Tower cylinder welding parameter optimization method based on low temperature environment

Abstract

The present application relates to the technical field of industrial intelligent control, and particularly relates to a tower drum welding parameter optimization method based on a low-temperature environment. The method comprises the following steps: performing tower drum welding equivalent transient heat input field analysis based on low-temperature welding multi-dimensional perception data to generate tower drum welding equivalent transient heat input field data; performing weld structure prediction performance evaluation processing based on the tower drum welding equivalent transient heat input field data to generate weld structure prediction performance evaluation data; performing tower drum welding structure risk feature analysis through the weld structure prediction performance evaluation data to generate tower drum welding structure risk feature data; and performing intelligent control strategy design for tower drum welding parameter optimization based on the weld structure prediction performance evaluation data and the tower drum welding structure risk feature data to generate a tower drum welding parameter optimization strategy. The present application realizes precise and efficient welding parameter intelligent optimization adjustment by comprehensively monitoring and analyzing the tower drum welding process under a low-temperature environment.

Description

Technical Field

[0001] This invention relates to the field of industrial intelligent control technology, and in particular to a method for optimizing tower welding parameters based on low-temperature environments. Background Technology

[0002] As the core load-bearing structure in new energy and power facilities such as wind power and power transmission, the manufacturing and installation quality of the tower directly determines the safe and stable operation, load-bearing capacity, and service life of the entire facility. With the rapid expansion of the new energy industry into high-latitude, high-altitude, and other low-temperature regions, the on-site installation and welding operations of equipment such as wind turbine towers and transmission towers often face low-temperature environments. Low temperatures have become a key constraint affecting the welding quality of towers. During tower welding in low-temperature environments, the initial temperature of the base material is low, and the heat loss rate in the welding area is rapid, easily leading to a large temperature gradient between the weld area and the base material. This can cause problems in tower welding, increasing the safety risks of tower structural failure and collapse. However, existing tower welding parameters are mostly designed based on ambient temperature environments, failing to fully consider the coupled influence of multiple factors such as substrate temperature, ambient temperature, and arc characteristics on welding quality under low-temperature conditions. This results in significant blind optimization of parameters and poor adaptability. Furthermore, the lack of multi-dimensional sensing and monitoring methods for the low-temperature welding process makes it impossible to comprehensively and in real-time acquire key data such as the welding substrate environment and arc state. Consequently, it is impossible to effectively predict the evolution of weld microstructure and mechanical properties, making it difficult to balance the quality stability and structural reliability of tower welding under low-temperature conditions. This fails to fundamentally solve the problems of frequent low-temperature welding defects and the difficulty in welding quality control, and ultimately fails to meet the actual needs of tower welding operations in low-temperature regions. Summary of the Invention

[0003] Based on this, the present invention provides a method for optimizing tower welding parameters in a low-temperature environment to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above objectives, a method for optimizing tower welding parameters based on a low-temperature environment, wherein the low-temperature environment is a low-temperature operating environment of no more than 0°C in the tower welding scenario, includes the following steps:

[0005] Step S1: Obtain basic data of tower welding; perform multi-dimensional sensing and monitoring of the low-temperature environment of tower welding through sensor monitoring integrated equipment and basic data of tower welding to generate multi-dimensional sensing data of low-temperature welding.

[0006] Step S2: Analyze the equivalent transient heat input field of tower welding based on low-temperature welding multidimensional sensing data to generate equivalent transient heat input field data for tower welding;

[0007] Step S3: Based on the equivalent transient heat input field data of tower welding, perform weld microstructure prediction performance evaluation processing to generate weld microstructure prediction performance evaluation data;

[0008] Step S4: Analyze the risk characteristics of the tower welded structure using the weld microstructure prediction performance evaluation data, and generate risk characteristic data of the tower welded structure;

[0009] Step S5: Based on the weld microstructure prediction performance evaluation data and the tower weld structure risk characteristic data, design an intelligent control strategy for tower welding parameter optimization and generate a tower welding parameter optimization strategy.

[0010] The beneficial effects of this invention are as follows: The method for optimizing tower welding parameters in a low-temperature environment systematically acquires basic tower welding data and conducts multi-dimensional sensing and monitoring processing under low-temperature conditions (≤0℃). It scientifically deploys monitoring nodes by combining global tower welding characteristics and geometric topology analysis. Then, through spatial interpolation, it fuses the monitoring and topology data to generate comprehensive and accurate low-temperature welding multi-dimensional sensing data. This solves the technical defects of existing low-temperature welding monitoring data, such as insufficient support, single dimension, and high degree of arbitrariness, providing a comprehensive and reliable basic data source for subsequent welding parameter optimization. It also improves the standardization and operability of low-temperature welding monitoring. Based on the low-temperature welding multi-dimensional sensing data, it conducts equivalent transient heat input field analysis. Through three-dimensional heat conduction finite element discrete modeling, low-temperature thermal energy boundary condition analysis, and welding heat source characteristic analysis, combined with equivalent heat input prior data, a correction model is established. This accurately captures the heat conduction law and dynamic distribution evolution characteristics of heat input in low-temperature welding, clearly reflecting the coupling relationship between heat input, low-temperature conditions, and welding structure. This overcomes the problem of large deviations in traditional analysis and provides accurate thermal support for weld microstructure prediction. Based on equivalent transient heat input field data, a predictive performance assessment of weld microstructure is conducted. By analyzing the characteristics of welding cooling and weld microstructure evolution, the low-temperature phase transformation mapping path is identified. Combined with microstructure distribution and structural trend analysis, the weld microstructure and mechanical properties are predicted. Then, a quantitative assessment is completed through a preset evaluation index, achieving early prediction and precise quantification of weld quality. This solves the problem that existing technologies cannot accurately predict the microstructure and mechanical properties of low-temperature welds and cannot avoid defects in advance, providing a core quality basis for subsequent risk analysis and parameter optimization. Risk characteristic analysis of tower welded structures is also conducted using weld microstructure predictive performance assessment data. First, weak points in weld performance are located based on the assessment data, and residual stress analysis is carried out. Then, combined with prior data on material fracture under low-temperature conditions, the weld crack sensitivity characteristics and structural risk analysis are completed. This achieves precise matching between crack-sensitive areas and stress concentration areas, quantifies the risk level, and solves the defects of weak risk analysis and disconnection from weld quality, clarifying a precise risk avoidance direction for parameter optimization. Based on weld microstructure prediction performance evaluation data and tower weld structure risk characteristic data, an intelligent control strategy for welding parameter optimization was designed. By analyzing optimization needs and clarifying optimization priorities, and combining weld quality with multi-objective optimization feature analysis, the welding parameters were accurately matched with low-temperature conditions, tower structure, and weld quality. This effectively solved the problems of blind optimization and poor adaptability of existing low-temperature welding parameters. The optimized parameters can fundamentally reduce defects such as welding cracks and incomplete penetration, reduce the risk of tower structure failure, and balance the quality stability, structural reliability, and operational efficiency of low-temperature welding.It improves the efficiency and accuracy of tower welding parameter optimization under low temperature conditions of ≤0℃, reduces the labor cost and quality control difficulty of welding operations, and can fully adapt to the actual needs of tower welding operations in low-temperature areas such as high latitude and high altitude, providing reliable technical support for tower welding quality control under low temperature conditions. Attached Figure Description

[0011] Figure 1 This is a schematic diagram of the steps of a method for optimizing tower welding parameters based on a low-temperature environment according to the present invention.

[0012] Figure 2 for Figure 1 A detailed flowchart illustrating the implementation steps of step S3.

[0013] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0014] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0015] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. Functional entities may be implemented in software, or in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods. The term "and / or" as used herein includes any and all combinations of one or more of the associated items listed.

[0016] To achieve the above objectives, please refer to Figures 1 to 2 This invention provides a method for optimizing tower welding parameters based on low-temperature environments. In the embodiments of this invention, please refer to... Figure 1 The diagram shown is a flowchart illustrating the steps of a tower welding parameter optimization method based on a low-temperature environment according to the present invention. The low-temperature environment refers to a low-temperature working condition where the tower welding scenario is no higher than 0°C. The tower welding parameter optimization method based on a low-temperature environment includes the following steps:

[0017] Step S1: Obtain basic data of tower welding; perform multi-dimensional sensing and monitoring of the low-temperature environment of tower welding through sensor monitoring integrated equipment and basic data of tower welding to generate multi-dimensional sensing data of low-temperature welding.

[0018] In this embodiment of the invention, comprehensive collection of basic data for tower welding is carried out. The acquired basic data is strictly limited to tower welding geometric data and tower welding material data. The geometric data includes the overall structural dimensions of the tower body, the weld bevel structure, the related geometric parameters of the bevel, the segment connection dimensions of the body, and the cross-sectional characteristics of the weld area to be welded. The material data includes the chemical composition of the tower base material and the welding filler material, the thermophysical properties under low temperature conditions, the critical parameters of solid phase transformation, and the low temperature mechanical properties. All basic data are derived from the tower design technical documents and the special test reports of the materials, ensuring complete consistency with the tower structure and material system of the actual welding operation. After acquiring the basic data, low-temperature multi-dimensional sensing and monitoring of the tower welding was immediately carried out. First, global feature analysis was performed on the basic data to extract global features of the geometric structure and material properties. Then, based on the global features, the geometric spatial topology of the tower welding was constructed to clarify the spatial partitioning, structural relationships, and heat conduction paths of the weld and surrounding areas. Subsequently, sensor monitoring integrated equipment was deployed based on the topology. The monitoring equipment covers monitoring types such as environmental parameters, base material temperature, and welding arc characteristics. The monitoring nodes strictly correspond to high-risk areas of heat loss, the core area of ​​the weld, and the heat-affected zone in the topology. The monitoring process was started synchronously and continuously with the welding operation. The system comprehensively collects environmental data, substrate temperature data, and arc characteristic data throughout the entire welding process. Finally, it performs spatial interpolation processing on the collected monitoring data to fill data gaps in areas where no monitoring nodes are deployed, constructing a continuous data distribution across the entire weld space and multiple monitoring dimensions. This generates multi-dimensional sensing data for low-temperature welding, which is fully associated with the topology, monitoring parameters, and spatial location. This provides comprehensive and accurate basic data support for subsequent equivalent transient heat input field analysis. The core logic of this system is to achieve accurate capture of multi-dimensional data throughout the entire tower welding process under low-temperature conditions through basic data modeling, topology construction, real-time monitoring, and data completion.

