A tower crane tower structure intelligent life evaluation and management method

Abstract

This invention belongs to the cutting-edge interdisciplinary field of structural safety assessment and intelligent operation and maintenance management of engineering machinery. It provides a method for intelligent life assessment and management of tower crane tower structures, including: establishing unique identifiers for each standard section of the tower; identifying operating parameters for each work cycle; establishing a finite element model of the tower crane tower; training a proxy model; inputting actual operating parameters into the trained proxy model to obtain the stress cycle spectrum of each standard section; calculating the fatigue damage value and remaining life of the main structure of each standard section; establishing weld sub-models for dangerous weld areas in the standard sections; establishing a standard section life cycle database; and performing reassembly judgment and installation management based on the standard section life cycle database and the stress conditions of the sections. Through the collaborative assessment of the main structure and dangerous weld areas, the life consumption characteristics of the overall load-bearing structure and local weld weakness areas can be reflected across scales, improving the completeness of fatigue life assessment results.

Description

Technical Field

[0001] This invention belongs to the cutting-edge interdisciplinary field of engineering machinery structural safety assessment and intelligent operation and maintenance management, and in particular relates to a method for intelligent life assessment and management of tower crane tower structure. Background Technology

[0002] Tower cranes are widely used in high-rise buildings, bridge engineering, and large-scale infrastructure construction, serving as key lifting equipment in the engineering equipment field. A tower crane typically consists of a tower body, jib, counterweight jib, slewing mechanism, and hoisting mechanism. The tower body structure, as the main load-bearing component, endures complex cyclic loads caused by hoisting, luffing, slewing, and wind loads during service. Under long-term alternating loads, the main tower structure and its connecting welds are at risk of fatigue failure. Therefore, reasonably assessing the fatigue life of the tower crane tower structure under actual operating conditions and effectively monitoring and managing its safety status during service are crucial for ensuring safe operation, extending equipment lifespan, and reducing accident risks. This is of great significance in preventing loss of life and property due to structural fatigue failure.

[0003] Currently, the design, manufacture, and operation of tower cranes are based on existing standards such as GB / T 13752-2017 "Design Code for Tower Cranes," primarily focusing on structural strength for verification. However, insufficient attention is paid to the cumulative fatigue damage under long-term alternating loads, making it difficult to accurately reflect the true lifespan of the structure under complex working conditions. Existing tower crane life assessment technologies typically conduct fatigue analysis based on the entire structure, a section of the tower body, or major load-bearing components, such as Chinese patents CN104568487A and CN115809577A, lacking detailed modeling and life assessment of weld seams. However, in actual engineering structures, weld areas often become the weakest points where fatigue cracks are most likely to initiate due to geometric discontinuities, material microstructure variations, and residual welding stress. Engineering practice and related research both indicate that fatigue failure of tower crane tower structures often first occurs at welded joints. Therefore, overall life assessment methods that ignore the local structural characteristics of weld seams cannot accurately reflect the true fatigue failure risk of the tower crane structure. Furthermore, since the tower body is composed of multiple standard sections assembled vertically, the stress levels experienced by different sections during the same working cycle vary, resulting in different degrees of fatigue damage between sections. In practical engineering applications, the same standard section may undergo multiple disassembly and repositioning within a single project, or be transferred between different projects and reused. Currently, fatigue analysis is also conducted based on standard sections. For example, the method proposed in Chinese patent CN119849181A considers individual differences in standard sections, but it estimates lifespan by combining single stress calculation results with a stress-life lookup table. Essentially, it is an empirical assessment method based on equivalent stress, failing to consider the stress time history and cyclic characteristics formed under complex working conditions, and also failing to analyze multi-level loads using fatigue damage accumulation theory. Therefore, it is difficult to accurately reflect the fatigue damage evolution of the tower crane structure during actual service.

