A multi-scale digital twin early warning system

Abstract

This invention relates to the field of situational awareness technology, and in particular to a multi-scale digital twin early warning system. The system collects target early warning data through a data acquisition module, performs hierarchical denoising and feature extraction through a data processing module, dynamically feeds micro-evolutionary information to a macro-model through a multi-scale modeling module, constructs a quantitative risk index through a risk assessment module, and integrates and dynamically visualizes multi-source data through a situational awareness module. This transforms complex equipment status information into intuitive situational awareness and presents risk levels in real time, significantly improving the efficiency of early warning.

Description

Technical Field

[0001] This invention relates to the field of situational awareness technology, and in particular to a multi-scale digital twin early warning system. Background Technology

[0002] In the energy and power sector, the health status of critical equipment connections such as high-voltage control valves directly impacts unit safety. Traditional monitoring methods primarily rely on single vibration or temperature signals, which has significant limitations. They suffer from limited monitoring dimensions, blind spots, and time-consuming model calculations, often relying on static design models that fail to reflect the dynamic damage evolution of equipment in real time. Risk assessment is also lagging; traditional threshold alarms lack multi-dimensional predictive information fusion, hindering early warning capabilities. Information is fragmented, and traditional monitoring interfaces cannot intuitively display multi-scale risk situations from macro to micro levels, thus limiting decision-making efficiency.

[0003] Chinese patent application CN120725331A discloses a safety early warning method for construction projects involving tunneling under existing important buildings based on digital twins. The method includes: collecting historical engineering case data to construct an engineering case database, whereby the historical engineering case data includes environmental data, construction parameters, and construction monitoring data; predicting the environmental data and construction parameters in the engineering case database using a risk prediction model to obtain prediction results; determining the relationship between risk levels based on the environmental data and construction monitoring data in the engineering case database, and constructing a risk assessment system based on a neural network; inputting the prediction results into the risk assessment system for safety early warning; and issuing early warning information if the corresponding risk level conditions are triggered. However, this approach still suffers from insufficient reliability and low efficiency in equipment risk early warning due to the lack of multi-scale modeling and situational awareness. Summary of the Invention

[0004] To address this, the present invention provides a multi-scale digital twin early warning system to overcome the problems of insufficient reliability and low efficiency in equipment risk early warning caused by the lack of multi-scale modeling and situational awareness in the prior art.

[0005] To achieve the above objectives, the present invention provides a multi-scale digital twin early warning system, the system comprising: The data acquisition module is used to collect target early warning data; The data processing module is used to preprocess the target warning data to obtain the target warning data characteristics; The multi-scale modeling module is used to construct a multi-scale digital twin model based on the characteristics of target early warning data, adjust the multi-scale digital twin model, and acquire a multi-scale comprehensive information set based on the multi-scale digital twin model. The risk assessment module is used to conduct risk assessment and early warning based on a multi-scale comprehensive information set, and obtain the risk assessment and early warning results; The situation awareness module is used to perform real-time situation awareness based on multi-scale integrated information sets and risk assessment and early warning results.

[0006] Furthermore, the data processing module preprocesses the target early warning data using a preprocessing method, which includes: Step A01: Perform preliminary filtering on the target warning data using a bandpass filter to obtain filtered warning data; Step A02: The filtered early warning data is precisely denoised using wavelet transform to obtain cleaned early warning data; Step A03: Extract features from the cleaned early warning data to obtain the target early warning data features.

[0007] Furthermore, the multi-scale modeling module includes: The model building unit is used to build a multi-scale digital twin model based on the characteristics of the target early warning data; The model evaluation unit is used to obtain the data overlap based on the target early warning data, and to adjust the model in the construction process of the multi-scale digital twin model based on the data overlap. The model output unit is used to acquire a multi-scale integrated information set based on the multi-scale digital twin model.