[0019] Step S2: Analyze the equivalent transient heat input field of tower welding based on low-temperature welding multidimensional sensing data to generate equivalent transient heat input field data for tower welding;

[0020] In this embodiment of the invention, a three-dimensional heat conduction analysis model that is completely consistent with the actual tower welding structure is constructed using the topological structure bound by the multi-dimensional sensing data of low-temperature welding as the spatial framework. The model strictly matches the geometric dimensions and material properties in the multi-dimensional sensing data. Subsequently, the three-dimensional model is discretized by finite element method, dividing the model into multiple discrete mesh units. The mesh division follows the principle of refining the core area of ​​the weld and the fusion line area, and coarsening the main body area of ​​the substrate, ensuring that the discrete model can meet the requirements of calculation accuracy and improve calculation efficiency. At the same time, the initial temperature of the substrate and the low-temperature thermophysical performance parameters of the material in the multi-dimensional sensing data are assigned to the corresponding mesh units one by one, clarifying the initial state and material properties of each mesh unit. After discrete modeling, low-temperature thermal boundary condition analysis is conducted to accurately identify convective heat transfer boundaries and internal heat conduction boundaries in the model. Combined with parameters such as ambient temperature and wind speed from multi-dimensional sensing data, heat exchange parameters for each boundary unit are calculated and assigned. These boundary parameters are updated synchronously with the welding time step, maintaining consistency with the monitoring data acquisition frequency. Subsequently, based on discrete grid data and boundary condition data, dynamic characteristics analysis of the low-temperature tower's thermal energy is performed. According to transient heat conduction theory, the dynamic changes in the temperature field of each grid unit throughout the entire welding time history are calculated, and core thermal characteristic parameters such as peak temperature, cooling rate, and temperature gradient are extracted. Simultaneously, welding arc monitoring characteristic data are extracted from the multi-dimensional sensing data to conduct welding heat source characteristic analysis, clarifying the spatial distribution, energy distribution law, and dynamic fluctuation characteristics of the welding heat source, and determining the core parameters of the heat source model. In addition, prior data of equivalent heat input for tower welding that matches the welding conditions were specifically acquired. Equivalent heat input offset correction relationship was established by combining multi-dimensional sensing data, and equivalent heat input correction model was constructed. Finally, dynamic thermal energy characteristic data and heat source characteristic data were input into the correction model to conduct equivalent transient heat input field analysis. The nominal heat input of welding was corrected in real time, and the effective heat input distribution and dynamic evolution data of the entire weld space and the entire time step under low temperature environment were obtained. Equivalent transient heat input field data of tower welding was generated. The core logic of its implementation is to accurately restore the spatial distribution and dynamic change law of the actual effective heat input of tower welding under low temperature environment through modeling, boundary analysis, heat source analysis and correction calculation.

[0021] Step S3: Based on the equivalent transient heat input field data of tower welding, perform weld microstructure prediction performance evaluation processing to generate weld microstructure prediction performance evaluation data;

[0022] In this embodiment of the invention, based on the full-grid cell and full-time-step temperature field data in the equivalent transient heat input field data, complete welding thermal cycle curves corresponding to each spatial location of the weld fusion zone, welding heat-affected zone, and substrate zone are extracted. Combined with the solid-state phase transformation critical parameters of the tower substrate and welding filler material, low-temperature phase transformation mapping path analysis of the weld is conducted. Using the welding thermal cycle time axis as a reference, the correspondence between temperature change and the solid-state phase transformation process is established, clarifying the start and end nodes, phase transformation type, and transformation process of each phase transformation stage. Simultaneously, the shift in critical phase transformation temperature caused by rapid heat dissipation and large supercooling under low-temperature conditions is corrected, establishing a three-dimensional temperature-time-phase transformation mapping relationship across the entire weld space, generating low-temperature phase transformation mapping path data for the weld. Subsequently, using the continuous cooling transformation curve of the material as a reference, the phase transformation path at each spatial location is matched with the transformation curve to determine the core microstructure parameters such as the type, phase ratio, and grain size of the phase transformation products after cooling. Simultaneously, the weld is divided into multiple feature regions, and the microstructure distribution characteristics of each region are extracted, clarifying the microstructure differences and spatial distribution patterns of different regions, generating weld microstructure distribution characteristic data. Next, based on the phase transformation mapping path data, the dynamic evolution characteristics of the microstructure are extracted along the welding thermal cycle time axis. The dynamic laws of austenite grain growth during the heating stage and solid-state phase transformation during the cooling stage are analyzed. Simultaneously, the microstructure change trends in different spatial directions of the weld and the secondary evolution characteristics of multi-layer, multi-pass welds are analyzed, generating weld microstructure trend characteristic data. Based on this, a quantitative mapping relationship between weld microstructure characteristics and mechanical properties is established. Combining microstructure distribution characteristics and microstructure trend characteristics, core mechanical property parameters such as strength, hardness, low-temperature toughness, and plasticity at each spatial location are calculated. The cold crack sensitivity index of the weld is calculated in particular, clarifying the crack sensitivity characteristics and generating weld microstructure mechanical prediction property distribution characteristic data. Finally, a preset low-temperature welding performance evaluation index is used to comprehensively evaluate the mechanical prediction performance data. This evaluation index includes multiple core evaluation sub-items, each with a fixed qualification threshold and weight coefficient, strictly matching the service safety requirements of tower welding in low-temperature environments. By comparing performance parameters with evaluation thresholds one by one, a comprehensive evaluation score is calculated to clarify the performance compliance status, weak areas, and performance shortcomings of each location, conduct overall regional performance evaluation, and generate weld microstructure prediction performance evaluation data. Its core logic is to comprehensively quantify the microstructure and performance level of tower welding welds in low-temperature environments through phase transformation analysis, microstructure analysis, mechanical prediction, and performance evaluation.

[0023] Step S4: Analyze the risk characteristics of the tower welded structure using the weld microstructure prediction performance evaluation data, and generate risk characteristic data of the tower welded structure;

[0024] In this embodiment of the invention, the weld microstructure prediction performance evaluation data is decomposed in all dimensions, and core contents such as mechanical performance parameters, performance compliance status, comprehensive evaluation score, weak area location, and performance defect type of the entire weld space are extracted. Based on the weld groove centerline, the entire weld is divided into continuous analysis units, each unit corresponding to a unique spatial coordinate. At the same time, partition analysis is carried out according to the weld characteristic regions, and the performance mean, dispersion, performance shortcomings and weak points distribution of each region are extracted. The performance differences between the windward and leeward sides of the tower and the surface and root of the weld are analyzed in detail under low temperature conditions, generating weld microstructure prediction performance status characteristic data, accurately presenting the spatial distribution and overall state of weld performance. Subsequently, based on the performance status characteristic data, the residual stress characteristics of the tower welding were analyzed. Using the three-dimensional geometric model of the tower welding as the calculation carrier, the material mechanical property parameters of each analysis unit were assigned to the corresponding calculation unit to clarify the material mechanical boundaries. Combining the welding thermal cycle temperature history and weld phase transformation mapping path data obtained above, and based on the relevant theories of thermo-elastic-plastic deformation, the residual stress of the entire welding process was numerically calculated. During the calculation, the superposition effects of large temperature gradient, uneven thermal expansion and contraction, and solid phase transformation volume change on residual stress under low temperature environment were fully considered. The non-uniform deformation effect caused by high cooling rate on the windward side was corrected. After the calculation was completed, the residual stress data in the longitudinal, transverse, and wall thickness directions of each position in the entire weld space were extracted to clarify the residual stress peak, stress concentration area, and stress gradient change. The spatial overlap between the residual stress concentration area and the performance weak area was analyzed to generate the residual stress characteristic data of the tower welding. To conduct a risk characteristic analysis of the tower welded structure, prior fracture data of the low-temperature tower welding material, which perfectly matches the welding conditions, is first obtained. Using the fracture critical parameters and crack resistance thresholds in this data as a benchmark, combined with residual stress characteristic data and performance status characteristic data, the crack sensitivity level of each analysis unit is analyzed, and the ratio of residual stress to the material's crack resistance critical value is calculated to clarify the risk of cold crack initiation. Simultaneously, considering the actual service conditions of the tower, the superposition effect of residual stress and service load is analyzed to assess the risks of brittle fracture and fatigue failure. Subsequently, using residual stress level, crack sensitivity level, and load superposition effect as core indicators, the entire weld is divided into different risk levels. The spatial location, distribution range, risk type, core causes, and severity of each risk level are clarified, establishing a structural risk distribution model for the entire weld space and generating risk characteristic data for the tower welded structure. The core logic is to accurately predict the potential failure risk of the tower welded structure under low-temperature conditions through performance status analysis, residual stress calculation, and risk assessment.

[0025] Step S5: Based on the weld microstructure prediction performance evaluation data and the tower weld structure risk characteristic data, design an intelligent control strategy for tower welding parameter optimization and generate a tower welding parameter optimization strategy.

[0026] In this embodiment of the invention, a tower welding parameter optimization requirement analysis is conducted. Structural risk characteristic data is decomposed, and the spatial location, risk type, triggering factors, and severity of high- and medium-risk areas are extracted. Combined with performance evaluation data showing weak performance areas, shortcoming types, and differences in qualification thresholds, optimization target priorities are divided based on risk level. Areas with high risk of cold cracking and insufficient low-temperature toughness are listed as the highest priority. Subsequently, the relationship between the risk root cause and welding parameters of each optimization target is analyzed, clarifying the welding parameter optimization direction corresponding to different risks and performance shortcomings. The parameter adjustment boundary for each optimization direction is determined. The optimization targets, priorities, root cause analysis, optimization directions, and boundary requirements are integrated to generate tower welding parameter optimization requirement data, clarifying the core objectives and specific requirements of parameter optimization. Subsequently, combining performance evaluation data and optimization requirement data, a multi-objective optimization characteristic analysis of tower welding quality was conducted. Core optimization variables such as welding current, arc voltage, welding travel speed, preheating temperature, and interpass temperature were identified, and the value range of each variable was set. The core objectives of multi-objective optimization were clarified, including weld performance compliance, crack resistance improvement, residual stress control, and operational efficiency assurance. Fixed weight coefficients were assigned to each optimization objective. At the same time, constraints such as welding parameter matching specifications, arc stable combustion boundary, and low-temperature operation safety requirements were clarified. The influence of optimization variables on each optimization objective and the multi-variable coupling effect were analyzed, generating multi-objective optimization characteristic data of tower welding quality and establishing a parameter optimization framework. Based on this, iterative simulation optimization of tower welding parameters was carried out. Using multi-objective optimization feature data as a basis, and relying on the three-dimensional heat conduction finite element model, microstructure prediction model and residual stress calculation model established above, a multi-objective iterative optimization method was adopted to carry out full-process iterative calculation. Each iteration generated multiple sets of parameter combinations. The performance and risk prediction results corresponding to each set of parameters were obtained through model calculation. The optimal parameter combination was selected by comparing the optimization objectives and gradually approaching the global optimal solution. After the iteration was completed, the parameter combination with the best comprehensive performance was selected, the optimization effect of each set of parameters was clarified, and iterative optimization data of tower welding parameters was generated. Finally, based on iterative optimization data, intelligent control strategies were designed. Combining the characteristics of the tower welding structure, the multi-layer and multi-pass welding process, and the dynamic changes in the low-temperature environment, a closed-loop control strategy was designed for each region, process, and the entire process. Differentiated optimal parameters were matched for regions with different heat dissipation conditions and performance requirements. Specific parameters were set for each process, including root pass welding, fill pass welding, and cap pass welding. A dynamic closed-loop adjustment mechanism was designed in conjunction with sensor monitoring equipment to respond in real time to fluctuations in environmental parameters such as temperature and wind speed, automatically adjusting welding parameters to compensate for heat input loss. The control logic, parameter matching rules, dynamic adjustment trigger conditions, and anomaly response measures were integrated to generate an optimized strategy for tower welding parameters. The core logic of this strategy is to achieve intelligent and precise control of tower welding parameters in a low-temperature environment through demand analysis, multi-objective optimization, iterative simulation, and strategy design, thereby reducing structural risks at the source and ensuring that weld performance meets standards.

