Complex equipment IETM system based on digital twinning

Abstract

The invention discloses a complex equipment IETM system based on digital twinning. The complex equipment IETM system comprises the following steps that S1, multi-source heterogeneous data are collected and preprocessed; s2, constructing a multi-scale space-time coupled digital twin model to realize real-time modeling of equipment; s3, utilizing non-Euclidean space mapping to establish a dynamic evolution track of the equipment; s4, the information entropy is calculated and the model parameters are dynamically adjusted by monitoring the equipment state in real time; s5, generating a fault prediction and maintenance scheme, and optimizing the operation state of the equipment; s6, displaying an equipment state and a maintenance suggestion through an IETM system, and providing equipment management support; and S7, continuously optimizing the twin model by the system. According to the invention, the equipment maintenance precision is improved, and the use efficiency, the maintenance convenience and the information acquisition accuracy of the equipment are improved in real time.

Description

Technical Field

[0001] The present invention relates to the field of digital twin technology, and in particular to a complex equipment IETM system based on digital twin. Background Art

[0002] As the intelligence and complexity of industrial equipment continue to increase, equipment operation status monitoring, fault prediction and maintenance optimization have become key links that affect the normal operation and production efficiency of equipment. Traditional equipment maintenance methods mostly rely on regular inspections and manual maintenance, which is not only inefficient, but may also increase the risk of equipment failure, especially under complex working conditions, and cannot achieve accurate maintenance and real-time optimization of equipment. In order to improve the level of intelligence in equipment maintenance, the interactive electronic technical manual (IETM) system was introduced into equipment management. The IETM system helps operators better maintain equipment by providing maintenance information, operating procedures and fault diagnosis solutions. However, existing IETM systems usually only support static information display, lack the ability to dynamically update real-time data, and are difficult to cope with emergencies and complex working conditions during equipment operation.

[0003] In recent years, digital twin technology has gradually become an important technology for improving equipment maintenance and management. By building a virtual digital model corresponding to the physical equipment, digital twin technology can simulate, monitor, and analyze the equipment status based on the real-time equipment operation data, and provide predictive maintenance support. However, in current applications, digital twin technology still has some shortcomings, especially in dealing with multi-source data fusion, nonlinear dynamic change analysis, and real-time feedback control of complex equipment, making it difficult to achieve comprehensive equipment status optimization.

[0004] It is difficult to integrate and process multi-source heterogeneous data. When complex equipment is in operation, it will generate multi-source data from different sensors, operation logs and maintenance records. These data have large differences in time, space and format. The existing equipment maintenance system cannot effectively integrate and process these heterogeneous data, resulting in low data analysis efficiency and inability to achieve real-time monitoring and prediction of equipment status.

[0005] Insufficient dynamic modeling capabilities. The operating status of complex equipment changes dynamically, especially at different time scales and spatial dimensions, and its state changes are highly complex and uncertain. Existing digital twin technology can usually only model the static or semi-dynamic state of the equipment, and cannot accurately capture the dynamic evolution of the equipment, resulting in low accuracy in equipment failure prediction under complex working conditions.

[0006] Lack of modeling capabilities for nonlinear state changes. The operating state of complex equipment under extreme conditions often exhibits significant nonlinear characteristics. Existing models are difficult to capture these nonlinear changes, especially when the equipment state changes suddenly or fluctuates violently. Existing modeling methods cannot accurately predict the state changes of the equipment, increasing the risk of equipment failure.

[0007] The information feedback mechanism is imperfect. Existing digital twin systems usually lack real-time feedback mechanisms and cannot dynamically adjust the key parameters of the model according to changes in equipment status. During the operation of the equipment, the changes in equipment status are continuous and dynamic. Traditional feedback mechanisms often lag behind, making it difficult to achieve real-time optimization of equipment status, resulting in reduced equipment operation efficiency.

[0008] The intelligent level of maintenance solutions is low. Existing equipment maintenance systems usually rely on preset maintenance plans and manual judgment, and lack real-time maintenance recommendations based on the actual operating status of the equipment. Even if digital twin technology is used for status monitoring, the system often still requires manual intervention and cannot autonomously generate fault predictions and maintenance solutions based on the real-time status of the equipment, affecting the efficiency and response speed of equipment maintenance.

[0009] Therefore, how to provide a complex equipment IETM system based on digital twins is a problem that technical personnel in this field urgently need to solve. Summary of the invention

[0010] One purpose of the present invention is to propose a complex equipment IETM system based on digital twins, aiming to solve the problems in the prior art such as inaccurate equipment status monitoring, insufficient fault prediction capability, and slow maintenance optimization response speed.