[0004] Therefore, in order to meet the needs of refined life assessment and segmented life management of tower crane structures during long-term service, it is urgent to establish an intelligent life assessment method that can combine operational condition monitoring data to conduct collaborative fatigue damage analysis of the tower structure and key welds, and to carry out individualized life tracking and management of standard tower sections, so as to better ensure the safety and reliability of tower crane operation. Summary of the Invention

[0005] This invention addresses the limitations of existing technologies that fail to adequately consider the high failure risk caused by weld fatigue and cannot achieve real-time differentiated fatigue damage assessment and life tracking based on the individual service history of each standard section. It proposes an intelligent life assessment and management method for tower crane tower structures, thereby preventing the structural safety hazards caused by the reinstallation of highly damaged standard sections in high-stress areas. This invention achieves collaborative assessment of the fatigue damage status of the main tower structure and key welded parts through the system integration of tower crane operating condition monitoring data, structural response prediction models, weld fatigue analysis methods, and a full life cycle information management mechanism for standard section components. This forms a closed-loop technical system from operating condition input, structural response calculation, fatigue damage assessment to component-level life status determination and installation control, providing technical support for the safe and reliable operation of tower cranes during long-term service.

[0006] To achieve the above objectives, this invention provides a method for intelligent life assessment and management of tower crane tower structures, including: S1. Collect tower crane operating condition data and establish a unique identifier for each standard section of the tower crane. The unique identifier can be implemented using a QR code, RFID electronic tag, or other traceable identification method. The basic information associated with this unique identifier includes at least: standard section model, historical operating conditions, cumulative fatigue damage value, and predicted remaining life value. Operating condition data includes lifting capacity. Work range Rotation angle Wind speed Lifting height ; S2, based on the obtained operating condition data, preprocesses and identifies the operating condition parameters for each working cycle; S3. Establish a finite element model of the tower crane tower body, and solve the stress response of each standard section under the corresponding working conditions based on the identified working condition parameters to construct a training sample dataset. S4. Train the proxy model based on the constructed training sample dataset to establish a fast mapping relationship between operating condition parameters and stress response at key locations of standard sections; S5. Input the actual operating condition parameters into the trained surrogate model to obtain the stress cycle spectrum of each standard section; S6, based on the stress cycle spectrum and combined with the fatigue damage accumulation theory, calculates the fatigue damage value and remaining life of the main structure of each standard section; S7, Establish weld sub-models for dangerous weld locations in standard sections and conduct weld life assessments; S8. The fatigue damage information of each standard section structure and weld is bound to a unique identifier to establish a standard section life cycle database. S9 performs reassembly determination and installation management based on the standard section life cycle database and the stress conditions of the section.

[0007] Furthermore, in step S2, the operating condition data is recorded according to a unified time base to form the original operating condition dataset. ,in, for The original runtime data set at each moment; for Lifting capacity at any given moment; for The range of work at any given moment; for The angle of rotation at any given moment; for The working wind speed at all times; for The lift height at any given moment; After preprocessing the initial operating condition dataset, the preprocessed line operating condition dataset is obtained. ; For preprocessing Lifting capacity at any given moment; For preprocessing The range of work at any given moment; For preprocessing The angle of rotation at any given moment; For preprocessing The working wind speed at all times; For preprocessing The lift height at any given moment; Preprocessing includes time synchronization, outlier removal, missing value repair, noise filtering, unit unification, and normalization. The work cycle refers to the time period during which a tower crane completes one typical lifting operation. During the time period Internal identification of operating parameters Operating parameters Including lifting capacity Work range Rotation angle Operating wind speed Lifting height .

[0008] Furthermore, the specific steps of step S3 are as follows: Based on the tower crane's tower body structural drawings, material parameters, geometric dimensions, and connection relationships, a finite element model of the tower crane's tower body was established. The identified operating parameters As the boundary conditions and load inputs of the finite element model, the solution yields the first... Stress response of each standard section under corresponding working conditions ; A training sample dataset is constructed by solving multiple sets of working conditions. .