[0008] Furthermore, the model building unit constructs a multi-scale digital twin model based on the characteristics of the target early warning data using a multi-scale digital twin model construction method, the multi-scale digital twin model construction method including: Step B01: Collect topological information using a 3D scanner, and construct a 3D geometric model based on the topological information and target warning data characteristics; Step B02: Perform macroscopic finite element analysis on the three-dimensional geometric model to obtain the first geometric model. Perform microscopic molecular dynamics simulation on the first geometric model to obtain the simulated geometric model. Perform multi-scale coupling on the simulated geometric model through the homogenization method to obtain the multi-scale digital twin model.

[0009] Furthermore, the model evaluation unit acquires the data overlap, compares the model's simulation output data with the target early warning data input data to obtain the data overlap Cs output by the data comparison model.

[0010] Furthermore, the model evaluation unit adjusts the construction process of the multi-scale digital twin model based on the data overlap degree. It compares the data overlap degree Cs with a preset data overlap degree Cs0, setting Cs0=0.9. Based on the comparison result, it judges the degree of data overlap compliance and adjusts the construction process of the multi-scale digital twin model accordingly. When Cs≥Cs0, the model evaluation unit determines that the data overlap meets the standard and does not adjust the model during the construction process of the multi-scale digital twin model; When Cs < Cs0, the model evaluation unit determines that the data overlap degree does not meet the standard, and adjusts the model in the construction process of the multi-scale digital twin model: the homogenization method is coupled to obtain the optimized homogenization method, and the simulated geometric model is re-coupled in multiple scales through the optimized homogenization method.

[0011] Furthermore, the model output unit inputs the target early warning data features into the multi-scale digital twin model to obtain the multi-scale comprehensive information set output by the multi-scale digital twin model.

[0012] Furthermore, the risk assessment module performs real-time situational awareness based on a multi-scale integrated information set, inputs the multi-scale integrated information set and target early warning data features into the risk assessment and early warning model, and obtains the risk value Fc output by the risk assessment and early warning model.

[0013] Furthermore, the risk assessment module compares the risk value Fc with a first preset risk value Fc1 and a second preset risk value Fc2, setting Fc1=0.3 and Fc2=0.6. Based on the comparison result, it judges the status of the risk value and outputs a risk assessment warning result based on the judgment result, wherein: When Fc≤Fc1, the risk assessment module determines the risk value to be in a low-risk state and does not output the risk assessment warning result; When Fc1 < Fc ≤ Fc2, the risk assessment module determines the risk value to be of medium risk and outputs the risk assessment warning result: increase the monitoring frequency and prepare maintenance plans. When Fc > Fc2, the risk assessment module determines that the risk value is in a high-risk state and outputs the risk assessment warning result: issuing a warning through an audible and visual alarm and initiating emergency maintenance.

[0014] Furthermore, the situation awareness module performs real-time situation awareness based on the multi-scale integrated information set and risk assessment and early warning results, and performs situation visualization processing on the multi-scale integrated information set and risk assessment and early warning results to obtain a situation visualization interface.

[0015] Compared with existing technologies, the beneficial effects of this invention are as follows: The system collects target early warning data through a data acquisition module and performs hierarchical denoising and feature extraction on the target early warning data through a data processing module. This effectively suppresses noise and impulse interference while improving the quality and value of early warning data, providing high-quality data input for subsequent prediction and early warning. The system also achieves dynamic feedback of micro-evolutionary information to macro-models through a multi-scale modeling module, forming a unified and evolving digital twin. This provides core model support for accurately predicting structural lifespan and achieving predictive maintenance, fundamentally improving the predictability and reliability of equipment risk early warning. Furthermore, the system constructs a quantitative risk index through a risk assessment module and classifies it into three levels based on clear physical thresholds. This transforms complex operating conditions into intuitive low, medium, and high risk levels, providing a decision-making basis for accurate early warning and hierarchical response. This significantly improves the scientific nature and operability of equipment safety management and effectively avoids over-maintenance or insufficient protection. Finally, the system integrates and dynamically visualizes multi-source data through a situational awareness module, transforming complex equipment status information into intuitive situational awareness and presenting risk levels in real time, significantly improving the efficiency of early warning. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the structure of the multi-scale digital twin early warning system in this embodiment. Detailed Implementation