[0011] A complex equipment IETM system based on digital twin according to an embodiment of the present invention includes the following steps:

[0012] S1. Collect multi-source heterogeneous data of complex equipment, including sensor data, operation logs and maintenance records, and divide the data into different time scales and spatial levels according to the equipment operation characteristics for pre-processing;

[0013] S2. Based on the preprocessed data, a multi-scale spatiotemporal coupled digital twin model is constructed to capture the dynamic changes of the equipment in different time and space dimensions and realize real-time modeling of the equipment status;

[0014] S3, using non-Euclidean space mapping, converting multi-dimensional equipment status data into a topological structure, establishing the dynamic evolution trajectory of the equipment, and reflecting the nonlinear changes of the equipment under complex working conditions;

[0015] S4. By real-time monitoring of device status, calculating the change value of information entropy, and using the information entropy feedback mechanism to dynamically adjust the key parameters in the twin model;

[0016] S5. Optimize the equipment operation status according to the adjusted model parameters and generate real-time fault prediction and maintenance plans;

[0017] S6. Display equipment status, fault prediction and maintenance suggestions to users through the IETM system to provide equipment management and decision support;

[0018] S7. During the entire life cycle of the equipment, the system continuously optimizes the twin model through the synergy of multi-scale spatiotemporal coupling and information entropy feedback.

[0019] Optionally, S2 includes the following steps:

[0020] S21, based on the multi-source heterogeneous data preprocessed in S1, including sensor data, operation logs and maintenance records, generates a time-space state matrix M0 (t, x1, x2, ..., x n ), where t is the time dimension, x1,x2,...,x n Represents the state variables of the device in the spatial dimension;

[0021] S22. Perform adaptive variational wavelet transform on the time series data D(t), decompose the time dynamics of the equipment at different scales, and extract the multi-time scale characteristics of the equipment operation. The formula is as follows:

[0022]

[0023] Among them, C i (t) is the variation coefficient of the i-th time scale, u i (t) represents the displacement function of the device time dynamics, s i (t) is the scale factor, ψ is the adaptive wavelet basis function;

[0024] S23, for spatial dimension data x1, x2, ..., x n Perform multi-scale dimensionality reduction, extract the key spatial features of the device through adaptive nonlinear dimensionality reduction, and generate the device spatial feature matrix S LLE (x1,x2,...,x n );

[0025] S24, the time feature matrix T s (t) and the spatial feature matrix S LLE (x1,x2,...,x n ) to conduct spatiotemporal coupling, build a dynamic state model of the device through multi-scale spatiotemporal tensors, and generate a spatiotemporal coupling tensor M st(t,x1,x2,...,x n ), the formula is as follows:

[0026]

[0027] in, Represents element-wise multiplication;

[0028] S25, by introducing a dynamic weight optimization mechanism, the spatiotemporal coupling tensor M st (t,x1,x2,...,x n ) Adaptively weight and dynamically adjust the importance of temporal and spatial features;

[0029] S26, through the optimized space-time coupling tensor M st (t,x1,x2,...,x n ), generate the device state prediction matrix P(t,x1,x2,...,x n ), the prediction formula is as follows:

[0030]

[0031] Among them, λ is the regularization parameter, which is used to control the smoothness of the model.

[0032] Optionally, S3 includes the following steps:

[0033] S31, extracting multi-dimensional state data M(t,x) of the device, where t represents the time dimension, x=(x1,x2,...,x n ) represents the state variables of the device in different spatial dimensions, forming a state matrix M0(t,x);

[0034] S32. Map the device status data based on the Riemann manifold in non-Euclidean geometry. The mapping kernel function is defined as:

[0035]

[0036] Among them, M(t,x) and M(t,x') represent the states of the device at the spatial points x and x' respectively. Represented on a Riemann manifold The metric distance on , σ is the scale parameter;

[0037] S33. Metric tensor g based on Riemannian manifold ij (x), calculate the change of device status in the spatial dimension, the formula is as follows:

[0038]

[0039] Among them, g ij(x) is the Riemann metric tensor, which represents the geometric change of the device in the spatial dimension, M k (t,x) is the state change of the device in the kth dimension, x i and x j are different dimensions of spatial coordinates;

[0040] S34, construct the dynamic evolution trajectory of the device, and describe the evolution trajectory of the device through the geodesic on the Riemann manifold;

[0041] S35. Dynamic topological characteristics of computing devices, using Betti number β k (t) to capture the topological changes of devices in different dimensions;

[0042] S36. Construct a nonlinear state change model of the equipment N e (t), based on the topological structure and dynamic evolution trajectory of the device;

[0043] S37, through the feedback mechanism, the nonlinear state change model N e (t) is fed back to the digital twin model to dynamically adjust the equipment state parameter θ(t).

[0044] Optionally, S4 includes the following steps:

[0045] S41. Extract real-time status data from the multidimensional status matrix M(t,x) of the device, where t is the time dimension, x=(x1,x2,...,x n ) is the state of the device in the spatial dimension;

[0046] S42. Perform probability density estimation on the device state data M(t,x), and use the kernel density estimation KDE method to calculate the probability density function P(M)t,x) of the device state distribution. The formula is as follows:

[0047]

[0048] Among them, Kh is the kernel function of bandwidth h, which represents the smooth distribution of state data, and Mi(t,x) represents the state of the i-th data point;

[0049] S43. Calculate the conditional information entropy H(t|x) of the device state. The formula is as follows:

[0050]

[0051] Among them, P(M(t,x)) represents the probability distribution of the device in the spatiotemporal state, and information entropy is used to quantify the complexity of the device state;

[0052] S44. Based on the change of the information entropy of the device state, define the information entropy change rate ΔH(t) as:

[0053] S45, feeding back the information entropy change rate ΔH(t) to the digital twin model, and adjusting the key parameter θ(t) of the twin model;

[0054] S46. Based on the information entropy feedback results, iteratively update the twin model of the device so that the model can adapt to the dynamic changes of the device status.