[0009] Furthermore, in step S4, the proxy model adopts a deep neural network structure that introduces an attention mechanism, including an input layer, a one-dimensional convolutional feature extraction layer, an attention enhancement layer, and a fully connected output layer. A residual connection structure is provided between the one-dimensional convolutional feature extraction layer and the attention enhancement layer. The residual connection structure is used to ensure that the network maintains stable convergence performance when the number of layers is increased. The input layer is used to input the training sample dataset; The one-dimensional convolutional feature extraction layer extracts features through a one-dimensional convolutional network and outputs a high-dimensional feature map. The attention enhancement layer includes a channel attention submodule and a spatial attention submodule. The channel attention submodule adaptively weights the feature channels to enhance the combination of key operating parameters; the spatial attention submodule weights the feature spatial location to highlight potential stress concentration areas. A one-dimensional convolutional feature extraction layer extracts features from the input working condition parameters using one-dimensional convolution operations to obtain a high-dimensional feature map, thereby uncovering local combination relationships between working condition parameters and acquiring the high-dimensional feature map. The channel attention submodule generates channel weight vectors by performing global average pooling and fully connected mapping on the high-dimensional feature map, and then performs channel-wise multiplication with the high-dimensional feature map to obtain a channel-weighted feature map, enhancing the responsiveness to key working condition parameter combinations. The spatial attention submodule constructs a spatial weight mapping by compressing the high-dimensional feature map's dimensions, and then performs element-wise weighting with the high-dimensional feature map to obtain a spatially weighted feature map, highlighting the feature representation of potential stress concentration areas. The channel-weighted feature map and the spatially weighted feature map are fused to obtain the enhanced feature map; the fusion method is either element-wise weighted summation or element-wise multiplication fusion. The purpose of introducing the attention mechanism is to enhance the ability to represent the differences in the contribution of different working condition parameter combinations to the structural stress response, thereby improving the prediction accuracy of the surrogate model for complex nonlinear load-stress mapping relationships. The enhanced feature map outputs the stress response at key locations of the corresponding standard section through a fully connected output layer.

[0010] Furthermore, in step S5, the operating parameters are arranged in chronological order. Input the trained proxy model to obtain the first Stress response of a standard section ; Rainflow counting was performed on the stress time history to obtain the first... Stress cycle spectrum of a standard section Its expression is: In the formula, Indicates the first Section 1 One stress amplitude range; Indicates the first The standard section in The number of cycles corresponding to each stress amplitude range Number of groups This is the group number.

[0011] Furthermore, in step S6, the fatigue damage value of each standard section's main structure is calculated using the Miner linear cumulative damage method, and the calculation formula is as follows: In the formula, For the first Fatigue damage values ​​of a standard section; For the first The standard section in The number of cycles corresponding to each stress amplitude range; The allowable number of cycles is determined by the material fatigue performance curve. No. The remaining life of each standard section for: In the formula, For the first The damage threshold for the main structure of each standard section is typically set to 1.

[0012] Furthermore, in step S7, based on the finite element model established in step S3, finite element analysis is performed on the overall structure to extract the dangerous points and dangerous areas of the weld dangerous parts. Using the displacement field of the finite element model as the boundary condition, a weld sub-model is established in the dangerous area, and the weld entity in it is finely modeled. Based on the service history data, stress cycle statistics and fatigue damage accumulation are performed on the weld dangerous parts. Among them, the Total fatigue damage of welds in each standard section for: In the formula, For the first In the standard section, the weld seam is in the first... The number of cycles corresponding to each stress amplitude range Number of groups The group number; For the first In the standard section, the weld seam is in the first... The allowable number of cycles for each stress amplitude range is determined according to the material fatigue performance curve; the welds include welds between the main chord and the connecting parts, welds connecting the web members, welds on the end plates, and other welded connection parts that are prone to fatigue crack initiation; The weld life analysis method selects the nominal stress method, hot spot stress method, or equivalent structural stress method according to the weld structure and stress characteristics to achieve adaptive analysis of fatigue damage for different welding connection forms.

[0013] Furthermore, in step S8, the standard section lifecycle database includes at least the following information: unique identifier, standard section model, installation history, dismantling history, current and historical section information, fatigue damage of the main structure, fatigue damage of the weld, remaining life of the main structure, remaining life of the weld, and load-bearing level history.

[0014] Furthermore, the section information is used to classify each installation section into high stress level, medium stress level and low stress level according to the stress level of different sections in the whole tower. The section closer to the bottom of the tower has a higher stress level and the section closer to the top of the tower has a lower stress level. The specific classification scheme should be determined according to the tower crane model, design life and enterprise management standards.

[0015] Furthermore, the specific steps of step S9 are as follows: When a standard section needs to be reinstalled after disassembly, its damage status and lifespan status information are retrieved from the standard section lifecycle database using its unique identifier. If the fatigue damage to the main structure and welds of the standard section are both below their respective preset thresholds, and the remaining lifespan of the main structure and welds meets the usage requirements, the standard section is allowed to be installed in the corresponding load-bearing capacity section. If the fatigue damage to the main structure is close to its preset threshold but still meets safety requirements, it is only allowed to be installed in a lower load-bearing capacity section. If the fatigue damage to the main structure or welds of the standard section exceeds its preset threshold, corresponding structural safety control measures are triggered, including re-inspection, repair, or cessation of use. The preset thresholds are set according to the tower crane model, design life, enterprise management standards, and safety regulations.

