Figure Description

[0051] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0052] Figure 1 This is a flowchart illustrating an embodiment of the present invention;

[0053] Figure 2 This is a schematic diagram of the system architecture of an embodiment of the present invention;

[0054] Figure 3 This is a schematic diagram of the intelligent evaluation and decision-making core module in an embodiment of the present invention. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0056] like Figure 1 As shown, the method for assessing and warning the load status of factory machinery based on digital twins includes the following steps:

[0057] Step S1: Construct a digital twin model of the manufacturing machine, wherein the digital twin model integrates the geometric information, physical attributes, and physical field behavior model of the manufacturing machine;

[0058] Step S2: Collect multi-source sensor data on the physical entity of the manufacturing machine, and map the multi-source sensor data to the digital twin model to drive the state of the digital twin model to be updated synchronously with the physical entity;

[0059] Step S3: Perform adaptive weighted fusion processing based on confidence assessment on the collected multi-source sensor data to generate a comprehensive load state vector;

[0060] Step S4: Calculate the comprehensive load state evaluation index based on the generated comprehensive load state vector;

[0061] Step S5: Generate the dynamic early warning threshold for the current moment based on the dynamic early warning threshold model, compare the calculated load state comprehensive evaluation index with the dynamic early warning threshold, and trigger graded early warnings based on the comparison results;

[0062] Step S6: Based on historical load data and the generated comprehensive load state vector, predict the remaining service life of key load-bearing components;

[0063] Step S7: Display the graded early warning information and the predicted remaining service life results in real time through a visual interface.

[0064] Step S1 specifically includes the following steps:

[0065] Step S1.1: Extract geometric skeleton information based on the building information model of the factory machine;

[0066] Step S1.2: Assign material properties to the structural components in the geometric skeleton and define the connection relationships between the components. The material properties include elastic modulus, density and yield strength, and the connection relationships include hinged or fixed connections.

[0067] Step S1.3: Integrate the structural finite element analysis model, the hydraulic system dynamic model, and the platform kinematic model into the digital twin model.

[0068] Based on the building information model of the mobile intelligent factory manufacturing machine, the precise geometric skeleton information constituting its physical structure is extracted. First, a 3D geometric model file containing all structural components (such as main beams, secondary beams, columns, supports, and platform slabs) is exported from BIM software. Then, the geometric model is cleaned and simplified using a model processing engine, removing non-structural components and details irrelevant to mechanical analysis, retaining the core wireframe and surface model used for structural calculations. Finally, the simplified model is converted into a format suitable for finite element analysis, and meshing is completed to generate a geometric skeleton composed of nodes and elements, which serves as the basis for the digital twin to bear physical properties.

[0069] Each structural component in the generated geometric skeleton is assigned realistic material physical properties. Based on the plant machine design drawings and material reports, the elastic modulus, density, and yield strength of the main components of the steel platform are defined; simultaneously, the connection relationships between components are clearly defined: bolted connections or nodes that can be considered to rotate relatively freely are defined as hinged, which only transmit force and not bending moment; welded or rigid flange connections are defined as fixed, which transmit both force and bending moment. In addition, contact properties, such as the coefficient of friction, need to be defined for the contact surfaces between hydraulic cylinders and the platform. This set of properties and connection relationships transforms the abstract geometric skeleton into a mechanical model with physical responsiveness.

[0070] By integrating specialized models describing behaviors in different physical domains into the aforementioned attributed structural model, a complete and computable multiphysics digital twin model is formed. Specifically, this includes: a structural finite element analysis model: based on assigned material properties and connection relationships, a finite element model is constructed to calculate the internal stress, strain, and deformation of the structure under load; a hydraulic system dynamic model: a model describing the working principle of the hydraulic lifting and walking system of the manufacturing machine is established, including the dynamic equations of the hydraulic pump, control valve, and hydraulic cylinder, used to simulate and predict the pressure, flow rate, and lifting speed of the hydraulic cylinder; and a platform kinematic model: a kinematic relationship model describing the overall movement of the manufacturing machine along the track is established, used to calculate the displacement, velocity, and acceleration of the platform during its movement.