[0027] Furthermore, step S1 includes the following steps:

[0028] Step S11: Obtain the basic data for tower welding, wherein the basic data for tower welding includes tower welding geometry data and tower welding material data;

[0029] In this embodiment of the invention, the basic data for tower welding includes tower welding geometry data and tower welding material data. The tower welding geometry data includes the nominal diameter of the tower body, the wall thickness, the bevel type, bevel angle, blunt edge height, spatial circumferential orientation of the bevel surface, the axial length of the tower body segments, the structural dimensions of the connection between the tower flange and the tower body, and the cross-sectional profile dimensions of the weld area. The tower welding material data includes the grade of the tower base material, the mass percentage of each chemical element, the yield strength, tensile strength, and impact strength at -40℃. Toughness, thermal conductivity, specific heat capacity, austenitic phase transformation initiation and termination temperatures, martensitic phase transformation critical temperature, welding filler material type, chemical composition of the weld metal, impact toughness at -40℃, thermophysical property parameters of the weld metal, and low-temperature welding compatibility parameters of the base material and filler material. All basic data correspond to the target tower structure and materials used in tower welding operations under low-temperature conditions. The data sources are tower design drawings and material factory performance test reports, ensuring that the basic data are completely consistent with the tower structure and material properties of the actual welding operation.

[0030] Step S12: Perform global feature analysis on the basic data of tower welding to generate global basic feature data of tower welding;

[0031] In this embodiment of the invention, for the tower welding geometry data, the overall structural load-bearing characteristics of the tower body, the continuous distribution characteristics of the weld bevel along the tower circumference and axis, the cross-sectional variation characteristics of the tower wall thickness, the geometric constraint characteristics of the bevel cross-section, the heat dissipation surface distribution characteristics of the inner and outer walls of the tower, and the heat conduction path characteristics in the axial and circumferential directions of the tower body are extracted. For the tower welding material data, the low-temperature environment adaptation characteristics of the base material and the welding filler material, the global distribution characteristics of the material's thermophysical properties with temperature changes, the critical parameter characteristics of the material's solid-state phase transformation, the variation characteristics of the material's low-temperature mechanical properties, and the fusion performance characteristics of the base material and the filler material are extracted. The extracted geometric global features and material global features are fused. During the fusion process, the weld area to be welded is used as the core associated node, and the geometric features and material features at the corresponding positions are bound together to form tower welding global basic feature data covering the entire structural range and all material properties of the tower welding, and adapting to the low-temperature welding operation scenario, providing feature basis for subsequent geometric space topology analysis.

[0032] Step S13: Perform geometrical topological analysis of the tower welding based on the global basic feature data of the tower welding, and generate tower welding topological data;

[0033] In this embodiment of the invention, a three-dimensional geometric space model of tower welding is constructed based on the global basic feature data of tower welding. The area to be welded at the weld bevel is taken as the core topological node. An independent topological unit is divided every 15° along the circumference of the tower body. In the axial direction, three layers of topological sub-units are divided along the wall thickness direction within a range of 200mm on both sides of the bevel, corresponding to the outer wall of the tower, the middle layer of the base material, and the inner wall of the tower, respectively. Each topological unit and topological sub-unit has clearly defined three-dimensional spatial coordinates, geometric dimensions, material grade, and heat dissipation surface attributes. The physical connection relationship and heat conduction path association relationship between adjacent topological units are established. In view of the strong heat loss characteristics of the welding process in low temperature environment, the topological sub-units are refined in the windward area of ​​the tower, the edge area of ​​the bevel, and the area of ​​abrupt change in wall thickness. The spatial boundary range, heat conduction attributes, and association relationship with adjacent topological sub-units of each refined topological sub-unit are defined. This forms tower welding topological structure data covering the entire tower welding structure and including spatial location information, geometric attributes, material attributes, and heat conduction path information, providing a spatial framework for subsequent monitoring node layout and spatial interpolation processing.

[0034] Step S14: Based on the sensor monitoring integrated equipment and tower welding topology data, perform environmental monitoring processing of the tower welding substrate to generate environmental monitoring data of the tower welding substrate;

[0035] In this embodiment of the invention, the topological attribute feature analysis of the tower welding topology data is first performed to extract the heat conduction sensitivity attribute, weld bevel location attribute, base material wall thickness attribute, and low-temperature heat loss high-risk attribute of each topological unit. Based on the extracted topological attribute features, tower welding sensing and monitoring nodes are designed. Base material temperature monitoring nodes are deployed in the windward topological unit with high low-temperature heat loss risk. In the topological unit corresponding to the weld bevel, a base material surface temperature monitoring node is deployed every 30° along the circumference. Two sets of environmental parameter monitoring nodes are deployed on the windward and leeward sides of the tower welding operation area. The arc feature acquisition module of the welding equipment moves synchronously with the welding torch. The sensor monitoring integrated device includes an ambient temperature sensor, a base material surface temperature sensor, a welding arc voltage and current acquisition module, and an ambient wind speed sensor. Sensors, with all monitoring nodes fixedly deployed according to the three-dimensional spatial positions of their corresponding topological units, initiate data acquisition synchronously with the welding operation. The acquisition frequency is matched to the welding travel speed; when the welding travel speed is 300 mm / min, the acquisition frequency is 10 Hz. During the acquisition process, the tower welding space environment data and the tower substrate welding arc monitoring characteristic data are acquired simultaneously. The tower welding space environment data includes the ambient temperature and ambient wind speed of the welding operation area, while the tower substrate welding arc monitoring characteristic data includes the initial temperature of the substrate, the real-time temperature of the substrate during the welding process, the voltage and current of the welding arc, and the welding travel speed. This generates tower welding substrate environmental monitoring data that is synchronous with the welding operation process and covers all monitoring dimensions, providing real-time monitoring data for subsequent spatial interpolation processing and equivalent transient heat input field analysis.

[0036] Step S15: Transmit the environmental monitoring data of the tower welding substrate to the tower welding topology data for spatial interpolation processing of the substrate environmental monitoring to generate low-temperature welding multidimensional sensing data.

[0037] In this embodiment of the invention, the environmental monitoring data of the tower welding substrate is matched to the corresponding topological units of the tower welding topology data according to the three-dimensional spatial position of the topological units corresponding to the monitoring nodes. Taking the data collected by each monitoring node as the baseline value, based on the spatial position relationship and heat conduction path correlation between the topological units, the Kriging spatial interpolation method is used to perform data interpolation calculation on the topological units without monitoring nodes. This yields spatially continuous distribution data of the ambient temperature, substrate temperature, and arc characteristic parameters covering the entire topology. Combining the material properties and geometric properties of the topology, the interpolated data is subjected to spatial dimension feature matching. The environmental parameters, substrate parameters, and arc parameters are bound to the corresponding topological units, so that each topological unit has corresponding environmental characteristics, substrate temperature characteristics, and welding heat source characteristics. This forms low-temperature welding multi-dimensional sensing data that covers the entire space range and all monitoring dimensions of tower welding and is deeply integrated with the topology. This data directly provides the initial and boundary conditions of the entire space for subsequent three-dimensional heat conduction finite element discrete modeling, eliminating the limitations of single-point monitoring data and avoiding deviations in subsequent equivalent transient heat input field analysis. This provides comprehensive and accurate basic data support for the entire welding parameter optimization process.

[0038] Furthermore, step S14 includes the following steps:

[0039] Step S141: Perform tower welding topology attribute feature analysis on the tower welding topology data to generate tower welding topology attribute data;

[0040] In this embodiment of the invention, the tower welding topology data is decomposed into units to clarify the spatial boundaries, unique spatial identifiers, and relationships between adjacent units of all topological units and sub-units within the entire structure. The tower welding topology data includes independent topological units divided along the circumference and axial direction of the tower, with the weld bevel as the core node, as well as refined topological sub-units of the outer wall, inner wall, and bevel regions of the tower. Then, multi-dimensional attribute feature extraction is performed on all the decomposed topological units. The first dimension extracts geometric structural attribute features, clarifying the three-dimensional spatial coordinates, wall thickness, bevel cross-sectional parameters, heat dissipation surface ratio, and heat conduction path length of each topological unit. The heat dissipation surface ratio of the tower outer wall topological unit is 100%, the heat dissipation surface ratio of the tower inner wall topological unit is 0%, and the heat dissipation surface of the bevel edge topological unit covers both sides of the facade and end face. The second dimension extracts material attribute features, clarifying the grade of the base material corresponding to each topological unit, the thermophysical performance parameters under low temperature conditions, etc. The critical temperature of solid-state phase transformation and the performance parameters of the weld filler metal corresponding to the topological units of the bevel welding area are used to extract low-temperature welding sensitive attribute features in the third dimension. The low-temperature heat loss risk level, the coverage range of the welding heat-affected zone, and the weld fusion correlation of each topological unit are clarified. Among them, the low-temperature heat loss risk level of the topological unit on the windward side of the tower is the highest. The welding heat-affected zone coverage and the weld fusion correlation of the topological units within 20mm on both sides of the bevel are 100%. After completing the full-dimensional feature extraction, the geometric structure attribute features, material attribute features, and low-temperature welding sensitive attribute features are bound to the unique spatial identifier of the corresponding topological unit. The attribute coupling mapping relationship between adjacent topological units is established, and the attribute correlation weight of heat conduction adjacent units is clarified. This forms tower welding topological structure attribute data that covers the entire tower welding topology, contains multi-dimensional attribute features, and is adapted to the needs of low-temperature welding monitoring, providing a core basis for the accurate deployment of subsequent monitoring nodes.

[0041] Step S142: Design tower welding sensing and monitoring node data through tower welding topology attribute data, and use the sensor monitoring integrated equipment configured with tower welding sensing and monitoring node data to perform tower welding substrate environmental monitoring processing to generate tower welding substrate environmental monitoring data.

