A crane whole life cycle evaluation method based on digital twinning

Abstract

The application discloses a kind of crane full life cycle evaluation methods based on digital twinning, it is related to crane technical field, the method includes: determining primary loss structure component set and secondary loss structure component set;Initial digital twin structure model is constructed;Initial digital twin structure model is time series updated, and digital twin structure monitoring model sequence is obtained;Digital twin structure monitoring model sequence is enhanced in multi-receptive field space-time state interaction, and target crane digital twin space-time state feature is determined;Life assessment is carried out based on target crane digital twin space-time state feature, and full life cycle stage and target crane life assessment result are determined.The application solves the technical problems that the depth of crane operation data analysis is not enough in the prior art, leading to full life cycle evaluation not comprehensive, there is a lag.The technical effects of multi-level component interaction analysis on crane are achieved, and the accuracy of full life cycle evaluation is improved.

Description

Technical Field

[0001] This invention relates to the field of crane technology, and more specifically to a method for assessing the entire lifecycle of a crane based on digital twins. Background Technology

[0002] Currently, crane lifecycle assessments primarily rely on traditional experience-based inspections and scheduled maintenance plans, which typically cannot obtain real-time status information for crane components. While some intelligent monitoring systems exist, most focus only on monitoring specific components and fail to fully consider the overall performance and lifecycle evolution of the crane structure, especially its comprehensive performance under dynamic loads, environmental influences, and various operating conditions. How to comprehensively monitor and accurately assess the lifespan and status of crane components, thereby optimizing maintenance plans and reducing unexpected failures, has become an important research direction for intelligent crane management.

[0003] Current technologies for crane life assessment primarily rely on experience-based wear prediction models, which cannot accurately capture the multi-level and multi-component collaborative evolution effects, especially the changing patterns under complex loads and environmental factors. Furthermore, traditional assessment methods often depend on manual inspections and periodic data collection, which frequently fail to reflect the crane's dynamic performance and real-time status during operation accurately and in real time. This results in a certain lag in fault prediction and equipment maintenance.

[0004] Existing technologies suffer from insufficient depth in analyzing crane operation data, resulting in incomplete and lagging full life cycle assessments. Summary of the Invention

[0005] This application provides a method for assessing the entire life cycle of a crane based on digital twins, which addresses the technical problem that insufficient depth of crane operation data analysis in existing technologies leads to incomplete and lagging life cycle assessments.

[0006] In view of the above problems, this application provides a method for assessing the entire life cycle of a crane based on digital twins, the method comprising:

[0007] Obtain the structural parameters, structural component attributes, and operating condition boundaries of the target crane. Based on the structural parameters, structural component attributes, and operating condition boundaries of the target crane, perform historical full life cycle loss data mining. Based on the historical full life cycle loss data set obtained from the mining, identify the primary loss structural components and secondary loss structural components in the set of structural components, and determine the set of primary loss structural components and the set of secondary loss structural components.

[0008] The set of first-level lossy structural components is used as the first-level digital twin node set, the set of second-level lossy structural components is used as the second-level digital twin node set, and the physical relationship between each structural component in the first-level lossy structural component set and the second-level lossy structural component set is represented as an edge to construct the initial digital twin structural model.

[0009] The initial digital twin structure model is deployed on a digital twin platform. The digital twin platform receives operational monitoring data collected from the set of primary loss structure components and the set of secondary loss structure components at a preset monitoring frequency, and updates the initial digital twin structure model in a time sequence based on the operational monitoring data to obtain a digital twin structure monitoring model sequence.

[0010] The digital twin structure monitoring model sequence is enhanced with multi-sensory field spatiotemporal state interaction to determine the spatiotemporal state characteristics of the target crane's digital twin;

[0011] Life assessment is performed based on the spatiotemporal state characteristics of the target crane's digital twin to determine the full life cycle stage and the target crane's life assessment results.

[0012] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0013] This application obtains the structural parameters, structural component attributes, and operating condition boundaries of the target crane. Based on these parameters, it performs historical lifecycle loss data mining. Using the mined historical lifecycle loss data set, it identifies primary and secondary loss structural components within the structural component set, thus determining the primary and secondary loss structural component sets. The primary loss structural component set is used as the primary digital twin node set, and the secondary loss structural component set is used as the secondary digital twin node set. The application also establishes the physical relationships between the structural components within each of the primary and secondary loss structural component sets. Relationships are represented as edges to construct an initial digital twin structural model. This initial digital twin structural model is deployed on a digital twin platform. The platform receives operational monitoring data collected from the primary and secondary lossy structural component sets at a preset monitoring frequency and updates the initial digital twin structural model temporally based on this data, obtaining a digital twin structural monitoring model sequence. Multi-sensory field spatiotemporal state interaction enhancement is applied to the digital twin structural monitoring model sequence to determine the spatiotemporal state characteristics of the target crane's digital twin. Based on these characteristics, a lifespan assessment is performed to determine the entire lifespan stage and the target crane's lifespan assessment results. This achieves the technical effect of performing multi-level component interaction analysis on the crane, improving the accuracy of the entire lifespan assessment. Attached Figure Description

[0014] Appendix Figure 1 This is a schematic diagram of a crane lifecycle assessment method based on digital twins provided in an embodiment of the present invention.