[0071] Step S2 specifically includes the following steps:

[0072] Step S2.1: Establish the mapping relationship between the real-time data stream collected by the physical sensors and the corresponding state parameters in the digital twin model;

[0073] Step S2.2: Based on the mapping relationship, input the real-time collected sensor data into the digital twin model to drive it to perform state calculation and update.

[0074] Establish a one-to-one data mapping relationship between the digital twin model and the physical sensor network. Specifically, for each physical sensor (such as a strain sensor located in the middle of the main beam span), find its corresponding logical location in the twin model, i.e., a specific node or element in the finite element model. Establish a mapping table to clarify the parameter type and location index in the twin model corresponding to each physical sensor channel ID;

[0075] Based on the established mapping relationship, the physical quantity data stream acquired in real time through the sensor network is used as known boundary conditions or state inputs to drive the digital twin model for rapid state calculation and updating. For example, a model-data fusion algorithm based on a Kalman filter can be used for state updating. The finite element model of the digital twin is discretized into state-space equations, and key point sensor data are used as observations. The model state variables are continuously corrected through the Kalman gain matrix, thereby achieving real-time and optimal estimation of the stress and strain states across the entire field.

[0076] Step S3 specifically includes the following steps:

[0077] Step S3.1: Collect strain and displacement data output by sensors deployed on the main steel platform, hydraulic support points and walking mechanism; at the same time, collect pressure, tilt angle and environmental data for independent system status monitoring and visualization.

[0078] Step S3.2: Perform normalization preprocessing on the collected strain and displacement data;

[0079] Step S3.3: Real-time assessment of the confidence level of each strain and displacement data source. The assessment criteria include signal-to-noise ratio, deviation between the data and the values ​​calculated by the digital twin model, and sensor health status.

[0080] Step S3.4: Dynamically allocate fusion weights according to the confidence level of each data source, perform weighted fusion calculation, and generate the comprehensive load state vector.

[0081] A distributed sensor network deployed at key stress-bearing components of the machine simultaneously collects multi-dimensional data reflecting its load and environmental conditions. This data primarily includes: strain data, reflecting local structural deformation; displacement data, reflecting overall structural or key point deformation; pressure data, reflecting the working load of the hydraulic system; and tilt angle data, used to display the real-time tilt status of the machine platform. When the tilt angle exceeds a safe threshold, the system issues an independent tilt angle over-limit alarm, reminding operators to adjust the platform's level. Environmental data is displayed in real-time on a visual interface, providing a reference for construction safety. For example, when the wind speed exceeds a certain threshold, it prompts a stop to high-altitude operations.

[0082] The collected raw data are normalized to eliminate the influence of different physical dimensions and orders of magnitude on subsequent fusion calculations. Among them, stress and displacement are normalized using design allowable values.

[0083] To ensure the reliability of the fused data, a real-time confidence assessment is performed on the data sources from strain and displacement sensors. The confidence level is a value between 0 and 1, where 0 represents completely unreliable and 1 represents completely reliable. The specific formula is as follows:

[0084]

[0085] in, This represents the confidence level of the i-th data source at time t. This represents the signal quality score of the i-th data source at time t, calculated based on the signal-to-noise ratio of the sensor signal. This represents the data deviation score of the i-th data source at time t. This represents the sensor health status score of the i-th data source at time t. , , These are the weight coefficients corresponding to the scores, and their sum is 1. They can be adjusted according to actual needs. For example, when the system is initially put into operation, the model may be inaccurate, so they can be set accordingly. Higher, such as 0.5. The initial value is relatively low, such as 0.2; as the model is continuously optimized, it can be gradually increased. The weight.