[0055] Optionally, S5 includes the following steps:

[0056] S51. Generate the device optimization state matrix M based on the adjusted twin model parameters θ(t) opt (t,x);

[0057] S52, using the device state matrix M opt (t,x), build a fault risk model based on state offset, the formula is as follows:

[0058]

[0059] Among them, λ is the failure risk control coefficient, M safe (x) is the state distribution of the device under the ideal safety state;

[0060] S53. Calculate the predicted time T of equipment failure according to the failure risk model fault , the formula is:

[0061] T fault = inf{t|P fault (t,x)≥P threshold};

[0062] Among them, P threshold The preset fault trigger threshold is used to predict the possible failure of the equipment at a certain point in time;

[0063] S54, Optimize maintenance strategy S maint The generation formula of (t) is as follows:

[0064]

[0065] Among them, C maint (t') is the maintenance cost at time t', γ is the weight factor;

[0066] S55, dynamically update the equipment status matrix M based on the real-time monitoring data opt (t,x) and maintenance strategy S maint (t);

[0067] S56. During the entire life cycle of the equipment, the key parameters θ(t) in the twin model are adjusted in real time by combining the state optimization and fault prediction models;

[0068] S57. Based on the optimized twin model and equipment status matrix, the dynamic operation of the equipment is continuously monitored, and the long-term stable operation and maintenance costs of the equipment are minimized through iterative updates of maintenance strategies.

[0069] Optionally, the S6 includes the following steps:

[0070] S61. Obtain the optimized state matrix and fault prediction data of the equipment from the twin model, and display the equipment status and real-time risk assessment through the multi-dimensional visualization interface of the IETM system;

[0071] S62. Generate an overall risk assessment of the equipment based on the status of each component of the equipment and the fault prediction results, and display it in the user interface of the IETM system through risk mapping;

[0072] S63. Dynamically generate maintenance priorities for equipment, adjust the priorities of various parts of the equipment based on risk assessment results and maintenance opportunities, and display them in the system;

[0073] S64. Through the real-time data interface of the IETM system, users can view equipment status, maintenance recommendations and fault predictions. All information is displayed through interactive graphics and text descriptions, and users can operate based on these data;

[0074] S65. Based on real-time monitoring of equipment status, dynamically adjust equipment management strategies and optimize equipment management decisions through risk assessment results;

[0075] S66, dynamically updating the equipment maintenance suggestions and fault prediction information as the equipment status changes, so that the maintenance plan can be adjusted in real time, and the user can confirm and modify the maintenance suggestions through the interface;

[0076] S67. Provide real-time equipment monitoring and fault prediction through the synergy between the IETM system and the digital twin model.

[0077] A complex equipment IETM system based on digital twins, characterized in that the system includes:

[0078] Data acquisition module, used to collect multi-source heterogeneous data of complex equipment, including sensor data, operation logs, maintenance records, etc., which provide input data for the spatiotemporal coupling modeling of equipment;

[0079] The time-space coupling processing module processes the collected data through a multi-scale time-space coupling mechanism, performs synchronous modeling at different time scales and spatial levels according to the operating status of the equipment, and captures the dynamic state changes of the equipment;

[0080] The digital twin model module uses the multi-scale data generated by the spatiotemporal coupling processing module to build a digital twin model of the equipment, integrating the microscopic state and macroscopic operation trend of the equipment in real time to provide accurate equipment state representation;

[0081] Non-Euclidean mapping module: maps the multi-dimensional state of the equipment to the non-Euclidean space through nonlinear topology, generates a dynamic evolution trajectory model of the equipment under complex working conditions, describes the nonlinear changes of the equipment under extreme conditions, and provides data support for optimizing the equipment state;

[0082] The information entropy feedback module monitors the changes in information entropy during equipment operation, provides real-time feedback on the complexity and uncertainty of equipment status, adjusts key parameters of the digital twin model, and achieves adaptive optimization of equipment status and fault prediction;

[0083] The IETM system module is responsible for integrating the results of the digital twin model, non-Euclidean mapping module and information entropy feedback module, presenting the equipment's status changes, fault predictions and maintenance optimization plans to users through an interactive interface, and providing equipment management and decision support tools.

[0084] The beneficial effects of the present invention are:

[0085] Based on digital twin technology, non-Euclidean space mapping and information entropy feedback mechanism, the present invention proposes an IETM system for complex equipment, which solves the problems of inaccurate equipment status monitoring, insufficient fault prediction capability and slow maintenance response speed in the prior art, and has the following beneficial effects:

[0086] The present invention effectively integrates the sensor data, operation logs and maintenance records of complex equipment through the collection and preprocessing of multi-source heterogeneous data, significantly improving the efficiency and accuracy of data processing. Traditional systems find it difficult to accurately capture dynamic changes in different time scales and spatial levels of equipment, while the multi-scale spatiotemporal coupling digital twin model constructed by the present invention can monitor the status of equipment in real time, especially providing more accurate dynamic feedback on the operating status of equipment under complex working conditions, thereby improving the reliability of status monitoring.