[0042] In this embodiment of the invention, based on the topological attribute data of the tower welding structure, three types of sensing and monitoring nodes are divided according to the low-temperature welding sensitivity attributes, weld fusion correlation, and low-temperature heat loss risk level of the topological unit. The first type is the base material temperature monitoring node, which is fixed at the topological unit with the highest low-temperature heat loss risk level and the highest weld fusion correlation. One base material temperature monitoring node is deployed every 30° along the circumference of the tower and within the corresponding topological unit at the bevel. In the high-risk area on the windward side, the density is increased to one node every 15°. Base material temperature monitoring nodes are deployed on the outer wall, the middle layer of the base material, and the inner wall of the tower in the wall thickness direction, respectively, covering the entire range of the welding heat-affected zone. The second type is the environmental parameter monitoring node. The monitoring nodes are fixed in the surrounding space of the corresponding topological units on the windward and leeward sides of the tower welding operation area, as well as at the top and bottom of the tower. Two sets of environmental parameter monitoring nodes are deployed in each area. A third type of node, the welding arc characteristic monitoring node, is synchronously bound to the welding torch and moves along the corresponding topological unit of the weld bevel. After designing the location, type, and quantity of the monitoring nodes, complete tower welding sensing and monitoring node data is generated. Based on this data, a sensor monitoring integration device is then configured. This device includes a patch-type substrate surface temperature sensor, an ambient temperature sensor, an ambient wind speed sensor, and a welding arc voltage and current acquisition module. All sensors are configured according to the monitoring... After the monitoring nodes are fixedly installed and their parameters configured, the substrate temperature sensor is attached to the substrate surface of the corresponding topology unit, while the ambient temperature sensor and ambient wind speed sensor are fixed in the surrounding space of the corresponding topology unit. The welding arc voltage and current acquisition module is directly connected to the welding power supply and welding torch, and the acquisition frequency is matched with the welding travel speed. When the welding travel speed is 300 mm / min, the acquisition frequency is set to 10 Hz. All monitoring nodes start data acquisition synchronously with the welding operation, and the acquisition is continuous and synchronous throughout the entire process. The substrate temperature monitoring node acquires the initial substrate temperature and the real-time substrate temperature during the welding process, while the environmental parameter monitoring node acquires the ambient temperature, ambient wind speed, and welding arc characteristics throughout the welding process. The monitoring nodes collect arc voltage, arc current, and welding travel speed during the welding process. After collection, all data are classified and aggregated according to the unique spatial identifier of the topological unit bound to the corresponding monitoring node, generating tower welding substrate environmental monitoring data that includes tower welding spatial environment data and tower substrate welding arc monitoring characteristic data. The tower welding spatial environment data includes continuously collected data of ambient temperature and ambient wind speed throughout the welding operation. The tower substrate welding arc monitoring characteristic data includes synchronously collected data of substrate initial temperature, substrate real-time temperature during welding, welding arc voltage, arc current, and welding travel speed, providing basic monitoring data for subsequent spatial interpolation processing and equivalent transient heat input field analysis.

[0043] Furthermore, the environmental monitoring data of the tower welding substrate mentioned in step S142 includes tower welding space environmental data and tower substrate welding arc monitoring characteristic data.

[0044] Furthermore, step S2 includes the following steps:

[0045] Step S21: Perform three-dimensional heat conduction finite element discrete modeling on the low-temperature welding multi-dimensional sensing data to generate low-temperature welding heat conduction discrete mesh data;

[0046] In this embodiment of the invention, the tower welding topology bound by low-temperature welding multi-dimensional sensing data is used as the spatial framework. The tower weld bevel region, welding heat-affected zone, and tower substrate body region are used as modeling objects to construct a three-dimensional heat conduction analysis geometric model that corresponds 1:1 to the actual tower welding structure. The size parameters and material properties of the geometric model are completely consistent with the tower welding geometric data and material data in the low-temperature welding multi-dimensional sensing data. Then, the three-dimensional geometric model is meshed and discretized using a hexahedral structured mesh. During the meshing process, the weld pool region and the bevel fusion line region are used as the core densification areas, and the mesh size matches the minimum weld pool size. The forming dimensions and the mesh size of the weld heat-affected zone gradually transition away from the fusion line. The main body area of ​​the tower substrate uses a coarsened mesh. After the mesh is divided, the initial temperature of the substrate and the low-temperature thermophysical property parameters of the material in the low-temperature welding multi-dimensional sensing data are assigned to each mesh unit. The spatial coordinates, material properties, and initial thermal state of each mesh unit are defined, and the heat conduction connection relationship between adjacent mesh units is established. This generates low-temperature welding heat conduction discrete mesh data that covers the entire structure of the tower welding, maps to the topology, and is adapted to the needs of low-temperature heat conduction calculation. This provides a discretized computational carrier for subsequent low-temperature thermal energy boundary condition analysis and transient heat conduction calculation.

[0047] Step S22: Perform low-temperature thermal energy boundary condition analysis based on the low-temperature welding heat conduction discrete mesh data to generate low-temperature thermal energy boundary condition data;

[0048] In this embodiment of the invention, boundary elements are identified in the discrete mesh data of low-temperature welding heat conduction. Three types of boundary elements are identified across the entire model: the first type is the convective heat transfer boundary element between the outer wall of the tower and the exposed surface of the bevel; the second type is the convective heat transfer boundary element between the inner wall of the tower; and the third type is the internal heat conduction boundary element between the tower substrate and the welding filler material. Combined with the welding space environment data from the low-temperature welding multi-dimensional sensing data, boundary condition parameters are calculated and assigned for each type of boundary element. For the convective heat transfer boundary element, the convective heat transfer coefficient is calculated based on the ambient temperature and wind speed from the low-temperature welding multi-dimensional sensing data. The convective heat transfer coefficient of the boundary element on the windward side of the outer wall of the tower is the best match. The calculated values ​​corresponding to high wind speeds are used. The convective heat transfer coefficient of the boundary element on the leeward side is matched with the calculated wind speed at the corresponding location. The convective heat transfer coefficient of the boundary element on the inner wall of the tower is matched with the air convection parameters in the enclosed space. At the same time, the fluid medium temperature corresponding to the ambient temperature is assigned to all convective heat transfer boundary elements. For the internal heat conduction boundary elements, the thermal conductivity coefficient of the corresponding mesh element material is assigned. The boundary condition parameters are updated synchronously with the time step of the welding process and are consistent with the acquisition frequency of the low-temperature welding multi-dimensional sensing data. Low-temperature thermal energy boundary condition data covering all boundary elements, the entire welding time history, and adapting to the heat dissipation characteristics of the low-temperature environment are generated, providing clear boundary constraints for subsequent analysis of the thermal dynamic characteristics of the low-temperature tower.

[0049] Step S23: Perform dynamic characteristic analysis of thermal energy of the low-temperature tower based on the discrete mesh data of heat conduction in low-temperature welding and the dynamic boundary condition data of low-temperature thermal energy, and generate dynamic characteristic data of thermal energy of the low-temperature tower.

[0050] In this embodiment of the invention, Fourier's transient heat conduction law is used as the calculation basis, low-temperature welding heat conduction discrete grid data is used as the calculation carrier, low-temperature thermal energy dynamic boundary condition data is used as the boundary constraint, and the low-temperature thermophysical property parameters of the material corresponding to each grid cell are used as the calculation basis. The transient heat conduction numerical calculation of the entire welding process is carried out. The calculation time step is fixed at 0.1s, perfectly matching the welding travel speed of 300mm / min and the 10Hz acquisition frequency of the low-temperature welding multi-dimensional sensing data. The temperature field distribution calculation of the entire grid cell is completed within each time step. After the calculation is completed, thermal energy dynamic features are extracted from the temperature field calculation results of the entire welding time history. The extracted features include those of each grid cell. The study analyzed the peak welding temperature of the unit, the high-temperature residence time in the austenitizing temperature range, the cooling rate in the 800℃ to 500℃ range, the temperature gradient near the fusion line, the temperature diffusion range of the weld heat-affected zone, the temperature conduction rate in the thickness direction of the tower wall, and the difference in cooling rate between the windward and leeward sides of the tower. It also clarified the dynamic evolution characteristics of the molten pool, the range variation characteristics of the heat-affected zone, and the diffusion law of the base material temperature during the welding process. This generated dynamic thermal energy characteristic data of the low-temperature tower, covering the entire grid unit and the entire welding time history, and including the dynamic distribution of the temperature field and the characteristics of heat conduction, accurately reflecting the thermal energy evolution and heat conduction laws during the tower welding process in a low-temperature environment. This provides a thermal characteristic basis for subsequent equivalent heat input correction.

[0051] Step S24: Based on the welding arc monitoring characteristic data of the tower substrate corresponding to the low-temperature welding multi-dimensional sensing data, perform welding heat source characteristic analysis to generate tower welding heat source characteristic data;

[0052] In this embodiment of the invention, welding arc voltage, welding current, and welding travel speed data synchronously collected throughout the entire welding process are extracted from multi-dimensional sensing data of low-temperature welding. Based on the welding heat input calculation formula, the nominal heat input value of welding within each time step is calculated to clarify the dynamic fluctuation characteristics of the nominal heat input during the welding process. Then, based on the double ellipsoidal heat source model of gas metal arc welding, the spatial distribution characteristics of the welding heat source are analyzed. Combining the tower welding bevel form and cylinder wall thickness parameters, the core parameters of the double ellipsoidal heat source model, including the length of the first half-ellipsoid, the length of the second half-ellipsoid, the heat source width, and the heat source depth, are determined. Simultaneously, combined with welding arc monitoring characteristic data, the analysis is further performed. The distribution coefficients of arc energy in the four directions of front, back, left, and right of the molten pool are analyzed to clarify the spatial distribution law of welding heat source energy. Then, considering the influence of low temperature environment, the correspondence between arc voltage and current fluctuations and heat source energy effective utilization rate is analyzed. The dynamic fluctuation characteristics and effective utilization rate change characteristics of heat source energy during welding process are extracted. Tower welding heat source characteristic data including core parameters of welding heat source model, dynamic data of nominal heat input throughout the welding process, heat source energy spatial distribution coefficient, and heat source effective utilization rate characteristics are generated. This accurately reflects the energy input characteristics and spatial distribution law of welding arc, and provides core heat source parameters for subsequent equivalent transient heat input field analysis.

[0053] Step S25: Obtain prior data of equivalent heat input for tower welding;

[0054] In this embodiment of the invention, the prior data of equivalent heat input for tower welding comes from a low-temperature welding test database under the same brand of tower base material, the same type of welding filler material, the same bevel structure, and the same welding process. The data includes the correspondence between nominal heat input and actual weld penetration, weld width, and forming state under normal temperature and different low-temperature gradient environments; benchmark data of effective utilization rate of welding heat input under different ambient temperatures, different initial temperatures of base materials, and different ambient wind speeds; benchmark values ​​of heat loss correction coefficient for welding heat input under low-temperature environments; benchmark range of equivalent heat input corresponding to the achievement of weld low-temperature impact toughness and tensile strength standards under normal temperature environments; critical thresholds of heat input corresponding to cold cracks, incomplete penetration, and coarse grain defects under low-temperature environments; and the correspondence between equivalent heat input and weld microstructure and properties under different combinations of welding parameters. All prior data are completely matched with the tower welding structure, materials, and welding methods of this embodiment, providing a benchmark reference system for the establishment of subsequent equivalent heat input offset correction relationships, clarifying the difference benchmark of welding heat input under low-temperature and normal temperature environments, and ensuring that the correction process has clear experimental data support.