[0015] Appendix Figure 2 This is a schematic diagram of the process for determining the set of primary loss structural components and the set of secondary loss structural components in a crane life cycle assessment method based on digital twins provided in an embodiment of the present invention.

[0016] Appendix Figure 3 This is a flowchart illustrating the process of determining the spatiotemporal state characteristics of a target crane's digital twin in a crane lifecycle assessment method based on digital twins, provided by an embodiment of the present invention. Detailed Implementation

[0017] The present invention will be further illustrated below with reference to specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. Furthermore, it should be understood that after reading the teachings of this invention, those skilled in the art can make various alterations or modifications to the invention, and these equivalent forms also fall within the scope defined by the appended claims.

[0018] It should be noted that the terms "comprising" and "having" are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to these processes, methods, products, or devices.

[0019] Examples, as shown in the appendix Figure 1 As shown, this application provides a method for assessing the entire lifecycle of a crane based on digital twins, wherein the method includes:

[0020] Step S1000: Obtain the structural parameters, structural component attributes, and working condition boundaries of the target crane; perform historical full life cycle loss data mining based on the structural parameters, structural component attributes, and working condition boundaries of the target crane; identify primary loss structural components and secondary loss structural components in the set of structural components based on the historical full life cycle loss data obtained from the mining; and determine the set of primary loss structural components and the set of secondary loss structural components.

[0021] Furthermore, such as Figure 2As shown, the structural parameters, structural component attributes, and operating condition boundaries of the target crane are obtained. Based on the structural parameters, structural component attributes, and operating condition boundaries of the target crane, historical full life cycle loss data mining is performed. Based on the historical full life cycle loss data set obtained through mining, primary loss structural components and secondary loss structural components in the structural component set are identified, and the primary loss structural component set and the secondary loss structural component set are determined. Step S1000 of this embodiment further includes:

[0022] Step S1100: Extract data from the historical full life cycle loss data set based on the loss structure components, loss degree, and usage duration to obtain M loss structure components, M loss degree sets, and M loss structure component usage duration sets.

[0023] Step S1200: Perform mean tracking and filtering on the M sets of loss degree and the M sets of usage time of loss structural components respectively to determine the M loss degree filtering values ​​and the M usage time filtering values ​​of loss structural components;

[0024] Step S1300: Identify the primary loss structural components and secondary loss structural components in the set of structural components based on the M loss degree screening values ​​and the M loss structural component usage time screening values, and determine the set of primary loss structural components and the set of secondary loss structural components.

[0025] In one possible embodiment, structural parameters represent the physical characteristics of each crane component, such as size, weight, material properties, and load-bearing capacity. Structural component attributes represent the functional characteristics, service life, manufacturing materials, and durability of each crane component. Operating condition boundaries represent the range of external conditions the crane endures during operation, including load, environmental factors, and work cycles. Primary loss structural components represent those experiencing high wear throughout the crane's lifespan; these are typically critical components bearing significant loads (such as the boom and tower). Secondary loss structural components are usually those bearing smaller loads or experiencing slower wear during use (such as electrical systems and control systems).

[0026] Preferably, using structural parameters, structural component attributes, and operating condition boundaries as indexes, a data mining and retrieval process is performed in the big data to obtain the wear and tear information of cranes with the same conditions as the target crane, which is the historical full life cycle wear and tear data set. Then, using worn structural components, wear degree, and usage duration as indexes, the historical full life cycle wear and tear data set is retrieved to obtain the corresponding data in the wear and tear data, namely, M worn structural components, M sets of wear degree data, and M sets of usage duration data for worn structural components. Here, M is a positive integer.

[0027] Furthermore, the obtained data needs to be filtered to determine the most representative data from the M sets of wear degree sets and the M sets of wearable structural component usage duration sets. Preferably, the M wear degree screening values ​​and M wearable structural component usage duration screening values ​​are obtained by mean tracking and filtering the M sets of wear degree sets and the M wearable structural component usage duration sets. The wear degree screening value reflects the general wear level of each wearable structural component. The wearable structural component usage duration screening value reflects the general usage duration of each wearable structural component when wear occurs. By analyzing the M wear degree screening values ​​and the M wearable structural component usage duration screening values, structural components that experience more wear, i.e., those requiring close monitoring, are identified as primary wear structural components. Then, based on the set of structural components, the remaining structural components are identified as secondary wear structural components.