[0087] The introduction of non-Euclidean space mapping enables the multi-dimensional state data of the equipment to be converted into a topological structure, thereby establishing the dynamic evolution trajectory of the equipment. This innovative space mapping method greatly enhances the system's ability to capture the nonlinear changes of complex equipment. Compared with traditional linear methods, the present invention can effectively deal with the nonlinear changes of equipment under extreme working conditions, and improve the accuracy and timeliness of fault prediction.

[0088] The present invention realizes real-time adaptive adjustment of the equipment status through the information entropy feedback mechanism. The system can dynamically adjust the key parameters in the twin model by monitoring the changes in information entropy during the operation of the equipment to ensure that the equipment operation status is always maintained at the optimal level. Compared with the static adjustment method in the prior art, the present invention significantly enhances the feedback capability of the system, making the fault prediction during the operation of the equipment more accurate. At the same time, the maintenance suggestions can also be updated in real time according to the actual situation, ensuring the continuous optimization of the equipment.

[0089] The present invention integrates the display of the IETM system, and users can intuitively view the status, fault prediction and maintenance suggestions of the equipment through the visual interface of the system. Different from the traditional static manual, the IETM system integrated in the present invention can provide the real-time status and dynamic maintenance strategy of the equipment, reduce manual intervention, and significantly improve the intelligent level of equipment management and the efficiency of maintenance decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0090] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0091] Figure 1 This is a flow chart of a complex equipment IETM system based on digital twin proposed by the present invention;

[0092] Figure 2 A flowchart for constructing the multi-scale spatiotemporal coupled digital twin model proposed in the present invention;

[0093] Figure 3 This is a device status topology structure generation diagram based on non-Euclidean space mapping proposed by the present invention. DETAILED DESCRIPTION

[0094] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0095] refer to Figure 1-3 , a complex equipment IETM system based on digital twins, including the following steps:

[0096] S1. Collect multi-source heterogeneous data of complex equipment, including sensor data, operation logs and maintenance records, and divide the data into different time scales and spatial levels according to the equipment operation characteristics for pre-processing;

[0097] S2. Based on the preprocessed data, a multi-scale spatiotemporal coupled digital twin model is constructed to capture the dynamic changes of the equipment in different time and space dimensions and realize real-time modeling of the equipment status;

[0098] S3, using non-Euclidean space mapping, converting multi-dimensional equipment status data into a topological structure, establishing the dynamic evolution trajectory of the equipment, and reflecting the nonlinear changes of the equipment under complex working conditions;

[0099] S4. By real-time monitoring of device status, calculating the change value of information entropy, and using the information entropy feedback mechanism to dynamically adjust the key parameters in the twin model;

[0100] S5. Optimize the equipment operation status according to the adjusted model parameters and generate real-time fault prediction and maintenance plans;

[0101] S6. Display equipment status, fault prediction and maintenance suggestions to users through the IETM system to provide equipment management and decision support;

[0102] S7. During the entire life cycle of the equipment, the system continuously optimizes the twin model through the synergy of multi-scale spatiotemporal coupling and information entropy feedback.

[0103] In this implementation, S2 includes the following steps:

[0104] S21, based on the multi-source heterogeneous data preprocessed in S1, including sensor data, operation logs and maintenance records, generates a time-space state matrix M0 (t, x1, x2, ..., x n ), where t is the time dimension, x1,x2,...,x n Represents the state variables of the device in the spatial dimension;

[0105] S22. Perform adaptive variational wavelet transform on the time series data D(t0), decompose the time dynamics of the equipment at different scales, and extract the multi-time scale characteristics of the equipment operation. The formula is as follows:

[0106]

[0107] Among them, C i (t) is the variation coefficient of the i-th time scale, u i (t0 represents the displacement function of the device time dynamics, s i (t0 is the scale factor, ψ is the adaptive wavelet basis function;

[0108] S23, for spatial dimension data x1, x2, ..., x n Perform multi-scale dimensionality reduction, extract the key spatial features of the device through adaptive nonlinear dimensionality reduction, and generate the device spatial feature matrix S LLE (x1,x2,...,x n );

[0109] S24, the time feature matrix Ts (t) and the spatial feature matrix S LLE (x1,x2,...,x n ) to conduct spatiotemporal coupling, build a dynamic state model of the device through multi-scale spatiotemporal tensors, and generate a spatiotemporal coupling tensor M st (t,x1,x2,...,x n ), the formula is as follows:

[0110]

[0111] in, Represents element-wise multiplication;

[0112] S25, by introducing a dynamic weight optimization mechanism, the spatiotemporal coupling tensor M st (t,x1,x2,...,x n ) Adaptively weight and dynamically adjust the importance of temporal and spatial features;

[0113] S26, through the optimized space-time coupling tensor M st (t,x1,x2,...,x n ), generate the device state prediction matrix P(t,x1,x2,...,x n ), the prediction formula is as follows:

[0114]

[0115] Among them, λ is the regularization parameter, which is used to control the smoothness of the model.