[0017] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0018] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0019] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0020] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0021] Please see Figure 1 As shown, this is a schematic diagram of the structure of the multi-scale digital twin early warning system of this embodiment. The system includes: The data acquisition module is used to collect target early warning data; A data processing module is used to preprocess the target warning data to obtain the target warning data features. The data processing module is connected to the data acquisition module. The multi-scale modeling module is used to construct a multi-scale digital twin model based on the characteristics of target early warning data, adjust the multi-scale digital twin model, and acquire a multi-scale comprehensive information set based on the multi-scale digital twin model. The multi-scale modeling module is connected to the data processing module. The risk assessment module is used to perform risk assessment and early warning based on a multi-scale integrated information set, and to obtain the risk assessment and early warning results. The risk assessment module is connected to the multi-scale modeling module. The situation awareness module is used to perform real-time situation awareness based on a multi-scale integrated information set and risk assessment and early warning results. The situation awareness module is connected to the risk assessment module.

[0022] Specifically, the multi-scale digital twin early warning system is applied to energy and power equipment, such as high-voltage valve connectors. The system collects target early warning data through a data acquisition module and performs hierarchical denoising and feature extraction through a data processing module. This effectively suppresses noise and impulse interference while improving the quality and value of early warning data, providing high-quality data input for subsequent prediction and early warning. The system also uses a multi-scale modeling module to dynamically feed micro-evolutionary information to a macro-model, forming a unified, evolving digital twin. This provides core model support for accurately predicting structural lifespan and achieving predictive maintenance, fundamentally improving the predictability and reliability of equipment risk early warning. Furthermore, the system constructs a quantitative risk index through a risk assessment module and classifies complex operating conditions into intuitive low, medium, and high risk levels based on clear physical thresholds. This provides a basis for decision-making for accurate early warning and graded response, significantly improving the scientific nature and operability of equipment safety management and effectively avoiding over-maintenance or insufficient protection. Finally, the system integrates and dynamically visualizes multi-source data through a situational awareness module, transforming complex equipment status information into intuitive situational awareness and presenting risk levels in real time, significantly improving early warning efficiency.

[0023] Specifically, the data acquisition module collects target early warning data.

[0024] Specifically, the target early warning data includes vibration, temperature, strain, and acoustic emission signals. The vibration refers to the macroscopic, overall mechanical oscillation phenomenon generated by high-adjustment valve connectors, such as valve bodies, bolts, and pins, under the action of internal steam excitation force and other dynamic loads. In this embodiment, vibration is collected by an accelerometer. The temperature refers to the thermal state of the working environment of the high-adjustment valve connector. In this embodiment, the temperature is collected by a thermocouple. The strain refers to the relative deformation of a material under stress. In this embodiment, the strain is collected by a fiber optic grating sensor. The acoustic emission signal refers to the transient elastic wave generated when energy is suddenly released from the material due to damage, such as crack propagation, dislocation movement, phase transition, or fiber breakage. In this embodiment, the acoustic emission signal is collected by a piezoelectric acoustic emission sensor.

[0025] Specifically, the data processing module preprocesses the target early warning data using a preprocessing method, which includes: Step A01: Perform preliminary filtering on the target warning data using a bandpass filter to obtain filtered warning data; Step A02: The filtered early warning data is precisely denoised using wavelet transform to obtain cleaned early warning data; Step A03: Extract features from the cleaned early warning data to obtain the target early warning data features.

[0026] Specifically, the preliminary filtering refers to the process of initially removing noise from the target warning data; the precise denoising refers to the process of further refining the removal of noise from the filtered warning data through wavelet transform; and the feature extraction refers to the process of extracting time-domain and frequency-domain features from the cleaned warning data. The time-domain features refer to statistical indicators calculated from the waveform of the signal changing over time, and the frequency-domain features refer to the frequency structure describing the signal. This embodiment does not limit the specific method of feature extraction, and those skilled in the art can freely choose according to actual needs, such as statistically analyzing the signal peak value and extracting the frequency-domain features through Fourier transform.

[0027] Specifically, the data processing module effectively suppresses noise and impulse interference while improving the quality and value of early warning data through hierarchical denoising and feature extraction, so as to provide high-quality data input for subsequent prediction and early warning, thereby improving the accuracy of prediction and early warning.