[0055] Step S26: Establish the correction relationship for equivalent heat input offset by using low-temperature welding multi-dimensional sensing data and tower welding equivalent heat input prior data, and generate the tower welding equivalent heat input correction model;

[0056] In this embodiment of the invention, based on prior data of equivalent heat input for tower welding, core offset influencing factors of equivalent heat input under low-temperature conditions are determined. These core offset influencing factors are all derived from multi-dimensional sensing data of low-temperature welding, including ambient temperature of the welding area, ambient wind speed, initial temperature of the tower substrate, fluctuations in welding arc voltage and current, and fluctuations in welding travel speed. For each core offset influencing factor, the quantitative mapping relationship between it and the effective heat input offset is analyzed. This clarifies the changes in effective heat input utilization rate and heat loss caused by decreased ambient temperature, increased ambient wind speed, and decreased initial substrate temperature, as well as the dynamic offset of heat input caused by fluctuations in arc parameters. After establishing the single-factor mapping relationship, multi-factor mapping is then conducted. Coupling effect analysis clarifies the superimposed impact of coupling effects between different influencing factors on heat input offset, and establishes a multi-dimensional coupling equivalent heat input correction relationship. The input of the correction relationship is the real-time monitoring parameters in the multi-dimensional sensing data of low-temperature welding, and the output is the welding equivalent heat input correction coefficient under the corresponding working condition. Based on this correction relationship, an equivalent heat input correction model for tower welding is constructed. The model can realize the real-time calculation of the heat input correction coefficient at each time step in the entire welding process. The generated tower welding equivalent heat input correction model can accurately quantify the impact of multi-factor coupling in the low-temperature environment on the offset of welding heat input, making up for the deficiency of traditional heat input calculation that does not consider the coupling effect of the low-temperature environment, and providing a correction carrier for the accurate analysis of the equivalent transient heat input field.

[0057] Step S27: Transmit the dynamic characteristic data of the thermal energy of the low-temperature tower and the characteristic data of the heat source of the tower welding to the equivalent heat input correction model of the tower welding to perform equivalent transient heat input field analysis of the tower welding, and generate equivalent transient heat input field data of the tower welding.

[0058] In this embodiment of the invention, the heat source model parameters, energy spatial distribution coefficient, and nominal heat input data for the entire time step from the heat source characteristic data of tower welding, along with the dynamic data of the temperature field of the entire grid cell and the heat conduction rate data from the dynamic characteristic data of thermal energy of the low-temperature tower, are synchronously input into the equivalent heat input correction model for tower welding. The correction model calculates the welding equivalent heat input correction coefficient for each time step and each grid cell, and then corrects the nominal heat input of the welding heat source in real time based on the correction coefficient to obtain the effective heat input value of each grid cell under low-temperature conditions. Combined with the transient heat conduction calculation results, an equivalent transient heat input field covering the entire welding time history and the entire welded structure range of the tower is constructed. Then, the distribution characteristics and dynamic evolution of this heat input field are analyzed. Feature analysis clarifies the effective heat input peak value in the core region of the molten pool, the heat input gradient distribution in the weld heat-affected zone, the heat input attenuation law along the circumference and wall thickness of the tower, the dynamic movement characteristics of the heat input field during welding, and the difference in effective heat input distribution between the windward and leeward sides of the tower. This generates equivalent transient heat input field data for tower welding, including effective heat input distribution data across the entire time step and all grid cells, dynamic evolution data of the heat input field, and heat input characteristic parameters of the molten pool and heat-affected zone. This accurately reflects the actual effective heat input distribution and evolution law of tower welding under low-temperature conditions, eliminates the heat input calculation bias caused by the low-temperature environment, and provides accurate thermal foundation data for subsequent weld microstructure prediction performance evaluation and weld structure risk characteristic analysis.

[0059] Furthermore, as an embodiment of the present invention, reference is made to... Figure 2 As shown, Figure 1 A detailed flowchart illustrating the implementation steps of step S3 is provided in this embodiment. Step S3 includes:

[0060] Step S31: Perform low-temperature phase transformation mapping path analysis on the weld seam using the equivalent transient heat input field data of the tower welding, and generate low-temperature phase transformation mapping path data for the weld seam.

[0061] In this embodiment of the invention, based on the dynamic temperature field data of the entire grid cell and the entire welding time step in the equivalent transient heat input field data of tower welding, complete welding thermal cycle curves corresponding to each spatial location of the weld fusion zone, welding heat-affected zone, and substrate zone are extracted. Core parameters are defined for each location during the welding process, including peak temperature, residence time in the austenitizing temperature range from Ac1 to Ac3, cooling rate in the range from 800℃ to 500℃, and cooling rate in the martensitic transformation critical temperature range from Ms to Mf. The austenitizing critical temperatures Ac1 and Ac3 are 730℃ and 910℃, respectively; the martensitic transformation initiation temperature Ms is 380℃ and the termination temperature Mf is 180℃. Combining these with the solid-state phase transformation critical temperature parameters of the tower substrate and welding filler material, and using the welding thermal cycle history as the time axis, the correspondence between temperature changes and solid-state phase transformation processes at each spatial location is established. The time-temperature correspondences for the austenitizing initiation and termination nodes, the ferrite transformation initiation node, the bainitic transformation range, and the martensitic transformation initiation and termination nodes are clarified. Simultaneously, addressing the core characteristics of rapid heat loss and significant undercooling during welding at -25℃, the offset of the critical phase transformation temperature under low-temperature conditions was corrected, clarifying the differences in phase transformation paths under different cooling rates. For example, in the windward, strongly convective heat dissipation area of ​​the tower, the cooling time from 800℃ to 500℃ is 6 seconds, with a cooling rate of 50℃ / s, and the phase transformation path is mainly martensitic transformation. In the leeward, weakly heat dissipation area, the cooling time from 800℃ to 500℃ is 12 seconds, with a cooling rate of 25℃ / s, and the phase transformation path is mainly bainitic-bonded iron. The raw material is transformed into the main body. Then, taking the center line of the weld groove as the reference, a three-dimensional mapping relationship of temperature-time-phase transformation process is established along the tower wall thickness direction, circumferential direction, and welding travel direction. This generates low-temperature phase transformation mapping path data of the weld that covers the entire weld space and includes the phase transformation type, phase transformation interval, phase transformation process, and cooling phase transformation characteristics of each location. This accurately reflects the intrinsic relationship between welding heat input and the solid-state phase transformation process of the weld under low-temperature environment, providing basic data of phase transformation process for subsequent weld microstructure analysis.

[0062] Step S32: Analyze the weld microstructure distribution characteristics of the low-temperature phase transformation mapping path data to generate weld microstructure distribution characteristic data;

[0063] In this embodiment of the invention, using the continuous cooling transformation curve of the tower substrate and the weld filler material as a benchmark, the phase transformation path at each spatial location in the low-temperature phase transformation mapping path data of the weld is matched with the continuous cooling transformation curve to determine the type of phase transformation products, the volume percentage of each phase transformation product, and the core microstructure parameters of grain size after welding cooling at each spatial location. For example, when the cooling rate is greater than 40℃ / s, the phase transformation product is mainly martensite, with a volume percentage of up to 60%. When the cooling rate is between 10℃ / s and 40℃ / s, the phase transformation product is mainly bainite. When the cooling rate is less than 10℃ / s, the phase transformation product is mainly ferrite + pearlite. Then, taking the center line of the weld pool as the core, the weld area is divided into six characteristic regions: weld fusion zone, fusion line, coarse-grained heat-affected zone, fine-grained heat-affected zone, incomplete phase transformation zone, and substrate region. Microstructure distribution features are extracted for each characteristic region to clarify the as-cast microstructure distribution features, columnar crystal growth direction, and average grain size of 120μm in the weld fusion zone. The peak temperature of 1350℃ in the coarse-grained heat-affected zone corresponds to an austenite grain size of 80μm and a martensitic hardened structure of 45% after cooling. The peak temperature of 950℃ in the fine-grained heat-affected zone corresponds to an equiaxed grain ratio of 90% and an average grain size of 10μm. The distribution characteristics of ferrite and pearlite in the incomplete phase transformation zone are also shown. In addition, considering the rapid cooling rate at low temperatures, the spatial distribution differences of the hardened structure along the tower wall thickness and on the windward and leeward sides are clearly defined. For example, the martensite ratio in the coarse-grained heat-affected zone on the windward side of the tower is 45%, while the martensite ratio in the same area on the leeward side is only 20%. The microstructure parameters of each spatial location are then bound to the three-dimensional spatial coordinates of the corresponding mesh unit to establish a continuous distribution model of microstructure characteristics across the entire weld space. This generates weld microstructure distribution characteristic data containing microstructure type, phase ratio, grain size, and microstructure uniformity parameters at each location in the entire weld space, comprehensively presenting the spatial distribution state of the microstructure of the tower weld under low-temperature conditions, and providing microstructure characteristic basis for subsequent mechanical property prediction.

[0064] Step S33: Analyze the trend characteristics of weld microstructure based on the low-temperature phase transformation mapping path data of the weld, and generate trend characteristic data of weld microstructure.

[0065] In this embodiment of the invention, based on the phase transformation dynamic process data of the full time step in the low-temperature phase transformation mapping path data of the weld, the dynamic characteristics of austenite grain growth during the welding heating stage, the characteristics of austenite homogenization process during the holding stage, and the dynamic transformation characteristics of solid phase transformation during the cooling stage are extracted along the time axis of the welding thermal cycle. For the heating stage, the influence of peak temperature and high-temperature dwell time on the austenite grain growth rate is analyzed, clarifying that when the high-temperature dwell time exceeds 5s, the austenite grain growth rate increases by 3 times, and when the peak temperature exceeds 1200℃, the austenite grain coarsening trend increases exponentially. For the cooling stage, the changes in phase transformation rate, hardened microstructure growth trend, and phase transformation completion rate under different cooling rates are analyzed, clarifying that for every 10℃ / s increase in cooling rate, the proportion of martensitic hardened microstructure increases by 12%. Then, using the weld bevel centerline as a reference, the continuous change trend of microstructure along the welding direction, the gradient change trend of microstructure along the tower wall thickness direction, and the trend along the tower... The study analyzes the microstructure differences between the windward and leeward sides in the circumferential direction. For multi-layer, multi-pass welding processes, it examines the tempering effect of a subsequent weld peak temperature of 650℃ on the preceding weld microstructure, the grain refinement trend of martensite decomposition into tempered sorbite, and the secondary grain coarsening trend caused by subsequent weld peak temperatures exceeding 1200℃. Combined with the core characteristic of uneven heat dissipation in low-temperature environments, the study clarifies that when the ambient wind speed increases from 3 m / s to 8 m / s, the weld cooling rate doubles, the proportion of hardened microstructure increases by 25%, and the microstructure stability changes due to fluctuations in welding heat input. This generates microstructure trend data including austenite grain growth trends, hardened microstructure growth trends, microstructure uniformity trends, secondary evolution trends of multi-layer weld microstructure, and spatial microstructure gradient trends. This data accurately predicts the evolution direction and potential degradation risks of tower weld microstructure in low-temperature environments, providing microstructure evolution trend support for subsequent mechanical property prediction.