[0086] The calculation formulas for the signal quality score, data deviation score, and sensor health status score are as follows:

[0087]

[0088] in, This represents the signal-to-noise ratio of the i-th sensor at time t, which is converted to a linear value during calculation;

[0089]

[0090] in, This represents the bias decay coefficient, used to control the rate at which the score decreases as the bias increases. It is usually set to 5 or 10. The larger the bias decay coefficient, the more sensitive the score is to bias. This represents the normalized measured value of the i-th sensor at time t. This represents the normalized theoretical value calculated by the digital twin model at the corresponding position and time t. This represents the reference deviation benchmark, used to normalize the absolute deviation. It is usually taken as an empirical value, such as 0.1.

[0091]

[0092] in, This represents the total number of health monitoring indicators. This represents the actual value of the k-th monitoring index of the i-th sensor at time t. This represents the rated value of the k-th monitoring indicator for the i-th sensor. This represents the allowable deviation threshold for the k-th indicator. This represents the unit step function. The value within the parentheses is 1 if it is greater than 0, and 0 otherwise. It is used to determine if an indicator is abnormal. This represents the health impact factor of the k-th indicator, which is the degree of discount on the overall health score when the indicator is abnormal. The value is 1 for key indicators such as power supply abnormalities and 0.2 for minor indicators.

[0093] Based on the calculated real-time confidence levels of each data source, their weights in the fusion calculation are dynamically allocated. The higher the confidence level of a data source, the greater its weight. The specific formula is as follows:

[0094]

[0095] in, Indicates the total number of data sources. This represents the weight of the i-th data source.

[0096] The normalized sensor data is weighted and fused using assigned weights to generate a comprehensive load state vector that can fully and robustly characterize the overall load state of the machine. The specific formula is as follows:

[0097]

[0098] in, This represents the normalized data vector of the i-th data source. This represents the combined load state vector at time t.

[0099] In step S4, the comprehensive load state evaluation index is a weighted sum of the stress safety factor, deformation safety factor, shear safety factor, and fatigue damage accumulation factor; the stress safety factor, deformation safety factor, and shear safety factor are the ratios of the real-time measured values ​​of the corresponding physical quantities to the design allowable values, respectively, and the specific formulas are as follows:

[0100]

[0101] in, This represents the comprehensive evaluation index of the load condition. The measured maximum normal stress at time t is calculated from the strain value measured by the strain sensor using the material constitutive relation. Indicates the allowable stress of the material. The measured displacement of the key point at time t is obtained by a displacement sensor. Indicates the maximum allowable displacement in the design. The measured maximum shear stress at time t is calculated from data from strain sensors placed at the critical shear points. Indicates the allowable shear stress of the material. Indicates the fatigue damage accumulation factor. , , , These represent the weighting coefficients for stress, deformation, shear, and fatigue, respectively, and their sum is 1. These coefficients are determined through optimization using historical data; for example, taking... It is 0.4. It is 0.2. It is 0.2. It is 0.2.

[0102] The formula for calculating the fatigue damage accumulation factor is:

[0103]

[0104] in, This represents the total number of stress levels in the load stress spectrum, that is, discretizing the continuous stress change into M different stress level intervals, where m represents the m-th stress level. This represents the actual number of load cycles that occurred within the monitoring period at stress level m. This represents the total number of cycles required to cause fatigue failure of the component at the m-th stress level, based on the material's SN curve. This represents the fatigue curve index of the material. For commonly used structural steels such as Q345, it is usually taken as 3.

[0105] In step S5, the dynamic early warning threshold model is constructed based on a baseline threshold, construction cycle, cumulative equipment operating time, and design life parameters. The output value fluctuates periodically with the construction phase and decreases linearly with increasing equipment usage time. The specific formula is as follows:

[0106]

[0107] in, This represents the dynamic warning threshold at time t. This represents the baseline warning threshold, a preset constant set based on engineering experience and safety regulations, typically 0.85. This represents the influence coefficient of periodic loads. , This represents a typical complete construction cycle. Indicates the aging and degradation coefficient of the equipment. , This indicates the cumulative operating time of the equipment since it was put into use. This indicates the total design life of the equipment.