[0116] In this implementation, S3 includes the following steps:

[0117] S31, extracting multi-dimensional state data M(t,x) of the device, where t represents the time dimension, x=x1,x2,...,x n ) represents the state variables of the device in different spatial dimensions, forming a state matrix M0(t,x0;

[0118] S32. Map the device status data based on the Riemann manifold in non-Euclidean geometry. The mapping kernel function is defined as:

[0119]

[0120] Among them, M(t,x0 and M(t,x') represent the states of the device at the spatial points x and x' respectively. Represented on a Riemann manifold The metric distance on , σ is the scale parameter;

[0121] S33. Metric tensor g based on Riemannian manifold ij(x), calculate the change of device status in the spatial dimension, the formula is as follows:

[0122]

[0123] Among them, g ij (x0 is the Riemann metric tensor, which represents the geometric change of the device in the spatial dimension, M k (t,x is the state change of the device in the kth dimension, x i and x j are different dimensions of spatial coordinates;

[0124] S34, construct the dynamic evolution trajectory of the device, and describe the evolution trajectory of the device through the geodesic on the Riemann manifold;

[0125] S35. Dynamic topological characteristics of computing devices, using Betti number β k (t) to capture the topological changes of devices in different dimensions;

[0126] S36. Construct a nonlinear state change model of the equipment N e (t), based on the topological structure and dynamic evolution trajectory of the device;

[0127] S37, through the feedback mechanism, the nonlinear state change model N e (t) is fed back to the digital twin model to dynamically adjust the equipment state parameter θ(t).

[0128] In this implementation, S4 includes the following steps:

[0129] S41. Extract real-time status data from the multidimensional status matrix M(t,x) of the device, where t is the time dimension, x=(x1,x2,...,x n ) is the state of the device in the spatial dimension;

[0130] S42. Perform probability density estimation on the device state data M(t,x0), and use the kernel density estimation KDE method to calculate the probability density function P9M(t,x) of the device state distribution. The formula is as follows:

[0131]

[0132] Among them, K h is the kernel function of bandwidth h, representing the smooth distribution of state data, M i (t,x) represents the state of the i-th data point;

[0133] S43. Calculate the conditional information entropy H(t|x) of the device state. The formula is as follows:

[0134]

[0135] Among them, P(M(t,x)) represents the probability distribution of the device in the spatiotemporal state, and information entropy is used to quantify the complexity of the device state;

[0136] S44. Based on the change of the information entropy of the device state, define the information entropy change rate ΔH(t) as:

[0137] S45, feeding back the information entropy change rate ΔH(t) to the digital twin model, and adjusting the key parameter θ(t) of the twin model;

[0138] S46. Based on the information entropy feedback results, iteratively update the twin model of the device so that the model can adapt to the dynamic changes of the device status.

[0139] In this implementation, S5 includes the following steps:

[0140] S51, based on the adjusted twin model parameters θ ( t ) , generate the device optimization state matrix M opt ( t,x ) ;

[0141] S52, using the device state matrix M opt ) t,x ( , construct a fault risk model based on state offset, the formula is as follows:

[0142]

[0143] Among them, λ is the failure risk control coefficient, M safe (x) is the state distribution of the device under the ideal safety state;

[0144] S53. Calculate the predicted time T of equipment failure according to the failure risk model fault , the formula is:

[0145] T fault = inf{t|P fault (t,x)≥P threshold};

[0146] Among them, P threshold The preset fault trigger threshold is used to predict the possible failure of the equipment at a certain point in time;

[0147] S54, Optimize maintenance strategy S maint The generation formula of (t) is as follows:

[0148]

[0149] Among them, C maint (t') is the maintenance cost at time t', γ is the weight factor;

[0150] S55, dynamically update the equipment status matrix M based on the real-time monitoring data opt ( t,x ) and maintenance strategy S maint ( t);

[0151] S56. During the entire life cycle of the equipment, the key parameters θ(t) in the twin model are adjusted in real time by combining the state optimization and fault prediction models;

[0152] S57. Based on the optimized twin model and equipment status matrix, the dynamic operation of the equipment is continuously monitored, and the long-term stable operation and maintenance costs of the equipment are minimized through iterative updates of maintenance strategies.

[0153] In this implementation, S6 includes the following steps:

[0154] S61. Obtain the optimized state matrix and fault prediction data of the equipment from the twin model, and display the equipment status and real-time risk assessment through the multi-dimensional visualization interface of the IETM system;

[0155] S62. Generate an overall risk assessment of the equipment based on the status of each component of the equipment and the fault prediction results, and display it in the user interface of the IETM system through risk mapping;

[0156] S63. Dynamically generate maintenance priorities for equipment, adjust the priorities of various parts of the equipment based on risk assessment results and maintenance opportunities, and display them in the system;

[0157] S64. Through the real-time data interface of the IETM system, users can view equipment status, maintenance recommendations and fault predictions. All information is displayed through interactive graphics and text descriptions, and users can operate based on these data;

[0158] S65. Based on real-time monitoring of equipment status, dynamically adjust equipment management strategies and optimize equipment management decisions through risk assessment results;

[0159] S66, dynamically updating the equipment maintenance suggestions and fault prediction information as the equipment status changes, so that the maintenance plan can be adjusted in real time, and the user can confirm and modify the maintenance suggestions through the interface;

[0160] S67. Provide real-time equipment monitoring and fault prediction through the synergy between the IETM system and the digital twin model.