[0066] Step S34: Analyze the distribution characteristics of weld microstructure based on the weld microstructure distribution characteristics data and weld microstructure trend characteristics data, and generate weld microstructure mechanical prediction performance distribution characteristics data.

[0067] In this embodiment of the invention, a quantitative mapping relationship between weld microstructure characteristics and mechanical properties is established. Based on the Hall-Page relationship, the correspondence between grain size and yield strength and tensile strength is clarified. That is, when the austenite grain size increases from 10 μm to 80 μm, the yield strength of the material decreases by 180 MPa and the tensile strength decreases by 120 MPa. Based on the proportion of hardened microstructure, the change law of material hardness and low-temperature impact toughness is clarified. That is, for every 10% increase in the proportion of martensitic hardened microstructure, the Brinell hardness of the weld increases by 25 HB, and the low-temperature impact toughness at -40℃ decreases by 15%. Based on the microstructure uniformity parameter, the change law of material hardness and low-temperature impact toughness is clarified. The correspondence between the material's elongation and reduction of area plasticity indices was determined. Then, the microstructure type, phase ratio, and grain size parameters at each spatial location in the weld microstructure distribution characteristic data were substituted into the quantization mapping relationship to calculate the core mechanical property parameters corresponding to each spatial location, including tensile strength, yield strength, Brinell hardness, -40℃ low-temperature impact toughness, and elongation. Furthermore, the grain growth trend, hardened microstructure growth trend, and microstructure uniformity change trend in the weld microstructure trend characteristic data were combined to correct the initially calculated mechanical property parameters, clarifying the decrease in low-temperature impact toughness caused by grain coarsening. The hardness increase reached up to 75%, with the hardening effect from the growth of the hardened structure reaching up to 80 HB. The plasticity index dispersion caused by the uneven structure reached up to 40%. Simultaneously, considering the service requirements of low-temperature tower welding, the cold cracking sensitivity index of different regions of the weld was calculated to clarify the crack sensitivity parameters under the combined effects of hardened structure, diffusible hydrogen content, and welding stress. For example, when the cold cracking sensitivity index Pcm is greater than 0.25%, the risk of cold crack initiation in the weld increases exponentially. Furthermore, the mechanical property parameters of each spatial location were bound to the three-dimensional spatial coordinates of the corresponding mesh element to establish a complete... A continuous distribution model of mechanical properties in the weld space is used to extract the gradient distribution characteristics of mechanical properties along the weld cross-section, wall thickness direction, and circumference direction. For example, the lowest low-temperature impact toughness of the coarse-grained heat-affected zone on the windward side of the tower is 12J at -40℃, while the low-temperature impact toughness of the same area on the leeward side is 48J. The model generates the distribution characteristics of the weld microstructure mechanical properties, including the strength, hardness, plasticity, low-temperature toughness, and crack sensitivity parameters at each location in the entire weld space. This comprehensively presents the spatial distribution of mechanical properties of the tower weld in low-temperature environment, providing a quantitative data basis for subsequent weld performance evaluation.

[0068] Step S35: The weld microstructure prediction performance data is processed by using the preset low-temperature welding performance evaluation index to evaluate the distribution characteristics of the weld microstructure mechanical performance, and weld microstructure prediction performance evaluation data is generated.

[0069] In this embodiment of the invention, the preset low-temperature welding performance evaluation index includes six core evaluation sub-items: the qualified range of tensile strength of tower weld, the minimum threshold of yield strength, the minimum limit of low-temperature impact toughness, the maximum limit of hardness, the maximum threshold of cold crack sensitivity index, and the minimum requirement of microstructure uniformity. Each evaluation sub-item is assigned a fixed weight coefficient and a qualification threshold. All thresholds and weights are matched to the service safety requirements of tower welding in low-temperature environments. First, the mechanical performance parameters of each spatial location in the weld microstructure mechanical prediction performance distribution characteristic data are compared one by one with the corresponding sub-item threshold of the low-temperature welding performance evaluation index to calculate the compliance coefficient of each sub-item. Then, the comprehensive performance evaluation score of each spatial location is calculated by combining the sub-item weight coefficients to clarify the performance compliance status, performance margin, and performance defects of each location. Based on the weld seam characteristic areas, a regional overall performance assessment is conducted on the weld fusion zone, heat-affected zone, and substrate area. This identifies the spatial location, size, and core performance weaknesses of areas that fail to meet performance standards. Simultaneously, focusing on the core requirements of low-temperature environments, the assessment prioritizes the low-temperature impact toughness compliance and cold crack sensitivity index control of the entire weld seam area. This clarifies the overall performance uniformity and stability of the weld seam, generating weld microstructure prediction performance assessment data that includes the performance compliance status of each location in the entire weld seam space, comprehensive evaluation score, regional performance assessment results, weak area location, core performance defect type, and overall weld seam performance level. This accurately quantifies the performance compliance and quality level of tower weld seams under low-temperature environments, providing clear performance guidance and optimization basis for subsequent welded structure risk characteristic analysis and welding parameter optimization.

[0070] Furthermore, step S31 includes the following steps:

[0071] Step S311: Perform tower welding cooling characteristic analysis on the equivalent transient heat input field data of tower welding to generate tower welding cooling characteristic data;

[0072] In this embodiment of the invention, the transient temperature field distribution data of the entire grid cell and the entire welding time step in the equivalent transient heat input field data are first retrieved. Complete welding thermal cycle curves corresponding to each spatial location in the weld fusion zone, weld heat-affected zone, and substrate zone are extracted. The thermal cycle is decomposed into three continuous segments: heating, peak temperature holding, and cooling. Focusing on the cooling segment, which has the most significant impact on the microstructure under low-temperature conditions, core parameters are extracted for each location, including the welding peak temperature, high-temperature residence time in the austenitizing temperature range, cooling time in the 800℃ to 500℃ range, cooling rate in the martensitic transformation critical temperature range, temperature gradient during cooling, and total cooling time from peak temperature to room temperature. Combining the inherent characteristics of strong convection heat dissipation in low-temperature welding, this study simultaneously analyzes the difference in cooling rates between the windward and leeward sides of the tower, the changes in cooling gradients along the tower wall thickness, and the tempering intervention characteristics of subsequent weld thermal cycles on the cooling process of the preceding weld during multi-layer, multi-pass welding. By binding all extracted cooling characteristic parameters with their corresponding three-dimensional spatial coordinates, a continuous distribution model of cooling characteristics covering the entire weld space is constructed. The resulting tower welding cooling characteristic data fully covers the entire weld space and the entire welding time history, accurately reproducing the dynamic changes and spatial distribution differences of the tower welding cooling process under low-temperature conditions. This provides core cooling process data support for subsequent weld microstructure evolution characteristic analysis.

[0073] Step S312: Analyze the weld microstructure evolution characteristics based on the tower welding cooling characteristic data and the equivalent transient heat input field data of tower welding, and generate weld microstructure evolution characteristic data;

[0074] In this embodiment of the invention, based on the peak temperature and high-temperature residence time in the equivalent transient heat input field data, the microstructure changes during the heating stage are analyzed to clarify the austenite nucleation initiation and termination temperatures, and to define the influence of peak temperature and high-temperature residence time on the austenite grain growth rate and grain homogenization. The austenite grain size and homogenization characteristics at each spatial location after heating are determined. Then, using the cooling rate and phase transformation interval cooling time in the cooling characteristic data obtained in the previous step as the core, the continuous transformation process of undercooled austenite during the cooling stage is analyzed. The dynamic changes in the nucleation initiation and termination temperatures, transformation rates, and phase transformation volume ratios of proeutectoid ferrite, pearlite, bainite, and martensite are clarified. This addresses the issues of rapid cooling and undercooling in low-temperature environments. The core characteristic of high density is the analysis of the nucleation and growth of hardened microstructures under high cooling rates. Combined with the differences in thermal cycling at different locations in the weld, the microstructure evolution differences in the weld fusion zone, coarse-grained heat-affected zone, fine-grained heat-affected zone, and incomplete phase transformation zone are analyzed. Simultaneously, the secondary evolution process of microstructure tempering, grain refinement, or coarsening caused by secondary thermal cycling during multi-layer, multi-pass welding is tracked. The microstructure evolution parameters of each spatial location throughout the entire welding cycle are bound to the corresponding three-dimensional spatial coordinates, forming weld microstructure evolution characteristic data that fully covers the entire weld space and welding time history. This clearly reconstructs the full-cycle evolution law of tower weld microstructure from heating to cooling under low-temperature conditions, providing core microstructure evolution basis for subsequent low-temperature phase transformation mapping path analysis of the weld.

[0075] Step S313: Perform low-temperature phase transformation mapping path analysis on the weld microstructure evolution characteristic data to generate low-temperature phase transformation mapping path data for the weld.

[0076] In this embodiment of the invention, based on the dynamic evolution data of microstructure at all spatial locations and throughout the entire time step, and combined with the continuous cooling transformation curves of the tower substrate and welding filler material, a corresponding mapping relationship is established for temperature change, time progression, and solid-state phase transformation process at each spatial location, using the time axis of the welding thermal cycle as a reference. Following the sequence of the welding thermal cycle, the entire phase transformation process is decomposed into four continuous phase transformation segments: austenitization, proeutectoid transformation, diffusion-type transformation, and diffusionless transformation. The start and end points of each segment are clearly defined, along with the corresponding core parameters of temperature, cooling rate, and microstructure transformation. Specifically, the austenitization segment corresponds to the complete transformation of the base metal microstructure to austenite during the heating process; the proeutectoid transformation segment corresponds to the precipitation of proeutectoid ferrite during the cooling process; the diffusion-type transformation segment corresponds to the nucleation and growth of pearlite and bainite; and the diffusionless transformation segment corresponds to the transformation of martensite. This approach is specifically designed for welding in low-temperature environments. To address the issue of phase transformation critical temperature deviation caused by rapid cooling rate and large supercooling, this paper corrects the phase transformation path deviation characteristics under different cooling rates. Then, using the weld groove centerline as a reference, a three-dimensional mapping relationship of temperature-time-phase transformation products is established in the entire three-dimensional space of the weld along the tower wall thickness direction, circumferential direction, and welding travel direction. The complete phase transformation path of each spatial location is clarified, including phase transformation type, phase transformation range, transformation rate, proportion of phase transformation products, and spatial phase transformation gradient characteristics. At the same time, the differences in phase transformation paths between the windward and leeward sides of the tower and between the weld surface and root are clarified. The resulting low-temperature phase transformation mapping path data of the weld completely covers the entire weld space, including the complete phase transformation process, phase transformation characteristic parameters, and spatial phase transformation distribution law at each location. It accurately reveals the intrinsic relationship between welding heat input and the solid-state phase transformation process of the weld under low-temperature environment, providing core phase transformation process data support for subsequent weld microstructure distribution characteristic analysis and mechanical property prediction.