[0108] In step S5, the tiered early warning includes:

[0109] The first-level warning is triggered when the comprehensive load status evaluation index exceeds the dynamic warning threshold, and the system provides a status prompt.

[0110] The second-level warning is triggered when the comprehensive evaluation index of the load state exceeds the first preset limit, and the system suggests adjusting or suspending the current operation.

[0111] The third-level warning is triggered when the comprehensive evaluation index of the load status exceeds the second preset limit, and the system forcibly locks the relevant equipment.

[0112] The trigger condition for the first-level warning is: the comprehensive evaluation index of the load status exceeds the dynamic warning threshold;

[0113] The trigger condition for a Level 2 warning is: the comprehensive load condition assessment index exceeds 110% of the dynamic warning threshold, which is the first preset limit of 1.1. ;

[0114] The trigger condition for a Level 3 warning is: the comprehensive load condition assessment index exceeds 120% of the dynamic warning threshold, which is equivalent to the second preset limit of 1.2. ;

[0115] The second and third level warning thresholds increase by a fixed percentage of 10%. Based on the analysis of historical accident data, it was found that the load exceeding the standard by 10% is controllable, while exceeding the standard by 20% will cause the risk to rise sharply.

[0116] In step S6, the remaining service life is predicted using an exponential decay model, which multiplies the total designed lifespan of the equipment by a decay factor. The decay factor is calculated based on a negative exponential function of the natural constant, with the specific formula as follows:

[0117]

[0118] in, This represents the predicted remaining lifespan of the equipment at time t. This represents the critical damage value, typically taken as 0.8. When accumulated damage reaches this value, the component is considered to have reached the end of its service life. Indicates time The equivalent stress, according to the fourth strength theory, converts the complex stress state into an equivalent uniaxial stress. The reference stress level is typically taken as a portion of the material's fatigue limit or yield strength. This represents the material fatigue curve index, which is the same as the index in the fatigue damage accumulation factor.

[0119] A digital twin-based system for assessing and warning the load status of manufacturing machinery is provided to implement a digital twin-based method for assessing and warning the load status of manufacturing machinery, including:

[0120] Sensor sensing modules are distributed across the stress-bearing parts of the manufacturing machine to collect strain, displacement, pressure, tilt angle, and environmental data in real time.

[0121] The edge computing and communication module, deployed on the main body of the manufacturing machine, is used for preprocessing, local fusion, and uploading of sensor data via wireless network;

[0122] The cloud-based digital twin engine module is used to build, maintain, and run digital twin models, and to receive data to drive model synchronization.

[0123] The core module of intelligent assessment and decision-making is used to calculate the comprehensive assessment index, manage dynamic early warning thresholds, trigger graded early warnings, and predict the remaining lifespan based on the fused load data.

[0124] The visualization module is used to display system status and early warning information.

[0125] like Figure 3 As shown, the core module for intelligent evaluation and decision-making specifically includes:

[0126] The load state comprehensive evaluation index calculation unit is used to receive the comprehensive load state vector and calculate the load state comprehensive evaluation index.

[0127] The dynamic early warning threshold management unit is used to dynamically calculate and update the dynamic early warning threshold based on the current construction cycle time, the cumulative equipment running time, and preset parameters.

[0128] A multi-level early warning triggering unit is used to compare the comprehensive load status evaluation index with the dynamic early warning threshold in real time, and trigger the corresponding early warning level and related handling suggestions.

[0129] The remaining service life prediction unit is used to calculate and output the predicted value of the remaining service life of key load-bearing components based on historical load data and an exponential decay model.

[0130] In this invention, the sensing data specifically includes:

[0131] Core assessment data includes strain and displacement data. Through normalization, confidence assessment, and weighted fusion, a comprehensive load state vector is generated, which serves as the data input for subsequent calculations of the comprehensive load state assessment index, dynamic early warning threshold, and remaining service life.