[0161] A complex equipment IETM system based on digital twins, characterized in that the system includes:

[0162] Data acquisition module, used to collect multi-source heterogeneous data of complex equipment, including sensor data, operation logs, maintenance records, etc., which provide input data for the spatiotemporal coupling modeling of equipment;

[0163] The time-space coupling processing module processes the collected data through a multi-scale time-space coupling mechanism, performs synchronous modeling at different time scales and spatial levels according to the operating status of the equipment, and captures the dynamic state changes of the equipment;

[0164] The digital twin model module uses the multi-scale data generated by the spatiotemporal coupling processing module to build a digital twin model of the equipment, integrating the microscopic state and macroscopic operation trend of the equipment in real time to provide accurate equipment state representation;

[0165] Non-Euclidean mapping module: maps the multi-dimensional state of the equipment to the non-Euclidean space through nonlinear topology, generates a dynamic evolution trajectory model of the equipment under complex working conditions, describes the nonlinear changes of the equipment under extreme conditions, and provides data support for optimizing the equipment state;

[0166] The information entropy feedback module monitors the changes in information entropy during equipment operation, provides real-time feedback on the complexity and uncertainty of equipment status, adjusts key parameters of the digital twin model, and achieves adaptive optimization of equipment status and fault prediction;

[0167] The IETM system module is responsible for integrating the results of the digital twin model, non-Euclidean mapping module and information entropy feedback module, presenting the equipment's status changes, fault predictions and maintenance optimization plans to users through an interactive interface, and providing equipment management and decision support tools.

[0168] Embodiment 1:

[0169] In the production workshop of a large industrial manufacturing enterprise, the maintenance and management of complex equipment has always been a crucial part of the production process. The company's equipment includes multiple subsystems and precision mechanical parts. For a long time, the traditional regular maintenance method has been adopted, which cannot timely detect potential failures that may occur in the equipment during operation. Especially when the equipment is running under high load and complex environment, the equipment status monitoring accuracy is low, and untimely maintenance often leads to production downtime. In order to solve these problems, the company decided to introduce a complex equipment IETM system based on digital twins to achieve real-time monitoring, fault prediction and maintenance optimization of equipment.

[0170] In actual application, the company deployed the system in a production line containing 20 complex equipment, and the operating status of each equipment was monitored in real time by multiple sensors. Through the system's data acquisition module, the multi-source heterogeneous data of the equipment, including temperature sensors, vibration sensors, operation logs, maintenance records and other information, were transmitted to the digital twin system in real time. The system preprocessed the data, classified and filtered the data according to different time scales and spatial levels, to ensure that the operating data of each device can be processed efficiently.

[0171] Next, the system constructs a digital twin model of each device through a multi-scale spatiotemporal coupling model. The model can dynamically capture the state changes of the device, especially when the device is working under different loads, the real-time evolution of its state is accurately simulated and analyzed. For example, under high temperature and high pressure environments, the temperature and vibration data of the equipment show obvious nonlinear changes, and traditional regular inspections are difficult to detect these hidden dangers. The present invention uses non-Euclidean space mapping technology to convert the multi-dimensional state data of the equipment into a topological structure, thereby establishing a dynamic evolution trajectory of the equipment. This technology can not only accurately capture the nonlinear changes of the equipment, but also predict the possible failure points of the equipment in advance.

[0172] During the operation of a key piece of equipment, the system monitors the status information of the equipment in real time and calculates the change value of information entropy based on the operating data of the equipment. When the information entropy value is abnormal, the system immediately starts the feedback mechanism and automatically adjusts the key parameters in the twin model. By adjusting the parameters, the system optimizes the operating status of the equipment and avoids potential failures of the equipment. In this case, the system also generates a corresponding maintenance plan and displays the maintenance suggestions to the equipment operators through the IETM system, prompting them of potential failures that are about to occur and guiding them to perform preventive maintenance.

[0173] Specific data proves the effectiveness of the system. In the first month of applying the system, the company's equipment failure rate dropped significantly. In the previous month, the equipment had an average of 3 shutdown failures per week, but after the introduction of the system, the equipment failure rate was reduced to only one shutdown every two weeks. In a typical case, the vibration data of equipment 1 suddenly fluctuated greatly. The system discovered an abnormality in the rotating parts of equipment 1 through dynamic analysis, predicted the possible mechanical failure of the equipment two days in advance, and generated a detailed maintenance plan. With the maintenance suggestions generated by the system, the operator replaced the part in time, avoiding production downtime caused by sudden equipment failure.

[0174] Another scenario is that device 3 runs for a long time in a high temperature environment. Traditional maintenance methods can only regularly check the surface temperature and operating status of the device, and cannot accurately capture changes inside the device. However, the system monitors the temperature changes of the device in real time and combines the information entropy feedback mechanism to find that there is an abnormality in the cooling system of device 3. The system generates a fault prediction and recommends that the cooling system be checked immediately, which ultimately leads to timely repairs before the fault occurs, avoiding serious equipment damage and production stagnation.