[0016] The present invention has the following beneficial effects: (1) The present invention takes the standard section of the tower body as an independent life management object, establishes a unique identity for each standard section, and records its installation history, working condition history, fatigue damage status and remaining life information in association, thereby realizing differentiated life assessment and management for individual standard sections, overcoming the life distortion problem caused by treating multiple standard sections as equivalent components in the prior art.

[0017] (2) Based on actual tower crane operating data, this invention, combined with finite element analysis and a proxy model, establishes a rapid mapping relationship between operating conditions and stress response at key locations in standard sections. This allows for the rapid acquisition of structural stress time histories without the need for successive complex finite element solutions, providing an feasible technical path for dynamic life assessment during tower crane service. Through the collaborative assessment of the main structure and critical weld areas, the model can reflect the life consumption characteristics of the overall load-bearing structure and local welded weak areas across scales, thereby improving the completeness of fatigue life assessment results and facilitating a more accurate reflection of the true fatigue failure risk of the tower crane structure.

[0018] (3) This invention stores the damage results in association with the unique identifier of the standard section and the historical installation location. When the standard section is reinstalled, it is matched and controlled according to its remaining life status and the stress level of the target installation location. This helps to avoid high-damage standard sections from continuing to be installed in high-stress locations. It realizes an integrated technical system for the entire process from input of working condition data to structural response calculation, fatigue damage assessment and installation control decision-making, which effectively improves the level of refinement of tower crane structural life management and engineering safety assurance capabilities.

[0019] (4) This invention integrates tower crane operating condition monitoring data, structural response prediction model, weld fatigue analysis method and standard section component full life cycle information management mechanism to achieve collaborative assessment of fatigue damage status of the main structure of the tower crane tower body and key welded parts, and establishes a standard section life cycle tracking and management mechanism at the component level, and establishes a system-level life cycle assessment and control method based on physical model and data-driven integration, thereby realizing refined life cycle assessment and life cycle management of tower crane structure under actual service conditions, and providing technical support for the safe and reliable operation of tower crane during long-term service. Attached Figure Description

[0020] Figure 1 This invention relates to the life assessment and management process.

[0021] Figure 2 This is a statistical chart of the working condition data collected by this invention.

[0022] Figure 3 This is a diagram showing the finite element analysis results of the tower structure in this invention.

[0023] Figure 4 This is a diagram showing the finite element analysis results of the sub-model in this invention.

[0024] Figure 5 This is a cloud diagram for calculating the fatigue life of the main structure of the standard section in this invention.

[0025] Figure 6 This is a cloud diagram for calculating the fatigue life of welds in this invention. Detailed Implementation

[0026] The technical solution of the present invention will be further described in detail below with reference to specific embodiments. However, these embodiments are not intended to limit the present invention. Any similar structures and similar variations of the present invention should be included in the protection scope of the present invention. The commas in the present invention all indicate the relationship between and. The English letters in the present invention are case-sensitive.

[0027] like Figure 1 As shown, this embodiment focuses on a 2230t·m tower crane, using a tower structure composed of 9 standard sections, and conducts intelligent life assessment and management. The specific methods include: S1 collects tower crane operating condition data through the tower crane's black box and establishes a unique identifier for each standard section of the tower; each standard section Corresponding to a unique identifier This unique identifier is used for individual identification during subsequent life assessment, disassembly and reassembly management. The unique identifier can be implemented using a QR code, RFID electronic tag, or other traceable identification method. The basic information associated with this unique identifier includes at least: standard section model, historical operating conditions, cumulative fatigue damage value, and remaining life prediction value; operating condition data includes lifting capacity. Work range Rotation angle Wind speed Lifting height ; S2, based on the obtained operating condition data, preprocesses and identifies the operating condition parameters for each working cycle; Operating condition data is recorded according to a unified time base to form the original operating condition dataset. ,in, for The original runtime data set at each moment; for Lifting capacity at any given moment; for The range of work at any given moment; for The angle of rotation at any given moment; for The working wind speed at all times; for The lifting height at any given time; operational data collected during a two-month construction period at a certain construction site, such as... Figure 2 As shown.