[0028] Furthermore, based on the lossy structural components, loss degree, and usage duration, data is extracted from the historical full lifecycle loss data set to obtain M lossy structural components, M loss degree sets, and M lossy structural component usage duration sets. In this embodiment, step S1100 further includes:

[0029] Step S1101: Extract data from the historical full life cycle loss data set based on the loss structure components, loss degree, and usage duration to obtain the loss structure component set, loss degree set, and usage duration set;

[0030] Step S1102: Aggregate the set of lossy structural components of the same type to obtain M lossy structural components;

[0031] Step S1103: Based on the M loss structural components, map and aggregate the loss degree set and the usage duration set to obtain the M loss degree set and the M loss structural component usage duration set.

[0032] In one embodiment, by extracting loss degree and usage duration data related to structural components from a historical loss data set, a set of lossy structural components, a set of loss degree data, and a set of usage duration data are obtained. This allows for the quantification of the loss status of each structural component, forming a comprehensive loss data set, which supports the identification of primary and secondary lossy components and subsequent evaluation. Components with similar loss characteristics, operating conditions, or structural features are grouped into the same category, and M lossy structural components are selected.

[0033] The system maps and aggregates M lossy structural components to their corresponding loss degree sets and usage duration sets. Through this mapping, the system can integrate the loss degree and usage duration information of each lossy structural component to form corresponding aggregated data. Aggregate analysis provides complete loss information for each structural component, serving as the basis for subsequent loss screening and evaluation.

[0034] Furthermore, based on the M loss degree screening values ​​and the M loss structural component usage time screening values, the primary loss structural components and secondary loss structural components in the set of structural components are identified, and the set of primary loss structural components and the set of secondary loss structural components are determined. In this embodiment, step S1300 further includes:

[0035] Step S1301: Weight and fuse the reciprocals of the M loss degree screening values ​​and the M loss structural component usage time screening values ​​respectively to determine the M loss structural component attention coefficients;

[0036] Step S1302: Add the loss structure components whose attention coefficients are greater than or equal to a preset coefficient threshold from the M loss structure component attention coefficients to the first-level loss structure component set.

[0037] Step S1303: Add the loss structure components whose attention coefficients are less than a preset coefficient threshold from the M loss structure component attention coefficients to the secondary loss structure component set.

[0038] In one embodiment, the set of wear level screening values ​​is first inversely divided, thus giving greater weight to components with less wear. Then, the usage duration screening value for each wear-prone structural component is combined with this value, and a weighted sum of these two values ​​is obtained to obtain the wear-prone structural component attention coefficient. By calculating the attention coefficient, different components are assigned different weights in the wear assessment, ensuring that important wear-prone components receive sufficient attention in the subsequent screening process.

[0039] A preset coefficient threshold is set by those skilled in the art. Components with a coefficient of concern greater than or equal to this threshold are identified as primary loss structural components and added to the primary loss structural component set. The selected primary loss structural components are removed from the component set, leaving secondary loss structural components. These secondary loss structural components typically have lower wear levels but still play a role in the overall system operation. By accurately classifying component wear levels, more precise component classification is provided for subsequent health assessments, lifespan predictions, and maintenance plans. This achieves the goal of focusing more attention on the primary loss components that have the greatest impact on the system, while ensuring that the assessment of secondary loss components is not overlooked.

[0040] Furthermore, mean tracking and filtering are performed on the M sets of loss degree and the M sets of usage time of lossy structural components respectively to determine the M loss degree filtering values ​​and the M usage time filtering values ​​of lossy structural components. In this embodiment, step S1200 further includes:

[0041] Step S1201: Extract the first loss set from the M loss set;

[0042] Step S1202: Calculate the mean of the first set of loss degrees, and use the mean as the first mean tracking screening value to construct the first mean tracking screening center neighborhood;

[0043] Step S1203: Randomly extract a first loss degree from the edge of the neighborhood of the first mean tracking screening center, use it as the first iterative tracking screening value, and construct the first iterative tracking screening center neighborhood of the first iterative tracking screening value;

[0044] Step S1204: Compare the neighborhood of the first iterative tracking screening center with the neighborhood of the first mean tracking screening center to determine the mean tracking screening direction and the first stage tracking screening value;

[0045] Step S1205: According to the mean tracking screening direction, the first stage tracking screening value is iteratively screened in the first loss degree set until the preset number of iterations is met, and the first loss degree screening value is determined.

[0046] Step S1206: By analogy, mean tracking and filtering are performed on the M sets of loss degree and the M sets of usage time of loss structural components to determine the M loss degree filtering values ​​and the M usage time filtering values ​​of loss structural components.

[0047] Furthermore, by comparing the neighborhood of the first iterative tracking screening center and the neighborhood of the first mean tracking screening center to determine the mean tracking screening direction and the first-stage tracking screening value, step S1204 of this embodiment further includes:

[0048] Step S1204-1: Calculate the neighborhood density of the first iterative tracking screening center neighborhood and the first mean tracking screening center neighborhood respectively, to obtain the first iterative tracking screening center neighborhood density and the first mean tracking screening center neighborhood density;

[0049] Step S1204-2: Determine whether the neighborhood density of the first iteration tracking screening center is greater than or equal to the neighborhood density of the first mean tracking screening center. If so, take the direction from the first mean tracking screening value to the first iteration tracking screening value as the mean tracking screening direction, and take the first iteration tracking screening value as the first stage tracking screening value.