[0175] According to statistics, in the first three months after the system was applied, equipment downtime was reduced by about 70%, and equipment maintenance costs were reduced by 30%. The company's production efficiency increased by 15%, especially in the maintenance of key equipment. Through the fault prediction and maintenance suggestions generated by the system, the maintenance response time of the equipment was shortened from 24 hours to within 8 hours, greatly improving the operating efficiency of the production line.

[0176] Table 1 Data table

[0177]

[0178] This table shows the data comparison of multiple key indicators such as equipment failure rate, prediction accuracy, maintenance response time and maintenance cost after the introduction of the digital twin-based IETM system, proving the significant effect of the system in practical applications.

[0179] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A complex equipment IETM method based on digital twins, characterized in that: The steps include: S1. Collect multi-source heterogeneous data of complex equipment, including sensor data, operation logs and maintenance records, and divide the data into different time scales and spatial levels according to the equipment operation characteristics for pre-processing; S2. Based on the preprocessed data, a multi-scale spatiotemporal coupled digital twin model is constructed to capture the dynamic changes of the equipment in different time and space dimensions and realize real-time modeling of the equipment status; S3, using non-Euclidean space mapping, converting multi-dimensional equipment status data into a topological structure, establishing the dynamic evolution trajectory of the equipment, and reflecting the nonlinear changes of the equipment under complex working conditions; S4. By real-time monitoring of device status, calculating the change value of information entropy, and using the information entropy feedback mechanism to dynamically adjust the key parameters in the twin model; S5. Optimize the equipment operation status according to the adjusted model parameters and generate real-time fault prediction and maintenance plans; S6. Display equipment status, fault prediction and maintenance suggestions to users through the IETM system to provide equipment management and decision support; S7. During the entire life cycle of the equipment, the system continuously optimizes the twin model through the synergy of multi-scale spatiotemporal coupling and information entropy feedback.

2. According to claim 1, a complex equipment IETM method based on digital twins is characterized in that: The S2 comprises the following steps: S21, based on the multi-source heterogeneous data preprocessed in S1, including sensor data, operation logs and maintenance records, generates a time-space state matrix M0 (t, x1, x2, ..., x n ), where t is the time dimension, x1,x2,...,x n Represents the state variables of the device in the spatial dimension; S22. Perform adaptive variational wavelet transform on the time series data D(t), decompose the time dynamics of the equipment at different scales, and extract the multi-time scale characteristics of the equipment operation. The formula is as follows: Among them, C i ( t) is the variation coefficient of the i-th time scale, u i ( t) represents the displacement function of the device time dynamics, s i (t) is the scale factor, ψ is the adaptive wavelet basis function; S23, for spatial dimension data x1, x2, ..., x n Perform multi-scale dimensionality reduction, extract the key spatial features of the device through adaptive nonlinear dimensionality reduction, and generate the device spatial feature matrix S LLE (x1,x2,...,x n ); S24, the time feature matrix T s ( t) and the spatial feature matrix S LLE (x1,x2,...,x n ) to conduct spatiotemporal coupling, build a dynamic state model of the device through multi-scale spatiotemporal tensors, and generate a spatiotemporal coupling tensor M st (t,x1,x2,...,x n ), the formula is as follows: in, Represents element-wise multiplication; S25, by introducing a dynamic weight optimization mechanism, the spatiotemporal coupling tensor M st (t,x1,x2,...,x n ) Adaptively weight and dynamically adjust the importance of temporal and spatial features; S26, through the optimized space-time coupling tensor M st (t,x1,x2,...,x n ), generate the device state prediction matrix P(t,x1,x2,...,x n ), the prediction formula is as follows: Among them, λ is the regularization parameter, which is used to control the smoothness of the model.

3. The IETM method for complex equipment based on digital twin according to claim 1, characterized in that: The S3 comprises the following steps: S31, extracting multi-dimensional state data M(t,x) of the device, where t represents the time dimension, x=(x1,x2,...,x n ) represents the state variables of the device in different spatial dimensions, forming a state matrix M0(t,x); S32. Map the device status data based on the Riemann manifold in non-Euclidean geometry. The mapping kernel function is defined as: Among them, M(t,x) and M(t,x') represent the states of the device at the spatial points x and x', respectively. Represented on a Riemann manifold The metric distance on , σ is the scale parameter; S33. Metric tensor g based on Riemannian manifold ij (x), calculate the change of device status in the spatial dimension, the formula is as follows: Among them, g ij (x) is the Riemann metric tensor, which represents the geometric change of the device in the spatial dimension, M k (t,x) is the state change of the device in the kth dimension, x i and x j are different dimensions of spatial coordinates; S34, construct the dynamic evolution trajectory of the device, and describe the evolution trajectory of the device through the geodesic on the Riemann manifold; S35. Dynamic topological characteristics of computing devices, using Betti number β k (t) to capture the topological changes of devices in different dimensions; S36. Construct a nonlinear state change model of the equipment N e (t), based on the topological structure and dynamic evolution trajectory of the device; S37, through the feedback mechanism, the nonlinear state change model N e (t) is fed back to the digital twin model to dynamically adjust the equipment state parameter θ(t).