[0028] After preprocessing the initial operating condition dataset, the preprocessed line operating condition dataset is obtained. ; For preprocessing Lifting capacity at any given moment; For preprocessing The range of work at any given moment; For preprocessing The angle of rotation at any given moment; For preprocessing The working wind speed at all times; For preprocessing The lift height at any given moment; Preprocessing includes time synchronization, outlier removal, missing value repair, noise filtering, unit unification, and normalization to eliminate the impact of time inconsistencies, abnormal fluctuations, and missing values ​​between different data sources on subsequent calculations.

[0029] The work cycle refers to the time period during which a tower crane completes one typical lifting operation. During the time period Internal identification of operating parameters Operating parameters Including lifting capacity Work range Rotation angle Operating wind speed Lifting height .

[0030] S3. Establish a finite element model of the tower crane tower body, and solve the stress response of each standard section under the corresponding working conditions based on the identified working parameters to construct a training sample dataset. The specific steps are as follows: Based on the tower crane's structural drawings, material parameters, geometric dimensions, and connection relationships, a finite element model of the tower crane's tower body is established. The finite element model should reproduce the actual structural condition as much as possible to improve the accuracy of stress response solutions for each standard section. The identified operating parameters The boundary conditions and load inputs of the finite element model are used for finite element simulation analysis, and the analysis results are as follows: Figure 3 As shown, the solution yields the first... Stress response of each standard section under corresponding working conditions ; A training sample dataset is constructed by solving multiple sets of working conditions. .

[0031] S4. Train the proxy model based on the constructed training sample dataset to establish a fast mapping relationship between operating condition parameters and stress response at key locations of standard sections; The surrogate model can be a feedforward neural network, a convolutional neural network, or other data-driven models suitable for regression prediction, denoted as... The model training process includes: dividing the sample data into training, validation, and test sets; setting the input, hidden, and output layer structures; optimizing parameters using the backpropagation algorithm; adjusting model parameters using the validation set to avoid overfitting; and verifying the model's prediction accuracy using the test set. The training objective is to enable the model to adapt to the input operating parameters. It can quickly output the stress response at key locations of the corresponding standard section, thereby avoiding the need to frequently call up time-consuming finite element solutions during field use.

[0032] The proxy model adopts a deep neural network structure with an attention mechanism, including an input layer, a one-dimensional convolutional feature extraction layer, an attention enhancement layer, and a fully connected output layer. A residual connection structure is provided between the one-dimensional convolutional feature extraction layer and the attention enhancement layer. The residual connection structure is used to ensure that the network maintains stable convergence performance when the number of layers is increased. The input layer is used to input the training sample dataset; The one-dimensional convolutional feature extraction layer extracts features through a one-dimensional convolutional network and outputs a high-dimensional feature map. The attention enhancement layer includes a channel attention submodule and a spatial attention submodule. The channel attention submodule adaptively weights the feature channels to enhance the combination of key operating parameters; the spatial attention submodule weights the feature spatial location to highlight potential stress concentration areas. A one-dimensional convolutional feature extraction layer extracts features from the input working condition parameters using one-dimensional convolution operations to obtain a high-dimensional feature map, thereby uncovering local combination relationships between working condition parameters and acquiring the high-dimensional feature map. The channel attention submodule generates channel weight vectors by performing global average pooling and fully connected mapping on the high-dimensional feature map, and then performs channel-wise multiplication with the high-dimensional feature map to obtain a channel-weighted feature map, enhancing the responsiveness to key working condition parameter combinations. The spatial attention submodule constructs a spatial weight mapping by compressing the high-dimensional feature map's dimensions, and then performs element-wise weighting with the high-dimensional feature map to obtain a spatially weighted feature map, highlighting the feature representation of potential stress concentration areas. The channel-weighted feature map and the spatially weighted feature map are fused to obtain the enhanced feature map; the fusion method is either element-wise weighted summation or element-wise multiplication fusion. The purpose of introducing the attention mechanism is to enhance the ability to represent the differences in the contribution of different working condition parameter combinations to the structural stress response, thereby improving the prediction accuracy of the surrogate model for complex nonlinear load-stress mapping relationships. The enhanced feature map outputs the stress response at key locations of the corresponding standard section through a fully connected output layer. The network structure parameters and attention weights of the surrogate model are obtained through offline training using a training sample dataset and are used consistently during actual operation. This achieves a high-precision approximation of the finite element calculation results, thus replacing the time-consuming finite element solution process in practical applications and enabling rapid prediction of stress response. Compared to the approximately 4 minutes typically required for a single finite element simulation calculation, the surrogate model described in this invention can output the structural stress response under the corresponding working condition within milliseconds, achieving real-time evaluation.