[0077] Furthermore, step S4 includes the following steps:

[0078] Step S41: Analyze the characteristics of the predicted performance of weld microstructure based on the predicted performance evaluation data, and generate the predicted performance characteristics data of weld microstructure.

[0079] In this embodiment of the invention, the weld microstructure prediction performance evaluation data is decomposed in all dimensions. This data includes the mechanical performance parameters, performance compliance status, comprehensive evaluation score, weak area location, and core performance defect type of the weld in all spatial locations. Taking the weld groove centerline as the reference, an analysis unit is divided every 10 mm along the welding direction, and the tower wall thickness direction is divided into three layers of analysis units: outer wall, substrate middle layer, and inner wall. An analysis unit is also divided every 15° in the circumferential direction, forming a continuous three-dimensional analysis unit array for the entire weld. Each analysis unit corresponds to a unique three-dimensional spatial coordinate. For each analysis unit, the corresponding tensile strength, yield strength, -40℃ low-temperature impact toughness, Brinell hardness, and cold crack sensitivity index Pcm core performance parameters are extracted, as well as the performance compliance status, the difference from the qualified threshold, and the performance margin. Then, the weld fusion zone, coarse-grained heat-affected zone, fine-grained heat-affected zone, incomplete phase transformation zone, and substrate zone are used as characteristic partitions for regional analysis. The system extracts performance state features, clarifying the performance mean, performance dispersion, performance bottleneck type, and weak point distribution of each feature partition. Simultaneously, considering the core characteristics of low-temperature welding, it focuses on extracting performance differences between the windward and leeward sides of the tower, performance gradient characteristics between the weld surface and root, interlayer performance fluctuation characteristics of multi-layer, multi-pass welds, and performance degradation characteristics of areas with concentrated hardened structures. All extracted performance state features are then bound to the three-dimensional spatial coordinates of the corresponding analysis units to establish a continuous distribution model of performance state across the entire weld space. The generated weld microstructure prediction performance state feature data comprehensively covers the performance parameter distribution, weak point location, partitioned performance characteristics, and performance degradation patterns across the entire weld space. This accurately reconstructs the overall performance state of the tower weld under low-temperature conditions, clarifying both the overall performance compliance of the weld and locating the specific location and severity of local performance defects. This provides precise material property boundaries and analysis targets for subsequent analysis of residual stress characteristics in tower welding.

[0080] Step S42: Perform residual stress characteristic analysis on the predicted performance state characteristic data of weld microstructure to generate residual stress characteristic data of tower welding;

[0081] In this embodiment of the invention, the generation and distribution of welding residual stress are directly determined by the welding thermal cycle process, the solid-state phase transformation characteristics of the material, and the distribution of weld mechanical properties. Under low-temperature conditions, the rapid cooling rate and large temperature gradient lead to a more significant residual stress concentration effect. During the analysis, a three-dimensional geometric model of the tower weld is used as the computational carrier. The material elastic modulus, yield strength, Poisson's ratio, coefficient of thermal expansion, and phase transformation volume change rate parameters corresponding to each analytical unit in the weld microstructure prediction performance state characteristic data are assigned to the corresponding computational units in the model. This clarifies the material mechanical property boundaries of each unit. Combined with the welding thermal cycle temperature history and weld low-temperature phase transformation mapping path data obtained earlier, and based on the thermo-elastic-plastic deformation theory, numerical calculations of residual stress throughout the entire welding process are performed. During the calculation, the superposition effects of thermal expansion and contraction deformation caused by the welding temperature gradient and volume changes caused by solid-state phase transformation on residual stress are considered simultaneously. For the strong heat dissipation characteristics under low-temperature conditions, the windward orientation of the tower is specifically corrected. The effects of non-uniform deformation caused by high cooling rates were investigated. After calculation, the distribution data of longitudinal residual stress, transverse residual stress, and residual stress in the wall thickness direction at each location in the entire weld space were extracted. The peak value of residual stress, stress concentration factor, stress gradient change, and spatial distribution range of high stress area were identified. Combined with the weld microstructure prediction performance status characteristic data, the spatial overlap between residual stress concentration area and performance weak area was analyzed to clarify the amplification effect of performance degradation on stress concentration. The residual stress characteristics of the weld fusion line, weld toe position, bevel root, and high cooling area on the windward side of the tower were extracted. All residual stress characteristic parameters were bound to the three-dimensional spatial coordinates of the corresponding locations. The generated tower welding residual stress characteristic data completely covered the residual stress distribution, stress peak value, stress concentration area, and stress gradient characteristics of the entire weld space. It accurately restored the generation law and spatial distribution state of tower welding residual stress under low temperature environment, providing core stress boundary data for subsequent tower welding structural risk characteristic analysis.

[0082] Step S43: Perform a risk characteristic analysis on the residual stress characteristic data of the tower welding structure to generate risk characteristic data of the tower welding structure.

[0083] In this embodiment of the invention, the core failure risks of the tower welded structure are concentrated in three categories: cold crack initiation, brittle fracture, and fatigue failure during service. The magnitude and distribution of residual stress are the core factors that determine the risk level. Prior data on low-temperature welding material fracture under the same type of tower substrate and welding process were retrieved. This data includes the core parameters of the material plane strain fracture toughness threshold of 120MPa・m^1 / 2 at -40℃, the critical stress for cold crack initiation of 320MPa, the ductile-brittle transition temperature of -40℃, and the fatigue strength limit of 180MPa under alternating load. All parameters have been verified by the -25℃ low-temperature welding process test and are completely matched with the current tower welding conditions. Then, combining the residual stress characteristic data of tower welding, we first carried out the weld crack sensitivity characteristic analysis. The peak value of residual tensile stress at each location was compared with the critical stress for cold crack initiation of the material (320 MPa). The ratio of residual stress to material yield strength (345 MPa) was calculated. Combined with the cold crack sensitivity index and hardened structure proportion in the weld microstructure prediction performance state characteristic data, the cold crack sensitivity level of each location was determined. For example, the peak value of residual tensile stress in a certain area reached 480 MPa, exceeding the critical stress for cold crack initiation by 50%. The cold crack sensitivity index Pcm reached 0.28%, exceeding the highest threshold of 0.25%. The martensite volume proportion reached 45%, which was directly judged as a high sensitivity level. Furthermore, a structural risk level classification was conducted for the entire weld space. Using residual stress concentration, performance degradation, and crack sensitivity as core indicators, the weld area was divided into three fixed levels: high-risk, medium-risk, and low-risk. The spatial location, distribution range, and core risk factors of each risk level area were clearly defined. Simultaneously, considering the actual service conditions of the tower, the superposition effect of welding residual stress with the tower's axial load and wind-induced alternating load was analyzed. The fatigue failure risk and brittle fracture risk at low temperatures in high-stress areas during service were identified, with a focus on the residual tensile stress concentration area at the weld toe, the root of the bevel, and... A special risk characteristic analysis was conducted on the hardened structure and high-stress superposition zone to clarify the specific types, severity, and triggering conditions of the risks. Then, all risk characteristic parameters were bound to the three-dimensional spatial coordinates of the corresponding locations to establish a structural risk distribution model for the entire weld space. The generated risk characteristic data of the tower welded structure fully covers the risk level distribution, high-risk area location, risk type, risk cause, and risk severity of the entire weld space. This accurately predicts the potential failure risk of the tower welded structure under low-temperature conditions, providing a clear risk avoidance direction and optimization target for subsequent welding parameter optimization needs analysis and multi-objective optimization strategy formulation.

[0084] Furthermore, step S43 includes the following steps:

[0085] Step S431: Obtain prior data on fracture of welding materials for cryogenic towers;

[0086] In this embodiment of the invention, prior fracture data of low-temperature tower welding materials are obtained through a special experimental database of low-temperature mechanical properties and fracture behavior. The data covers the entire low-temperature gradient range from -40℃ to 0℃, including the plane strain fracture toughness threshold, critical stress for cold crack initiation, ductile-brittle transition temperature, and fatigue strength limit under alternating loads at different temperatures. It also includes the material crack sensitivity variation law corresponding to different hardened microstructure proportions, the influence coefficient of weld diffuseable hydrogen content on the critical stress for cold cracking, and the critical stress and microstructure matching parameters corresponding to historical cracking cases of similar tower welding structures. All data have been verified by welding process tests under low-temperature conditions and are fully matched with the material system, structural form, and working environment of this tower welding. This provides a unified and feasible benchmark threshold and reference system for subsequent weld crack sensitivity characteristic analysis and structural risk level determination, avoiding misjudgments caused by benchmark deviations during risk analysis.

[0087] Step S432: Perform weld crack sensitivity feature analysis using prior fracture data of low temperature tower welding materials and weld microstructure prediction performance state characteristic data to generate weld crack sensitivity feature data.

[0088] In this embodiment of the invention, the critical stress for cold crack initiation, the critical proportion of hardened structure, and the minimum fracture toughness in the prior fracture data of low-temperature tower welding materials are used as the core judgment criteria. Then, the predicted performance characteristics data of the weld microstructure are spatially decomposed. Using the weld groove centerline as a reference, the entire weld is divided into continuous analysis units along the tower wall thickness direction, circumferential direction, and welding direction. Each analysis unit corresponds to a unique three-dimensional spatial coordinate. For each analysis unit, the corresponding core parameters such as the volume proportion of hardened structure, weld metal hardness, low-temperature impact toughness, cold crack sensitivity index, and material yield strength are extracted and compared with the benchmark thresholds of the prior fracture data. The crack sensitivity coefficient of each analysis unit is calculated to clarify the material's inherent influence on crack initiation. The analysis focuses on the sensitivity of welds to cracks. Combined with the inherent characteristics of low-temperature welding, a detailed analysis is conducted on four high-incidence crack areas: the weld fusion zone, the coarse-grained heat-affected zone, the weld toe, and the root of the groove. The analysis emphasizes the amplifying effect of the decrease in toughness caused by austenite grain growth in the coarse-grained zone and the increased proportion of hardened microstructure on crack sensitivity. Simultaneously, the analysis corrects for the increase in sensitivity coefficient caused by the ductile-brittle transition of materials under low-temperature conditions. The crack sensitivity coefficient, sensitivity level, and core sensitivity triggers of each analysis unit are bound to the corresponding three-dimensional spatial coordinates to generate weld crack sensitivity characteristic data covering the entire weld space. This fully presents the distribution pattern of crack sensitivity at different locations in the weld, accurately locates highly sensitive areas, and provides core material sensitivity boundaries and analysis targets for subsequent risk characteristic analysis of tower welded structures.

[0089] Step S433: Analyze the risk characteristics of the tower welded structure by using the weld crack sensitive feature data to analyze the residual stress feature data of the tower welded structure, and generate the risk feature data of the tower welded structure.