[0050] Step S1204-3: If not, then take the direction from the first iteration mean tracking filter value to the first mean tracking filter value as the mean tracking filter direction, and take the first mean tracking filter value as the first stage tracking filter value.

[0051] Furthermore, using the first mean tracking filter value as the filter center and the preset mean tracking filter bandwidth as the filter radius, the first loss degree, which is within the range of the filter radius, is added to the neighborhood of the first mean tracking filter center.

[0052] In one embodiment, a first loss degree set is randomly selected from the M loss degree sets and used as an example. The mean of the first loss degree set is calculated, and this mean is used as a first mean tracking filter value. The first mean tracking filter value reflects the general loss characteristics of the first loss degree set. A first mean tracking filter center neighborhood is constructed based on the first mean tracking filter value. The first mean tracking filter center neighborhood reflects the loss degree characteristics clustered around the first mean tracking filter value.

[0053] Preferably, using the first mean tracking filter value as the filter center, and a preset mean tracking filter bandwidth pre-defined by those skilled in the art as the filter radius, if the distance from the first loss degree to the filter center is within the filter radius, it indicates that it is clustered near the filter center, and it is added to the neighborhood of the first mean tracking filter center. For example, if a component has a first loss degree of 9, and its distance from the first mean tracking filter value 10 is calculated to be 1, this value is within the filter radius of 2. Therefore, the loss degree of this component will be considered to be clustered near the filter center and added to the neighborhood of the filter center. As another example, if another component has a first loss degree of 7, and its distance from the filter center 10 is calculated to be 3, this distance exceeds the filter radius of 2, therefore this component will not be added to the neighborhood of the filter center.

[0054] A first loss value is randomly extracted from the edge of the neighborhood of the first mean tracking screening center and used as the first iterative tracking screening value. Based on the same principle as obtaining the neighborhood of the first mean tracking screening center, a first iterative tracking screening center neighborhood of the first iterative tracking screening value is constructed. Furthermore, by comparing the neighborhood densities of the first iterative tracking screening center neighborhood and the first mean tracking screening center neighborhood, the value that is more representative between the first iterative tracking screening value and the first mean tracking screening value is used as the first stage tracking screening value.

[0055] Preferably, the amount of data within the neighborhood of the first mean-tracking filter center is counted, and the data amount is divided by twice the filter radius to obtain the density of the neighborhood of the first iterative tracking filter center. The density of the neighborhood of the first iterative tracking filter center reflects the clustering of data around the first iterative tracking filter value. Based on the same principle, the density of the neighborhood of the first mean-tracking filter center is calculated.

[0056] Then, it is determined whether the neighborhood density of the first iteration tracking screening center is greater than or equal to the neighborhood density of the first mean tracking screening center. If so, the first iteration tracking screening value is more representative. The direction from the first mean tracking screening value to the first iteration tracking screening value is taken as the mean tracking screening direction, and the first iteration tracking screening value is taken as the first stage tracking screening value.

[0057] If not, the first mean tracking filter value is more representative. The direction from the first iteration mean tracking filter value to the first mean tracking filter value is taken as the mean tracking filter direction, and the first mean tracking filter value is taken as the first stage tracking filter value.

[0058] Following the mean-tracking filtering direction, the first-stage tracking filtering value is iteratively filtered within the first loss degree set until a preset number of iterations, such as 30 or 50, is met, thus determining the first loss degree filtering value. Based on the same principle used to obtain the first loss degree filtering value, mean-tracking filtering is performed on the M loss degree sets and the M loss structural component usage duration sets to determine the M loss degree filtering values ​​and the M loss structural component usage duration filtering values.

[0059] Step S2000: Take the set of first-level loss structure components as the set of first-level digital twin nodes, take the set of second-level loss structure components as the set of second-level digital twin nodes, and represent the physical relationship between each structure component in the set of first-level loss structure components and the set of second-level loss structure components as edges to construct the initial digital twin structure model.

[0060] In one embodiment, selected primary and secondary lossy structural components are used as nodes in a digital twin, and the physical relationships between these nodes are represented as edges, thus constructing a preliminary digital twin structural model. First, the primary and secondary lossy structural components are used as the primary and secondary node sets of the digital twin model, respectively; these nodes represent various components in the physical crane. Then, based on the physical relationships between these components (such as load-bearing, transmission, and motion), corresponding edges are created in the model to illustrate how these components interact and cooperate.