[0028] Specifically, the multi-scale modeling module includes: The model building unit is used to build a multi-scale digital twin model based on the characteristics of the target early warning data; The model evaluation unit is used to acquire the data overlap and adjust the model construction process of the multi-scale digital twin model based on the data overlap. The model output unit is used to acquire a multi-scale integrated information set based on the multi-scale digital twin model.

[0029] Specifically, the model building unit constructs a multi-scale digital twin model based on the characteristics of the target early warning data using a multi-scale digital twin model construction method. The multi-scale digital twin model construction method includes: Step B01: Collect topological information using a 3D scanner, and construct a 3D geometric model based on the topological information and target warning data characteristics; Step B02: Perform macroscopic finite element analysis on the three-dimensional geometric model to obtain the first geometric model. Perform microscopic molecular dynamics simulation on the first geometric model to obtain the simulated geometric model. Perform multi-scale coupling on the simulated geometric model through the homogenization method to obtain the multi-scale digital twin model.

[0030] Specifically, the topological information refers to structured data describing the connections, assembly, and spatial relationships between the various components of the valve body. The macroscopic finite element analysis refers to the process of simulating the mechanical response of the high-adjustment valve as a whole structure under complex loads. This embodiment does not limit the specific method of macroscopic finite element analysis. Those skilled in the art can freely choose according to actual needs, such as discretizing the high-adjustment valve into a large number of small and simple units, such as tetrahedrons and hexahedrons, to test the response of each unit. The microscopic molecular dynamics simulation refers to the process of studying the dynamic behavior and time evolution of the high-adjustment valve material at the atomic and molecular scale. The multi-scale coupling refers to the process of establishing a bidirectional information transmission channel between the macroscopic and microscopic models using a homogenization method, such as passing the local average quantity and updated material properties output by the microscopic model to the macroscopic model, and replacing or modifying the original material parameters of the corresponding region in the macroscopic FEA model.

[0031] Specifically, the model building unit uses a multi-scale coupling algorithm to achieve dynamic feedback of micro-evolutionary information to the macro-model, forming a unified and evolving digital twin. This provides core model support for accurately predicting structural lifespan and achieving predictive maintenance, fundamentally improving the predictability and reliability of equipment health management.

[0032] Specifically, the model evaluation unit acquires the data overlap, compares the model's simulation output data with the target early warning data input data, and obtains the data overlap Cs output by the data comparison model.

[0033] Specifically, the simulation output data of the model refers to the target early warning data obtained by running the multi-scale digital twin model, and the data comparison model refers to the cross-attention network model that takes the simulation output data and the target early warning data as input data and the data overlap as output data. The data overlap is a numerical value that measures the degree of overlap between the data of the multi-scale digital twin model and the real physical data.

[0034] Specifically, the model evaluation unit adjusts the construction process of the multi-scale digital twin model based on the data overlap. It compares the data overlap Cs with a preset data overlap Cs0, setting Cs0=0.9. Based on the comparison result, it judges the degree of data overlap compliance and adjusts the construction process of the multi-scale digital twin model accordingly. When Cs≥Cs0, the model evaluation unit determines that the data overlap meets the standard and does not adjust the model during the construction process of the multi-scale digital twin model; When Cs < Cs0, the model evaluation unit determines that the data overlap degree does not meet the standard, and adjusts the model in the construction process of the multi-scale digital twin model: the homogenization method is coupled to obtain the optimized homogenization method, and the simulated geometric model is re-coupled in multiple scales through the optimized homogenization method.

[0035] Specifically, the preset data overlap refers to a preset value for judging the degree of data overlap. The degree of data overlap includes meeting the standard and not meeting the standard. This embodiment does not limit the specific way of coupling the homogenization method. Those skilled in the art can freely choose according to actual needs, such as tightening the time step of information transmission in the homogenization method.

[0036] Specifically, the model evaluation unit judges the degree of data overlap to facilitate timely correction of the construction process of the multi-scale digital twin model and improve the accuracy of the multi-scale digital twin model.