[0033] S5. Input the actual operating condition parameters into the trained surrogate model to obtain the stress cycle spectrum of each standard section; Operating parameters are arranged in chronological order. Input the trained proxy model to obtain the first Stress response of a standard section ; Rainflow counting was performed on the stress time history to obtain the first... Stress cycle spectrum of a standard section Its expression is: In the formula, Indicates the first Section 1 One stress amplitude range; Indicates the first The standard section in The number of cycles corresponding to each stress amplitude range Number of groups This is the group number.

[0034] S6, based on the stress cycle spectrum and combined with the fatigue damage accumulation theory, calculates the fatigue damage value and remaining life of the main structure of each standard section; The fatigue damage values ​​of the main structure of each standard section were calculated using the Miner linear cumulative damage method, and the calculation formula is as follows: In the formula, For the first Fatigue damage values ​​of a standard section; For the first The standard section in The number of cycles corresponding to each stress amplitude range; The allowable number of cycles is determined by the material fatigue performance curve. No. The remaining life of each standard section for: In the formula, For the first The damage threshold for the main structure of each standard section is typically set to 1.

[0035] During actual service, the fatigue damage value and remaining life of each standard section's main structure can be dynamically updated based on monitoring data.

[0036] S7, Establish weld sub-models for dangerous weld locations in standard sections and conduct weld life assessments; Based on the finite element model established in step S3, finite element analysis is performed on the overall structure to extract the critical points and regions of the weld. Using the displacement field of the finite element model as boundary conditions, a weld sub-model is established in the critical region, and the weld entity within it is modeled in detail. Figure 4 As shown; after obtaining the simulation results, the weld file was extracted using nCode weldline, and the weld life was analyzed using nCode Weld CAE Fatigue. The equivalent structural stress method was used to obtain the weld life. Combined with the results of operational condition monitoring Stress cycle statistics and fatigue damage accumulation were performed on dangerous areas of the weld. Among them, the Total fatigue damage of welds in each standard section for: In the formula, For the first In the standard section, the weld seam is in the first... The number of cycles corresponding to each stress amplitude range Number of groups The group number; For the first In the standard section, the weld seam is in the first... The number of permissible cycles for each stress amplitude range is determined by the material fatigue performance curve.

[0037] The fatigue life calculation cloud diagram of the main structure of the standard section is as follows: Figure 5 As shown in the figure, the fatigue life calculation cloud diagram of the weld is as follows: Figure 6 As shown, from Figure 5 It can be seen that the maximum allowable number of cycles (fatigue life) of the standard section's main structure is 3.521 × 10⁻⁶. 13 The lowest value was 5.199 × 10⁻⁶. 6 ,like Figure 6 The minimum permissible number of cycles (fatigue life) for the weld area shown is 4.802 × 10⁻⁶. 6 Secondly, the weld life is significantly lower than that of the main structure, thus enabling a coordinated fatigue damage assessment of the main structure and weld of the standard section, and clearly identifying the weld as a weak link in the fatigue control of the structure.

[0038] S8. The fatigue damage information of each standard section structure and weld is bound to a unique identifier to establish a standard section life cycle database. The standard section lifecycle database includes at least the following information: unique identifier, standard section model, installation history, dismantling history, current and historical section information, fatigue damage of the main structure, fatigue damage of the weld, remaining life of the main structure, remaining life of the weld, and history of load-bearing level.

[0039] The section information is based on the stress level of different sections in the whole tower, classifying each installation section into high stress level, medium stress level, and low stress level; the closer the section is to the bottom of the tower, the higher the stress level, and the closer the section is to the top of the tower, the lower the stress level. The specific classification scheme should be determined according to the tower crane model, design life, and enterprise management standards.