4. The IETM method for complex equipment based on digital twin according to claim 1, characterized in that: The S4 comprises the following steps: S41, multi-dimensional state matrix M of slave device ( t,x), where t is the time dimension and x = (x1, x2, ..., x n ) is the state of the device in the spatial dimension; S42, perform probability density estimation on the device state data M(t,x), and use the kernel density estimation KDE method to calculate the probability density function P(M ( t,x)), the formula is as follows: Among them, K h is the kernel function of bandwidth h, representing the smooth distribution of state data, M i (t,x) represents the state of the i-th data point; S43. Calculate the conditional information entropy H(t|x) of the device state. The formula is as follows: Among them, P(M(t,x)) represents the probability distribution of the device in the spatiotemporal state, and information entropy is used to quantify the complexity of the device state; S44. Based on the change of the information entropy of the device state, define the information entropy change rate ΔH(t) as: S45, feeding back the information entropy change rate ΔH(t) to the digital twin model, and adjusting the key parameter θ(t) of the twin model; S46. Based on the information entropy feedback results, iteratively update the twin model of the device so that the model can adapt to the dynamic changes of the device status.

5. The IETM method for complex equipment based on digital twin according to claim 1, characterized in that: The S5 comprises the following steps: S51. Generate the device optimization state matrix M based on the adjusted twin model parameters θ(t) opt (t,x); S52, using the device state matrix M opt (t,x), build a fault risk model based on state offset, the formula is as follows: Among them, λ is the failure risk control coefficient, M safe (x) is the state distribution of the device under the ideal safety state; S53. Calculate the predicted time T of equipment failure according to the failure risk model fault , the formula is: T fault =inf{t∣P fault (t,x(≥P threshold }; Among them, P threshold The preset fault trigger threshold is used to predict the possible failure of the equipment at a certain point in time; S54, Optimize maintenance strategy S maint The generation formula of (t) is as follows: Among them, C maint (t') is the maintenance cost at time t', γ is the weight factor; S55, dynamically update the equipment status matrix M based on the real-time monitoring data opt ( t,x ) and maintenance strategy S maint ( t); S56. During the entire life cycle of the equipment, the key parameters θ(t) in the twin model are adjusted in real time by combining the state optimization and fault prediction models; S57. Based on the optimized twin model and equipment status matrix, the dynamic operation of the equipment is continuously monitored, and the long-term stable operation and maintenance costs of the equipment are minimized through iterative updates of maintenance strategies.

6. The IETM method for complex equipment based on digital twin according to claim 1, characterized in that: The S6 comprises the following steps: S61. Obtain the optimized state matrix and fault prediction data of the equipment from the twin model, and display the equipment status and real-time risk assessment through the multi-dimensional visualization interface of the IETM system; S62. Generate an overall risk assessment of the equipment based on the status of each component of the equipment and the fault prediction results, and display it in the user interface of the IETM system through risk mapping; S63. Dynamically generate maintenance priorities for equipment, adjust the priorities of various parts of the equipment based on risk assessment results and maintenance opportunities, and display them in the system; S64. Through the real-time data interface of the IETM system, users can view equipment status, maintenance recommendations and fault predictions. All information is displayed through interactive graphics and text descriptions, and users can operate based on these data; S65. Based on real-time monitoring of equipment status, dynamically adjust equipment management strategies and optimize equipment management decisions through risk assessment results; S66, dynamically updating the equipment maintenance suggestions and fault prediction information as the equipment status changes, so that the maintenance plan can be adjusted in real time, and the user can confirm and modify the maintenance suggestions through the interface; S67. Provide real-time equipment monitoring and fault prediction through the synergy between the IETM system and the digital twin model.

7. A complex equipment IETM system based on digital twins, characterized in that: The system comprises: Data acquisition module, used to collect multi-source heterogeneous data of complex equipment, including sensor data, operation logs, maintenance records, etc., which provide input data for the spatiotemporal coupling modeling of equipment; The time-space coupling processing module processes the collected data through a multi-scale time-space coupling mechanism, performs synchronous modeling at different time scales and spatial levels according to the operating status of the equipment, and captures the dynamic state changes of the equipment; The digital twin model module uses the multi-scale data generated by the spatiotemporal coupling processing module to build a digital twin model of the equipment, integrating the microscopic state and macroscopic operation trend of the equipment in real time to provide accurate equipment state representation; Non-Euclidean mapping module: maps the multi-dimensional state of the equipment to the non-Euclidean space through nonlinear topology, generates a dynamic evolution trajectory model of the equipment under complex working conditions, describes the nonlinear changes of the equipment under extreme conditions, and provides data support for optimizing the equipment state; The information entropy feedback module monitors the changes in information entropy during equipment operation, provides real-time feedback on the complexity and uncertainty of equipment status, adjusts key parameters of the digital twin model, and achieves adaptive optimization of equipment status and fault prediction; The IETM system module is responsible for integrating the results of the digital twin model, non-Euclidean mapping module and information entropy feedback module, presenting the equipment's status changes, fault predictions and maintenance optimization plans to users through an interactive interface, and providing equipment management and decision support tools.

Images