[0061] After constructing the initial digital twin structural model, the model will have a virtual copy of the crane, capable of reflecting the status, behavior, and interactions of each component in real time. This step provides an accurate virtual structural foundation for subsequent dynamic monitoring, loss assessment, and lifespan prediction, and supports the real-time updating and feedback of the digital twin model. In this way, the virtual copy of the crane can be continuously updated in sync with the physical system, enabling comprehensive lifecycle monitoring and intelligent decision support.

[0062] Step S3000: Deploy the initial digital twin structure model on the digital twin platform. The digital twin platform receives operation monitoring data collected from the set of primary loss structure components and the set of secondary loss structure components at a preset monitoring frequency, and updates the initial digital twin structure model in time according to the operation monitoring data to obtain a digital twin structure monitoring model sequence.

[0063] In one possible embodiment, the initial digital twin structural model represents a virtual model built based on the crane's structure, component attributes, and physical connections, representing the crane's complete structure and state. Operational monitoring data represents real-time data collected from various crane components (such as the boom, wire ropes, tower, etc.), typically including sensor data for temperature, stress, vibration, and load. The preset monitoring frequency represents the data acquisition frequency set on the digital twin platform, such as per minute, per hour, or per day, used to define the time interval for data updates.

[0064] The digital twin platform begins receiving real-time operational monitoring data from primary lossy structural components (such as the boom and tower) and secondary lossy structural components (such as electrical systems and sensors) at a preset monitoring frequency (e.g., data collected once per hour). This data may include various real-time sensor data such as temperature, vibration, stress, and load. Assume that at a certain moment, the digital twin platform receives stress sensor data from the boom (e.g., the boom is bearing a load of 500 kg) and vibration frequency data from the wire rope (e.g., the vibration frequency is 20 Hz). The received operational monitoring data is input into the initial digital twin structural model and updated sequentially based on the latest data. During each update, the platform calculates the state of each component at the current moment and compares it with the previous state, forming a dynamic, time-series updated model.

[0065] Suppose that at the previous moment, the stress on the boom was 490 kg and the vibration frequency of the wire rope was 19 Hz. At the next moment, data shows that the stress has changed to 500 kg and the vibration frequency has changed to 20 Hz. The digital twin platform updates the model over time, updating the state of the boom and wire rope through comparison and weighting to ensure that the model reflects the current state.

[0066] After each time-series update, the digital twin platform saves the current update results as new time-point data, forming a sequence of digital twin structural monitoring models. This sequence represents the health status and performance changes of various crane components at different points in time. Over time, the monitoring model sequence grows continuously, providing in-depth analysis of the crane's long-term health status.

[0067] Over time, the digital twin platform accumulates data at multiple points in time. For example, at time t1, the stress on the boom is 490 kg; at time t2, the stress on the boom is 500 kg; and at time t3, the stress on the boom is 510 kg. Each time the data is updated, the platform generates a new monitoring model and records it, thus forming the trajectory of the crane's state evolution.

[0068] By deploying the initial digital twin structural model to the digital twin platform and continuously updating it with collected operational monitoring data, the resulting digital twin structural monitoring model sequence provides comprehensive dynamic status feedback for the crane.

[0069] Step S4000: Perform multi-sensory-field spatiotemporal state interaction enhancement on the digital twin structure monitoring model sequence to determine the spatiotemporal state characteristics of the target crane digital twin;

[0070] Furthermore, such as Figure 3 As shown, the digital twin structure monitoring model sequence is enhanced with multi-sensory field spatiotemporal state interaction to determine the spatiotemporal state characteristics of the target crane's digital twin. Step S4000 in this embodiment includes:

[0071] Step S4100: Using the first-level digital twin nodes as indexes, perform long-term receptive field spatial state analysis on the digital twin structure monitoring model sequence to determine the spatial state feature set of the first-level digital twin nodes;

[0072] Step S4200: Using the secondary digital twin nodes as indexes, perform short-term receptive field time state analysis on the digital twin structure monitoring model sequence to determine the time state feature set of the secondary digital twin nodes;

[0073] Step S4300: Preprocess the spatial state feature set of the first-level digital twin node and the temporal state feature set of the second-level digital twin node respectively to obtain the preprocessed spatial state features of the first-level digital twin node and the preprocessed temporal state features of the second-level digital twin node.

[0074] Step S4400: Perform spatiotemporal state interaction enhancement on the preprocessed spatial state features of the first-level digital twin node and the preprocessed temporal state features of the second-level digital twin node to determine the spatiotemporal state features of the target crane digital twin.

[0075] Furthermore, spatiotemporal state interaction enhancement is performed on the preprocessed spatial state features of the first-level digital twin node and the preprocessed temporal state features of the second-level digital twin node to determine the spatiotemporal state features of the target crane digital twin. In this embodiment, step S4400 further includes:

[0076] Step S4401: Calculate the similarity between the preprocessed spatial state features of the first-level digital twin node and the preprocessed temporal state features of the second-level digital twin node, and determine the spatiotemporal state feature similarity set;

[0077] Step S4402: Normalize the spatiotemporal state feature similarity set to construct a spatiotemporal adjacency matrix;

[0078] Step S4403: Use the spatiotemporal adjacency matrix to interactively enhance the preprocessed spatial state features of the first-level digital twin node, and determine the spatiotemporal state features of the target crane digital twin.