[0132] Auxiliary monitoring data includes pressure data, tilt angle, and environmental data (temperature, humidity, wind speed). This data is not involved in fusion and evaluation calculations but is used to display independently on the visualization interface, providing comprehensive operating condition information. Independent alarm thresholds (such as platform tilt angle exceeding limits or wind speed exceeding limits) can be set to trigger special safety alarms unrelated to the system.

[0133] This classification approach ensures the focus and accuracy of the core assessment model while also taking into account other dimensions of construction safety.

[0134] like Figure 2 As shown, the sensor sensing module is deployed in a distributed manner at key stress-bearing parts of the manufacturing machine, including: strain sensors: monitoring local structural deformation for stress calculation; displacement sensors: monitoring overall structural deformation; pressure sensors: monitoring the working status of the hydraulic system; tilt sensors: monitoring the platform's horizontal status; and environmental sensors: monitoring environmental parameters such as temperature, humidity, and wind speed.

[0135] The edge computing and communication module is deployed in the control cabinet of the manufacturing machine. Its main functions include: data preprocessing, local data fusion, protocol conversion and communication transmission, and data is uploaded via 5G / industrial WiFi.

[0136] The cloud-based digital twin engine module is deployed on a cloud server cluster. Its main functions include: digital twin construction, assigning material properties and connection relationships, and multiphysics simulation: integrating finite element, hydraulic, and kinematic models, and real-time data-driven synchronization between the twin and the physical entity.

[0137] The core module of intelligent assessment and decision-making is deployed on a cloud server. Its main functions include: calculation of the comprehensive load status assessment index, generation and management of dynamic early warning thresholds, generation of multi-level early warning triggers and handling suggestions, and prediction of remaining service life.

[0138] The visualization module is deployed in the cloud and can be accessed through multiple terminals. Its main functions include: 3D digital twin model, real-time data monitoring panel, visualization of early warning information, and historical data query and playback.

[0139] Any combination of one or more computer-readable media may be used. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device.

[0140] The examples described herein are merely preferred embodiments of the invention and are not intended to limit the concept and scope of the invention. Any modifications and improvements made by those skilled in the art to the technical solutions of the invention without departing from the design concept of the invention should fall within the protection scope of the invention.

Claims

1. A method for assessing and warning the load status of factory machinery based on digital twins, characterized in that, Includes the following steps: Step S1: Construct a digital twin model of the manufacturing machine; Step S2: Collect multi-source sensor data on the physical entity of the manufacturing machine, and map the multi-source sensor data to the digital twin model to drive the state of the digital twin model to be updated synchronously with the physical entity; Step S3: Perform adaptive weighted fusion processing based on confidence assessment on the collected multi-source sensor data to generate a comprehensive load state vector; Step S4: Calculate the comprehensive load state evaluation index based on the generated comprehensive load state vector; Step S5: Generate the dynamic early warning threshold for the current moment based on the dynamic early warning threshold model, compare the calculated load state comprehensive evaluation index with the dynamic early warning threshold, and trigger graded early warnings based on the comparison results; Step S6: Based on historical load data and the generated comprehensive load state vector, predict the remaining service life of key load-bearing components; Step S7: Display the graded early warning information and the predicted remaining service life results in real time through a visual interface.

2. The method according to claim 1, characterized in that, Step S1 specifically includes the following steps: Step S1.1: Extract geometric skeleton information based on the building information model of the factory machine; Step S1.2: Assign material properties to the structural components in the geometric skeleton and define the connection relationships between the components. The material properties include elastic modulus, density and yield strength, and the connection relationships include hinged or fixed connections. Step S1.3: Integrate the structural finite element analysis model, the hydraulic system dynamic model, and the platform kinematic model into the digital twin model.

3. The method according to claim 2, characterized in that, Step S2 specifically includes the following steps: Step S2.1: Establish the mapping relationship between the real-time data stream collected by the physical sensors and the corresponding state parameters in the digital twin model; Step S2.2: Based on the mapping relationship, input the real-time collected sensor data into the digital twin model to drive it to perform state calculation and update.