[0037] Specifically, the model output unit inputs the target early warning data features into the multi-scale digital twin model to obtain the multi-scale comprehensive information set output by the multi-scale digital twin model.

[0038] Specifically, the risk assessment module performs real-time situational awareness based on a multi-scale integrated information set, inputs the multi-scale integrated information set and target early warning data features into the risk assessment and early warning model, and obtains the risk value Fc output by the risk assessment and early warning model.

[0039] Specifically, the risk assessment and early warning model refers to a recurrent neural network model that uses a multi-scale integrated information set and target early warning data features as input data and risk values ​​as output data. This embodiment does not limit the specific construction method of the risk assessment and early warning model. Those skilled in the art can freely choose according to actual needs, such as using historical multi-scale integrated information sets and target early warning data features as training sets to train the recurrent neural network model to obtain the risk assessment and early warning model. The risk value refers to the numerical value obtained from the risk assessment and early warning model that measures the degree of risk of component failure.

[0040] Specifically, the risk assessment module compares the risk value Fc with a first preset risk value Fc1 and a second preset risk value Fc2, setting Fc1=0.3 and Fc2=0.6. Based on the comparison result, it judges the status of the risk value and outputs a risk assessment warning result based on the judgment result, wherein: When Fc≤Fc1, the risk assessment module determines the risk value to be in a low-risk state and does not output the risk assessment warning result; When Fc1 < Fc ≤ Fc2, the risk assessment module determines the risk value to be of medium risk and outputs the risk assessment warning result: increase the monitoring frequency and prepare maintenance plans. When Fc > Fc2, the risk assessment module determines that the risk value is in a high-risk state and outputs the risk assessment warning result: issuing a warning through an audible and visual alarm and initiating emergency maintenance.

[0041] Specifically, the first preset risk value refers to the lower limit of the preset value for judging the state of the risk value, the second preset risk value refers to the upper limit of the preset value for judging the state of the risk value, the state of the risk value includes low risk, medium risk and high risk, and the maintenance plan refers to the preset plan for maintenance personnel to troubleshoot faults based on the state of the risk value.

[0042] Specifically, the risk assessment module constructs a quantitative risk index and classifies it into three levels based on clear physical thresholds, transforming complex working conditions into intuitive low, medium, and high risk levels. This provides a basis for decision-making for accurate early warning and graded response, significantly improving the scientific nature and operability of equipment safety management, and effectively avoiding over-maintenance or insufficient protection.

[0043] Specifically, the situation awareness module performs real-time situation awareness based on the multi-scale integrated information set and risk assessment and early warning results, and performs situation visualization processing on the multi-scale integrated information set and risk assessment and early warning results to obtain a situation visualization interface.

[0044] Specifically, the situation visualization processing refers to the process of displaying the changing situation in an intuitive form by integrating multi-scale comprehensive information sets and risk assessment and early warnings. The changing situation refers to the state of multi-scale risk distribution. This embodiment does not limit the specific method of situation visualization processing. Those skilled in the art can freely choose according to actual needs, such as displaying the multi-scale comprehensive information sets and risk assessment and early warning results through a dashboard and using color coding, such as green, yellow and red, to represent the situation level.

[0045] Specifically, the situational awareness module transforms complex equipment status information into intuitive situational awareness through multi-source data integration and dynamic visualization, and presents the risk level in real time, significantly improving the efficiency of early warning.

[0046] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A multi-scale digital twin early warning system, characterized in that, The system includes: The data acquisition module is used to collect target early warning data; The data processing module is used to preprocess the target warning data to obtain the target warning data characteristics; The multi-scale modeling module is used to construct a multi-scale digital twin model based on the characteristics of target early warning data, adjust the multi-scale digital twin model, and acquire a multi-scale comprehensive information set based on the multi-scale digital twin model. The risk assessment module is used to conduct risk assessment and early warning based on a multi-scale comprehensive information set, and obtain the risk assessment and early warning results; The situation awareness module is used to perform real-time situation awareness based on multi-scale integrated information sets and risk assessment and early warning results.