[0079] In one embodiment, firstly, using primary digital twin nodes (such as components with heavy loads) as indices, a long-term sensing field spatial state analysis is performed on the digital twin structure monitoring model sequence to extract a set of state features related to the spatial location of these key components. Secondly, using secondary digital twin nodes (such as auxiliary components or components with lighter loads) as indices, a short-term sensing field temporal state analysis is performed on the digital twin structure monitoring model sequence to extract features related to the short-term dynamic state of these components. The short-term sensing field can help capture rapidly changing operating conditions, such as sudden load changes and short-term losses caused by frequent operations. Then, the extracted primary digital twin node spatial state feature set and secondary digital twin node temporal state feature set are subjected to mean normalization to standardize the data, ultimately obtaining the preprocessed spatial state features of the primary digital twin nodes and the preprocessed temporal state features of the secondary digital twin nodes.

[0080] For example, long-term sensing field spatial state analysis is performed on the boom (a primary digital twin node) to identify the gradual fatigue state of the material by analyzing the stress changes under load over a long period. For instance, after prolonged operation, the boom's load-bearing capacity may gradually decrease, resulting in higher load stress. Short-term sensing field temporal state analysis is performed on the electrical control system (such as a secondary digital twin node) to analyze the load fluctuations of the system over a short period. For instance, instantaneous current overloads may cause short-term losses in electrical components, which cannot be detected through long-term observation but require attention to instantaneous fluctuations.

[0081] Preferably, spatial features are extracted from the monitoring data sequence corresponding to the first-level digital twin nodes in the digital twin structure monitoring model sequence according to a long-term receptive field preset by those skilled in the art. Optionally, a model constructed with a neural network having a long-term receptive field is used for extraction to obtain the spatial state feature set of the first-level digital twin nodes. Similarly, a model constructed with a neural network having a short-term receptive field is used to extract the monitoring data sequence corresponding to the second-level digital twin nodes in the digital twin structure monitoring model sequence to obtain the temporal state feature set of the second-level digital twin nodes.

[0082] Then, the mean values ​​of the spatial state feature set of the first-level digital twin node and the temporal state feature set of the second-level digital twin node are calculated respectively to determine the state of the first-level and second-level components in the structural component set of the target crane, and to obtain the preprocessed spatial state features of the first-level digital twin node and the preprocessed temporal state features of the second-level digital twin node.

[0083] Furthermore, the spatial state features preprocessed by the first-level digital twin node and the temporal state features preprocessed by the second-level digital twin node will undergo spatiotemporal state interaction enhancement. This interaction enhancement method captures the coordinated changes of each component under different temporal and spatial conditions.

[0084] Preferably, the similarity between the preprocessed spatial state features of the first-level digital twin node and the preprocessed temporal state features of the second-level digital twin node is calculated to determine a spatiotemporal state feature similarity set. This set reflects the degree of similarity between the features of the first-level and second-level digital twin nodes. Furthermore, the spatiotemporal state feature similarity set is normalized using the softmax function, and the processed data is added to an initially empty matrix to construct a spatiotemporal adjacency matrix.

[0085] By utilizing a graph convolutional neural network model to interactively analyze the spatiotemporal adjacency matrix and the preprocessed spatial state features of the first-level digital twin nodes, the spatiotemporal state features of the target crane's digital twin are determined. The spatiotemporal adjacency matrix reflects the similarity between components of the first-level and second-level digital twin nodes. Graph data processing methods such as graph convolution enhance the dependencies between nodes and optimize the extraction of spatiotemporal state features. For example, if there is a high similarity between the boom and the electrical control system, their state features will be more closely linked, ensuring that the final output spatiotemporal state features accurately reflect the overall health of the crane.

[0086] Step S5000: Based on the spatiotemporal state characteristics of the target crane's digital twin, perform a life assessment to determine the full life cycle stage and the target crane's life assessment results.

[0087] Furthermore, a pre-constructed full life cycle evaluator is used to assess the life cycle of the target crane's digital twin spatiotemporal state characteristics, thereby determining the full life cycle stage and the target crane's life cycle assessment results.

[0088] In one possible embodiment, the lifecycle estimator includes an LSTM layer, a CNN layer, and a fully connected layer. The input data is the spatiotemporal state features of the target crane's digital twin, and the output data includes the lifecycle stages and the target crane's lifespan assessment results. The LSTM layer processes the time-series data, extracting important features along the time dimension. The CNN layer processes spatial features, extracting patterns within local regions. The fully connected layer maps the extracted features to the final output, including the lifespan assessment and the lifecycle stages.