4. The method according to claim 3, characterized in that, Step S3 specifically includes the following steps: Step S3.1: Collect strain and displacement data output by sensors deployed on the main steel platform, hydraulic support points and walking mechanism; simultaneously collect pressure, tilt angle and environmental data for independent system status monitoring and visualization. Step S3.2: Perform normalization preprocessing on the collected raw strain, displacement, and pressure data; Step S3.3: Real-time assessment of the confidence level of each strain and displacement data source. The assessment criteria include signal-to-noise ratio, deviation between the data and the values ​​calculated by the digital twin model, and sensor health status. Step S3.4: Dynamically allocate fusion weights according to the confidence level of each data source, perform weighted fusion calculation, and generate the comprehensive load state vector.

5. The method according to claim 4, characterized in that, In step S4, the comprehensive evaluation index of the load state is a weighted sum of the stress safety factor, deformation safety factor, shear safety factor, and fatigue damage accumulation factor; wherein the stress safety factor, deformation safety factor, and shear safety factor are the ratios of the real-time measured values ​​of the corresponding physical quantities to the design allowable values.

6. The method according to claim 5, characterized in that, In step S5, the dynamic early warning threshold model is constructed based on a baseline threshold, construction cycle, cumulative equipment operating time, and design life parameters. The output value fluctuates periodically with the construction phase and decreases linearly with increasing equipment usage time. The specific formula is as follows: ； in, This represents the dynamic warning threshold at time t. Indicates the baseline warning threshold. This represents the influence coefficient of periodic loads. This represents a typical complete construction cycle. Indicates the aging and degradation coefficient of the equipment. This indicates the cumulative operating time of the equipment since it was put into use. This indicates the total design life of the equipment.

7. The method according to claim 6, characterized in that, In step S5, the tiered early warning includes: The first-level warning is triggered when the comprehensive load status evaluation index exceeds the dynamic warning threshold, and the system provides a status prompt. The second-level warning is triggered when the comprehensive evaluation index of the load state exceeds the first preset limit, and the system suggests adjusting or suspending the current operation. The third-level warning is triggered when the comprehensive evaluation index of the load status exceeds the second preset limit, and the system forcibly locks the relevant equipment.

8. The method according to claim 7, characterized in that, In step S6, the remaining service life is predicted using an exponential decay model, which multiplies the total designed life of the equipment by a decay factor; the decay factor is calculated based on a negative exponential function of the natural constant.

9. A plant machinery load condition assessment and early warning system based on digital twins, used to implement the plant machinery load condition assessment and early warning method based on digital twins as described in any one of claims 1-8, characterized in that, include: Sensor sensing modules are distributed across the stress-bearing parts of the manufacturing machine to collect strain, displacement, pressure, tilt angle, and environmental data in real time. The edge computing and communication module, deployed on the main body of the manufacturing machine, is used for preprocessing, local fusion, and uploading of sensor data via wireless network; The cloud-based digital twin engine module is used to build, maintain, and run digital twin models, and to receive data to drive model synchronization. The core module of intelligent assessment and decision-making is used to calculate the comprehensive assessment index, manage dynamic early warning thresholds, trigger graded early warnings, and predict the remaining lifespan based on the fused load data. The visualization module is used to display system status and early warning information.

10. The system according to claim 9, characterized in that, The core module for intelligent assessment and decision-making specifically includes: The load state comprehensive evaluation index calculation unit is used to receive the comprehensive load state vector and calculate the load state comprehensive evaluation index. The dynamic early warning threshold management unit is used to dynamically calculate and update the dynamic early warning threshold based on the current construction cycle time, the cumulative equipment running time, and preset parameters. A multi-level early warning triggering unit is used to compare the comprehensive load status evaluation index with the dynamic early warning threshold in real time, and trigger the corresponding early warning level and related handling suggestions. The remaining service life prediction unit is used to calculate and output the predicted value of the remaining service life of key load-bearing components based on historical load data and an exponential decay model.

Images