[0040] S9, based on the standard section lifecycle database and section stress conditions, performs reassembly determination and installation management. The specific steps are as follows: When a standard section needs to be reinstalled after disassembly, its damage and lifespan information is retrieved from the standard section lifecycle database using its unique identifier. If the fatigue damage to the main structure and welds of the standard section are both below their respective preset thresholds, and the remaining lifespan of the main structure and welds meets the usage requirements, the standard section is allowed to be installed in the corresponding stress-bearing section position. If the fatigue damage to the main structure is close to its preset threshold but still meets safety requirements, it is only allowed to be installed in a lower stress-bearing section position. If the fatigue damage to the main structure or welds of the standard section exceeds its preset threshold, corresponding structural safety control measures are triggered, including re-inspection, repair, or cessation of use. The preset thresholds are set according to the tower crane model, design life, enterprise management standards, and safety regulations. Through reinstallation judgment and installation management, standard sections with significant fatigue damage can be prevented from being reinstalled in high-stress areas for continued service. Simultaneously, standard sections with better lifespan conditions can be prioritized for placement in positions with higher load-bearing requirements, thereby improving the scientific rigor and safety of the tower crane's full lifecycle management.

[0041] This invention can transform actual operating condition monitoring data of tower cranes into stress response, main structural fatigue damage, and weld fatigue damage information at the individual standard section level. Furthermore, it correlates this information with the standard section's identification, installation history, and section stress level to form a continuously updated standard section lifecycle archive. Based on this, lifespan constraint management can be implemented for the reinstallation and relocation of standard sections, thereby achieving refined lifespan assessment and individualized lifespan management of the tower crane's tower structure during long-term service.

[0042] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

Claims

1. A method for intelligent life assessment and management of tower crane tower structure, characterized in that, include: S1, collect tower crane operating condition data and establish a unique identification for each standard section of the tower body; S2, based on the obtained operating condition data, preprocesses and identifies the operating condition parameters for each working cycle; Operating condition data is recorded according to a unified time base to form the original operating condition dataset. ,in, for The original runtime data set at each moment; for Lifting capacity at any given moment; for The range of work at any given moment; for The angle of rotation at any given moment; for The working wind speed at all times; for The lift height at any given moment; After preprocessing the original operating condition dataset, the preprocessed line operating condition dataset is obtained. ; For preprocessing Lifting capacity at any given moment; For preprocessing The range of work at any given moment; For preprocessing The angle of rotation at any given moment; For preprocessing The working wind speed at all times; For preprocessing The lift height at any given moment; Preprocessing includes time synchronization, outlier removal, missing value repair, noise filtering, unit unification, and normalization. The work cycle refers to the time period during which a tower crane completes one typical lifting operation. During the time period Internal identification of operating parameters Operating parameters Including lifting capacity Work range Rotation angle Operating wind speed Lifting height ; S3. Establish a finite element model of the tower crane tower body, and solve the stress response of each standard section under the corresponding working conditions based on the identified working condition parameters to construct a training sample dataset. Based on the tower crane's tower body structural drawings, material parameters, geometric dimensions, and connection relationships, a finite element model of the tower crane's tower body was established. The identified operating parameters As the boundary conditions and load inputs of the finite element model, the solution yields the first... Stress response of each standard section under corresponding working conditions ; A training sample dataset is constructed by solving multiple sets of working conditions. ; S4. Train the proxy model based on the constructed training sample dataset to establish a fast mapping relationship between operating condition parameters and stress response at key locations of standard sections; The surrogate model adopts a deep neural network structure with an attention mechanism, including an input layer, a one-dimensional convolutional feature extraction layer, an attention enhancement layer, and a fully connected output layer. There is a residual connection structure between the one-dimensional convolutional feature extraction layer and the attention enhancement layer. The input layer is used to input the training sample dataset; The one-dimensional convolutional feature extraction layer extracts features through a one-dimensional convolutional network and outputs a high-dimensional feature map. The attention enhancement layer includes a channel attention submodule and a spatial attention submodule; The one-dimensional convolutional feature extraction layer extracts features from the input working parameters using one-dimensional convolution operations to obtain a high-dimensional feature map. The channel attention submodule generates channel weight vectors by performing global average pooling and fully connected mapping on the high-dimensional feature map, and then performs channel-wise multiplication with the high-dimensional feature map to obtain a channel-weighted feature map. The spatial attention submodule constructs a spatial weight mapping by compressing the high-dimensional feature map, and then performs element-wise weighting with the high-dimensional feature map to obtain a spatially weighted feature map. The channel-weighted feature map and the spatially weighted feature map are fused to obtain an enhanced feature map. The enhanced feature map is then output through a fully connected output layer to output the stress response at the key position of the corresponding standard section. S5. Input the actual operating condition parameters into the trained surrogate model to obtain the stress cycle spectrum of each standard section; Operating parameters are arranged in chronological order. Input the trained proxy model to obtain the first Stress response of a standard section ; Rainflow counting was performed on the stress time history to obtain the first... Stress cycle spectrum of a standard section Its expression is: In the formula, Indicates the first Section 1 One stress amplitude range; Indicates the first The standard section in The number of cycles corresponding to each stress amplitude range Number of groups The group number; S6, based on the stress cycle spectrum and combined with the fatigue damage accumulation theory, calculates the fatigue damage value and remaining life of the main structure of each standard section; S7, Establish weld sub-models for dangerous weld locations in standard sections and conduct weld life assessments; S8. The fatigue damage information of each standard section structure and weld is bound to a unique identifier to establish a standard section life cycle database. S9 performs reassembly determination and installation management based on the standard section life cycle database and the stress conditions of the section.