[0089] An Adam optimizer is employed to adaptively adjust the learning rate during the learning process. The dataset includes a set of spatiotemporal state features of digital twins of sample target cranes, a set of sample full lifecycle stages, and a set of lifespan assessment results for sample target cranes. The dataset is divided into training and validation sets; typically, 80% of the data is used for training, and 20% is used to validate the model's performance. The full lifecycle estimator is trained using the Adam optimizer on both the training and validation sets. Preferably, mini-batch gradient descent is used during training, with a batch size typically set to 32 or 64. Training is stopped early if the loss on the validation set no longer decreases after several consecutive training iterations.

[0090] For example, by inputting the spatiotemporal state characteristics of the target crane's digital twin into a trained lifecycle evaluator, the model outputs a lifecycle assessment result of 2000 hours of remaining service life and a gradual decline stage in the lifecycle phase, requiring close attention to the health status of the boom and control system. By extracting the equipment's health information from the spatiotemporal state characteristics of the digital twin and predicting the equipment's remaining service life and lifecycle stage, the technical effect of reliably assessing the crane's entire lifecycle and improving the accuracy of the assessment is achieved.

[0091] In summary, the embodiments of this application have at least the following technical effects:

[0092] 1. This application identifies primary and secondary loss structural components of a structural component set based on historical full life-cycle loss data, accurately identifying key components (such as the boom and tower) that significantly impact crane performance and lifespan. This achieves the technical effect of providing high-quality data support for subsequent digital twin model construction and lifespan assessment.

[0093] 2. This application utilizes the time-series update mechanism of a digital twin platform to receive real-time operational monitoring data from primary and secondary wear components. Based on this data, the initial digital twin structural model is updated, and a lifespan assessment is performed according to the spatiotemporal state characteristics of the target crane's digital twin. This enables accurate prediction of remaining lifespan and determination of the entire lifecycle stages. It achieves the technical effect of obtaining the latest status of the target crane and providing reliable and timely response to the entire lifecycle assessment.

[0094] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0095] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0096] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A method for a crane full life cycle assessment based on digital twinning, characterized by, The method comprises: obtaining the structure parameters, structure component attributes and working condition boundaries of the target crane, performing historical full life cycle loss data mining based on the structure parameters, structure component attributes and working condition boundaries of the target crane, identifying the primary loss structure components and the secondary loss structure components in the structure component set according to the obtained historical full life cycle loss data set, and determining the primary loss structure component set and the secondary loss structure component set; taking the primary loss structure component set as a primary digital twin node set, taking the secondary loss structure component set as a secondary digital twin node set, and representing the physical association relationship between the structure components in the primary loss structure component set and the secondary loss structure component set as an edge to construct an initial digital twin structure model; deploying the initial digital twin structure model on a digital twin platform, the digital twin platform receiving operation monitoring data collected at a preset monitoring frequency from the primary loss structure component set and the secondary loss structure component set, and performing time sequence updating on the initial digital twin structure model according to the operation monitoring data to obtain a digital twin structure monitoring model sequence; performing multi-receptive field spatiotemporal state interaction enhancement on the digital twin structure monitoring model sequence to determine the target crane digital twin spatiotemporal state feature; performing life assessment based on the target crane digital twin spatiotemporal state feature to determine the full life cycle stage and the target crane life assessment result; performing multi-receptive field spatiotemporal state interaction enhancement on the digital twin structure monitoring model sequence to determine the target crane digital twin spatiotemporal state feature, comprising: taking the primary digital twin node as an index, performing long-term receptive field spatial state analysis on the digital twin structure monitoring model sequence to determine a primary digital twin node spatial state feature set; taking the secondary digital twin node as an index, performing short-term receptive field temporal state analysis on the digital twin structure monitoring model sequence to determine a secondary digital twin node temporal state feature set; respectively preprocessing the primary digital twin node spatial state feature set and the secondary digital twin node temporal state feature set to obtain a primary digital twin node preprocessed spatial state feature and a secondary digital twin node preprocessed temporal state feature; performing spatiotemporal state interaction enhancement on the primary digital twin node preprocessed spatial state feature and the secondary digital twin node preprocessed temporal state feature to determine the target crane digital twin spatiotemporal state feature.

2. The crane life cycle assessment method based on digital twinning of claim 1, wherein, obtaining the structure parameters, structure component attributes and working condition boundaries of the target crane, performing historical full life cycle loss data mining based on the structure parameters, structure component attributes and working condition boundaries of the target crane, identifying the primary loss structure components and the secondary loss structure components in the structure component set according to the obtained historical full life cycle loss data set, and determining the primary loss structure component set and the secondary loss structure component set, comprising: performing data extraction on the historical full life cycle loss data set based on the loss structure components, loss degrees and use time lengths to obtain M loss structure components, M loss degree sets and M loss structure component use time length sets; respectively, the M loss degree sets and the M loss structure component use time sets are subjected to mean tracking screening to determine M loss degree screening values and M loss structure component use time screening values; According to the M loss degree screening values and the M loss structure component use time screening values, the primary loss structure components and the secondary loss structure components in the structure component set are identified, and a primary loss structure component set and a secondary loss structure component set are determined.