2. The multi-scale digital twin early warning system according to claim 1, characterized in that, The data processing module preprocesses the target early warning data using a preprocessing method, which includes: Step A01: Perform preliminary filtering on the target warning data using a bandpass filter to obtain filtered warning data; Step A02: The filtered early warning data is precisely denoised using wavelet transform to obtain cleaned early warning data; Step A03: Extract features from the cleaned early warning data to obtain the target early warning data features.

3. The multi-scale digital twin early warning system according to claim 2, characterized in that, The multi-scale modeling module includes: The model building unit is used to build a multi-scale digital twin model based on the characteristics of the target early warning data; The model evaluation unit is used to obtain the data overlap based on the target early warning data, and to adjust the model in the construction process of the multi-scale digital twin model based on the data overlap. The model output unit is used to acquire a multi-scale integrated information set based on the multi-scale digital twin model.

4. The multi-scale digital twin early warning system according to claim 3, characterized in that, The model building unit constructs a multi-scale digital twin model based on the characteristics of the target early warning data using a multi-scale digital twin model construction method. The multi-scale digital twin model construction method includes: Step B01: Collect topological information using a 3D scanner, and construct a 3D geometric model based on the topological information and target warning data characteristics; Step B02: Perform macroscopic finite element analysis on the three-dimensional geometric model to obtain the first geometric model. Perform microscopic molecular dynamics simulation on the first geometric model to obtain the simulated geometric model. Perform multi-scale coupling on the simulated geometric model through the homogenization method to obtain the multi-scale digital twin model.

5. The multi-scale digital twin early warning system according to claim 4, characterized in that, The model evaluation unit acquires the data overlap, compares the model's simulation output data with the target early warning data input data, and obtains the data overlap Cs output by the data comparison model.

6. The multi-scale digital twin early warning system according to claim 5, characterized in that, The model evaluation unit adjusts the construction process of the multi-scale digital twin model based on the data overlap. It compares the data overlap Cs with a preset data overlap Cs0, setting Cs0=0.

9. Based on the comparison result, it judges the degree of data overlap compliance and adjusts the construction process of the multi-scale digital twin model accordingly. When Cs≥Cs0, the model evaluation unit determines that the data overlap meets the standard and does not adjust the model during the construction process of the multi-scale digital twin model; When Cs < Cs0, the model evaluation unit determines that the data overlap degree does not meet the standard, and adjusts the model in the construction process of the multi-scale digital twin model: the homogenization method is coupled to obtain the optimized homogenization method, and the simulated geometric model is re-coupled in multiple scales through the optimized homogenization method.

7. The multi-scale digital twin early warning system according to claim 6, characterized in that, The model output unit inputs the target early warning data features into the multi-scale digital twin model to obtain the multi-scale comprehensive information set output by the multi-scale digital twin model.

8. The multi-scale digital twin early warning system according to claim 7, characterized in that, The risk assessment module performs real-time situational awareness based on a multi-scale integrated information set, inputs the multi-scale integrated information set and target early warning data features into the risk assessment and early warning model, and obtains the risk value Fc output by the risk assessment and early warning model.

9. The multi-scale digital twin early warning system according to claim 8, characterized in that, The risk assessment module compares the risk value Fc with a first preset risk value Fc1 and a second preset risk value Fc2, setting Fc1=0.3 and Fc2=0.

6. Based on the comparison result, it judges the status of the risk value and outputs a risk assessment warning result based on the judgment result. When Fc≤Fc1, the risk assessment module determines the risk value to be in a low-risk state and does not output the risk assessment warning result; When Fc1 < Fc ≤ Fc2, the risk assessment module determines the risk value to be of medium risk and outputs the risk assessment warning result: increase the monitoring frequency and prepare maintenance plans. When Fc > Fc2, the risk assessment module determines that the risk value is in a high-risk state and outputs the risk assessment warning result: issuing a warning through an audible and visual alarm and initiating emergency maintenance.

10. The multi-scale digital twin early warning system according to claim 9, characterized in that, The situation awareness module performs real-time situation awareness based on the multi-scale integrated information set and risk assessment and early warning results, and performs situation visualization processing on the multi-scale integrated information set and risk assessment and early warning results to obtain a situation visualization interface.

Images