2. The intelligent life assessment and management method for tower crane tower structure according to claim 1, characterized in that, In step S6, the fatigue damage value of each standard section's main structure is calculated using the Miner linear cumulative damage method, and the calculation formula is as follows: In the formula, For the first Fatigue damage values ​​of a standard section; For the first The standard section in The number of cycles corresponding to each stress amplitude range; For the first The standard section in The allowable number of cycles for each stress amplitude range is determined by the material fatigue performance curve; No. The remaining life of each standard section for: In the formula, For the first Damage threshold of each standard section of the main structure.

3. The intelligent life assessment and management method for tower crane tower structure according to claim 1, characterized in that, In step S7, based on the finite element model established in step S3, finite element analysis is performed on the overall structure to extract the dangerous points and dangerous areas of the weld dangerous parts. The displacement field of the finite element model is used as the boundary condition to establish a weld sub-model in the dangerous area and to perform fine modeling of the weld entity in it. Based on service history data, stress cycle statistics and fatigue damage accumulation were performed on dangerous areas of the weld. Among them, the Total fatigue damage of welds in each standard section for: In the formula, For the first In the standard section, the weld seam is in the first... The number of cycles corresponding to each stress amplitude range Number of groups The group number; For the first In the standard section, the weld seam is in the first... The allowable number of cycles for each stress amplitude range is determined by the material fatigue performance curve; The welds include welds between the main chord and the connecting parts, welds connecting the web members, and welds on the end plates, as well as welded connection parts that are prone to fatigue crack initiation. The weld life analysis method selects the nominal stress method, hot spot stress method, or equivalent structural stress method according to the weld structure and stress characteristics to achieve adaptive analysis of fatigue damage for different welding connection forms.

4. The intelligent life assessment and management method for tower crane tower structure according to claim 1, characterized in that, In step S8, the standard section lifecycle database includes at least the following information: unique identifier, standard section model, installation history, dismantling history, current and historical section information, fatigue damage of main structure, fatigue damage of weld, remaining life of main structure, remaining life of weld, and load-bearing level history.

5. The intelligent life assessment and management method for tower crane tower structure according to claim 4, characterized in that, The section information is based on the stress level of different sections in the whole tower, dividing each installed section into high stress level, medium stress level and low stress level; the section closer to the bottom of the tower has a higher stress level, and the section closer to the top of the tower has a lower stress level.

6. The intelligent life assessment and management method for tower crane tower structure according to claim 5, characterized in that, The specific steps of step S9 are as follows: When a standard section needs to be reinstalled after disassembly, its damage status and lifespan status information are retrieved from the standard section lifecycle database using its unique identifier. When the fatigue damage of the main structure and the weld fatigue damage of the standard section are both below their respective preset thresholds, and the remaining lifespan of the main structure and the weld of the standard section meets the usage requirements, the standard section is allowed to be installed in the corresponding stress level section. When the fatigue damage of the main structure is close to its preset threshold but still meets the safety requirements, it is only allowed to be installed in the low stress level section. When the fatigue damage of the main structure or the weld fatigue damage of the standard section exceeds its preset threshold, the corresponding structural safety control measures are triggered, including re-inspection, repair, or cessation of use.

Images