3. The crane life cycle assessment method based on digital twinning of claim 2, wherein, Based on the loss structure components, the loss degrees and the use time, the historical full life cycle loss data set is subjected to data extraction to obtain M loss structure components, M loss degree sets and M loss structure component use time sets, including: Based on the loss structure components, the loss degrees and the use time, the historical full life cycle loss data set is subjected to data extraction to obtain a loss structure component set, a loss degree set and a use time set; The loss structure component set is subjected to same type aggregation to obtain M loss structure components; Based on the M loss structure components, the loss degree set and the use time set are subjected to mapping aggregation to obtain the M loss degree set and the M loss structure component use time set.

4. The crane life cycle assessment method based on digital twinning of claim 2, wherein, According to the M loss degree screening values and the M loss structure component use time screening values, the primary loss structure components and the secondary loss structure components in the structure component set are identified, and a primary loss structure component set and a secondary loss structure component set are determined, including: The reciprocal of the M loss degree screening values and the M loss structure component use time screening values are respectively subjected to weighted fusion to determine M loss structure component attention coefficients; The loss structure components corresponding to the loss structure component attention coefficients in the M loss structure component attention coefficients that are greater than or equal to a preset coefficient threshold are added into the primary loss structure component set; The loss structure components corresponding to the loss structure component attention coefficients in the M loss structure component attention coefficients that are less than the preset coefficient threshold are added into the secondary loss structure component set.

5. The crane life cycle assessment method based on digital twinning of claim 2, wherein, respectively, the M loss degree sets and the M loss structure component use time sets are subjected to mean tracking screening to determine M loss degree screening values and M loss structure component use time screening values, including: A first loss degree set is extracted from the M loss degree sets; The mean value of the first loss degree set is calculated, and the mean value is taken as a first mean tracking screening value and a first mean tracking screening center neighborhood is constructed; A first loss degree is extracted from the edge of the first mean tracking screening center neighborhood at random as a first iterative tracking screening value, and a first iterative tracking screening center neighborhood of the first iterative tracking screening value is constructed; The first iterative tracking screening center neighborhood and the first mean tracking screening center neighborhood are compared to determine a mean tracking screening direction and a first stage tracking screening value; According to the mean tracking screening direction, the first stage tracking screening value is subjected to iterative screening in the first loss degree set until a preset iterative screening number is met, and a first loss degree screening value is determined; Similarly, the M loss degree sets and the M loss structure component use time length sets are subjected to mean tracking screening to determine M loss degree screening values and M loss structure component use time length screening values.

6. The crane life cycle assessment method based on digital twinning of claim 5, wherein, The first iterative tracking screening center neighborhood and the first mean tracking screening center neighborhood are compared to determine a mean tracking screening direction and a first-stage tracking screening value, including: The neighborhood densities of the first iterative tracking screening center neighborhood and the first mean tracking screening center neighborhood are calculated respectively to obtain a first iterative tracking screening center neighborhood density and a first mean tracking screening center neighborhood density; It is determined whether the first iterative tracking screening center neighborhood density is greater than or equal to the first mean tracking screening center neighborhood density, and if so, the direction from the first mean tracking screening value to the first iterative tracking screening value is taken as the mean tracking screening direction, and the first iterative tracking screening value is taken as the first-stage tracking screening value; If not, the direction from the first iterative mean tracking screening value to the first mean tracking screening value is taken as the mean tracking screening direction, and the first mean tracking screening value is taken as the first-stage tracking screening value.

7. The crane life cycle assessment method based on digital twinning of claim 6, wherein, The first loss degree within the screening radius from the first mean tracking screening value is added to the first mean tracking screening center neighborhood.

8. The crane life cycle assessment method based on digital twinning of claim 1, wherein, The spatiotemporal state interaction enhancement of the first-level digital twin node pretreatment spatial state feature and the second-level digital twin node pretreatment time state feature is performed to determine the target crane digital twin spatiotemporal state feature, including: The similarity of the first-level digital twin node pretreatment spatial state feature and the second-level digital twin node pretreatment time state feature is calculated to determine a spatiotemporal state feature similarity set; The spatiotemporal state feature similarity set is subjected to normalization processing to construct a spatiotemporal adjacency matrix; The first-level digital twin node pretreatment spatial state feature is interactively enhanced using the spatiotemporal adjacency matrix to determine the target crane digital twin spatiotemporal state feature.

9. The crane life cycle assessment method based on digital twinning of claim 1, wherein, A full life cycle evaluator is pre-constructed, and the target crane digital twin spatiotemporal state feature is subjected to life assessment using the full life cycle evaluator to determine a full life cycle phase and a target crane life assessment result.

Images