[0090] In this embodiment of the invention, the failure risk of the tower welded structure is essentially the result of the superposition of material crack sensitivity and welding residual stress. At low temperatures, material toughness decreases and the residual stress concentration effect intensifies; the superposition of these two factors significantly increases the probability of structural failure. Weld crack sensitivity characteristic data and tower weld residual stress characteristic data are matched according to three-dimensional spatial coordinates to ensure that the crack sensitivity parameters and residual stress parameters of each analysis unit correspond completely. Then, using the fracture critical stress and fatigue strength limit in the prior data of low-temperature tower welded material fracture as a benchmark, the ratio of the peak residual tensile stress to the critical stress for cold crack initiation in each analysis unit is calculated. When the ratio is greater than 1, it is determined that the residual stress in that area has exceeded the material's crack resistance critical value, and combined with the crack sensitivity level, it is directly classified as a high-risk area. Subsequently, multi-dimensional risk feature extraction is carried out for the entire weld. On the one hand, combined with the actual service conditions of the tower, the superposition effect of welding residual stress and tower axial load and wind-induced alternating load is analyzed to clarify the fatigue failure of high-stress areas during service. This study focuses on two main areas: effective risk and brittle fracture risk under low-temperature conditions. On the other hand, it emphasizes the spatial overlap between highly sensitive crack areas and areas with high residual stress concentration. For areas with the highest overlap, such as the weld toe, bevel root, and coarse-grained heat-affected zone on the windward side of the tower, specific risk tracing is conducted to clarify the specific types, severity, and triggering conditions of the risks. Using residual stress level, crack sensitivity level, and load superposition effect as core judgment indicators, the entire weld is divided into three fixed levels: high-risk, medium-risk, and low-risk. The spatial location, distribution range, risk type, core causes, and severity of each level are bound to corresponding three-dimensional spatial coordinates to establish a structural risk distribution model for the entire weld space. The generated risk characteristic data of the tower welded structure fully covers the risk level distribution of the entire weld, accurately locates high-risk areas, and predicts failure modes. This accurately reconstructs the full picture of potential failure risks of the tower welded structure under low-temperature conditions, providing clear risk avoidance directions and optimization targets for subsequent welding parameter optimization needs analysis and multi-objective optimization strategy formulation.

[0091] Furthermore, step S5 includes the following steps:

[0092] Step S51: Based on the risk characteristic data of the tower welding structure, perform tower welding parameter optimization requirement analysis and generate tower welding parameter optimization requirement data;

[0093] In this embodiment of the invention, the risk characteristic data of the tower welded structure is decomposed in all dimensions. The spatial location, distribution range, core risk type, risk triggering factors, and risk severity of high-risk and medium-risk areas within the entire weld space are extracted. Simultaneously, the data on weak performance areas, performance deficiency types, and differences from the acceptable threshold in the weld microstructure prediction performance evaluation data are combined. Using risk level as the core priority criterion, priority levels for optimization targets are divided. High-risk areas for cold cracking and areas with substandard low-temperature impact toughness are designated as first-priority optimization targets; medium-risk areas with residual stress concentration and fatigue risk areas caused by poor weld formation are designated as second-priority optimization targets; and areas with excessively low performance margins and insufficient uniformity are designated as third-priority optimization targets. Furthermore, for each priority optimization target, a correlation mapping analysis between risk root causes and welding parameters is conducted. It is clarified that the core trigger for high-risk cold cracking on the windward side of the tower is excessively rapid cooling due to strong convection heat dissipation and insufficient effective heat input; the core trigger for brittle fracture risk in the coarse-grained heat-affected zone is high-temperature dwell time. Excessive interpass length leads to excessive austenite grain growth. The core causes of incomplete penetration at the bevel root are insufficient arc penetration and excessively fast welding speed. The core causes of fatigue failure at the weld toe are poor weld formation and excessively high residual tensile stress peak. For each type of core cause, corresponding feasible welding parameter optimization directions are decomposed. For example, for high-risk targets of cold cracking, the optimization direction is to increase the lower limit of preheating temperature, control the upper limit of welding speed, and match an appropriate welding current and voltage combination to ensure that the t8 / 5 cooling time meets the requirements for low-temperature welding crack resistance. For targets with insufficient toughness in the coarse-grained region, the optimization direction is to control the upper limit of welding heat input, shorten the high-temperature dwell time, and match a reasonable interpass temperature range. All optimization targets, priority levels, core problem roots, parameter optimization directions, and optimization boundary requirements are integrated to generate tower welding parameter optimization requirement data. This accurately clarifies the core targets, priorities, adjustment directions, and constraint boundaries for tower welding parameter optimization under low-temperature conditions, providing a clear demand guide and optimization framework for subsequent multi-objective optimization feature analysis.

[0094] Step S52: Perform multi-objective optimization feature analysis on tower welding quality using weld microstructure prediction performance evaluation data and tower welding parameter optimization requirement data to generate multi-objective optimization feature data for tower welding quality;

[0095] In this embodiment of the invention, guided by the optimization requirements data for tower welding parameters and combined with the qualification threshold requirements in the weld microstructure prediction performance evaluation data, three core dimensions of multi-objective optimization are determined: optimization variables, optimization objectives, and constraints. For the optimization variables dimension, welding current, arc voltage, welding travel speed, preheating temperature, interpass temperature, and shielding gas flow rate are identified as core optimization variables. A fixed value range is set for each variable, with the range boundaries matching welding process specifications and on-site operation requirements. For the optimization objectives dimension, five core optimization objectives are set: the first objective is that the low-temperature impact toughness of the entire weld area meets the minimum limit requirement; the second objective is that the hardness of the weld and heat-affected zone does not exceed the maximum limit; the third objective is that the cold cracking sensitivity index of the entire weld area is lower than the critical threshold; the fourth objective is that the peak value of welding residual tensile stress is lower than a fixed proportion of the material yield strength; and the fifth objective is that the welding operation efficiency is not lower than the baseline operation requirements. Simultaneously, based on the priority hierarchy in the optimization requirements data, each optimization objective is further defined. Fixed weight coefficients are assigned to the optimization objectives. The crack resistance and toughness objectives corresponding to the first priority optimization targets have the highest weights, followed by stress control objectives, forming quality objectives, and operational efficiency objectives. In terms of constraints, the matching specifications for welding parameters, the parameter boundaries for stable arc combustion, operational safety requirements in low-temperature environments, and the upper and lower limits of interpass temperature for multi-layer, multi-pass welding are clearly defined. For each optimization variable, the single-factor influence on each optimization objective and the coupling effects between multiple variables are analyzed. For example, the coupling effect of welding current and welding speed on effective heat input, and the coupling effect of preheating temperature and ambient wind speed on cooling rate are analyzed. The core features of multi-objective optimization are extracted, including the value range of optimization variables, the threshold and weight of optimization objectives, the mapping relationship between variables and objectives, and the multi-objective coupling constraints. These are integrated to generate multi-objective optimization feature data for tower welding quality, providing a clear optimization framework, calculation boundaries, and judgment criteria for subsequent parameter iterative simulation optimization.

[0096] Step S53: Perform iterative simulation optimization of tower welding parameters based on the multi-objective optimization feature data of tower welding quality to generate iterative optimization data of tower welding parameters;

[0097] In this embodiment of the invention, the optimization variables in the multi-objective optimization feature data of tower welding quality are used as inputs, the optimization objective is used as the output, and the constraints are used as the calculation boundaries. A multi-objective particle swarm optimization algorithm is employed to perform iterative calculations throughout the entire process. The core carriers of the iterative calculations are the three-dimensional heat conduction finite element model of tower welding, the weld microstructure phase transformation prediction model, and the residual stress calculation model established earlier, ensuring that the iterative process is fully matched with the preceding thermal field analysis, performance prediction, and risk identification processes. In each iteration, multiple sets of welding parameter combinations that conform to the variable value ranges are randomly generated, and each set of parameter combinations is sequentially input into three... A finite element model of heat conduction was used to calculate the equivalent transient heat input field data of the tower welding under the given parameter combination. Based on the heat input field data, the following were calculated sequentially: weld low-temperature phase transformation mapping path data, weld microstructure distribution characteristics data, weld microstructure mechanical prediction performance distribution characteristics data, and tower welding residual stress characteristics data. Each calculated result was compared with the threshold of the optimization objective, and the multi-objective comprehensive fitness value of each parameter combination was calculated. A higher fitness value indicates a better overall optimization effect of the parameter combination. During the iteration process, the parameter combination with the optimal comprehensive fitness value was retained in each round, and combinations completely inconsistent with the target were excluded. Parameter combinations that do not meet the constraints are directly eliminated. Simultaneously, the values ​​of optimization variables are continuously adjusted based on the optimal combination to gradually approach the global optimum. The iteration terminates when the change in the optimal fitness value is lower than a fixed threshold after 30 consecutive iterations, or when the preset maximum number of iterations is reached. For the first-priority high-risk cold cracking optimization target, the preheating temperature is gradually increased and the welding speed is decreased during the iteration process. At the same time, the upper limit of the welding current is strictly controlled to avoid excessive growth of austenite grains in the coarse-grained region, precisely balancing the weld's crack resistance and low-temperature impact toughness. For target points where the toughness in the coarse-grained region is insufficient, the welding current and welding parameters are iteratively optimized. By matching the welding speed, the high-temperature dwell time is controlled within the critical range for austenite grain growth, while a reasonable interpass temperature is matched to achieve weld microstructure refinement. After iteration, three sets of welding parameter combinations with the best multi-objective comprehensive performance are selected. The weld performance prediction results, residual stress control effect, risk level reduction degree, and operation efficiency corresponding to each set of parameters are clarified. The tower welding parameter iterative optimization data are integrated and generated, including the optimal parameter combination, corresponding performance and risk prediction results, and optimization effect verification data. This provides core parameter basis and execution benchmark for the subsequent design of intelligent control strategy for welding parameter optimization.

[0098] Step S54: Design an intelligent control strategy for optimizing tower welding parameters by using iterative optimization data of tower welding parameters, and generate a tower welding parameter optimization strategy.

[0099] In this embodiment of the invention, the optimal welding parameter combination in the iterative optimization data of tower welding parameters is used as the core benchmark. Combined with the structural characteristics of tower welding, the multi-layer, multi-pass welding process, the dynamic characteristics of the low-temperature environment, and the tower welding topology and monitoring node layout scheme established earlier, a regional, process-specific, and full-process closed-loop intelligent control strategy is designed to ensure that the optimized parameters are implementable and executable, adapting to the on-site operation scenario of low-temperature tower welding. First, a regional precise matching control strategy is designed, matching the corresponding optimal parameters to the different heat dissipation conditions of the windward side, leeward side, top, and bottom of the tower circumference, as well as the different performance requirements of the weld fusion zone, heat-affected zone, and substrate area. Parameter combinations, such as those used in the windward, highly convective heat dissipation area of ​​the tower, employ higher preheating temperatures and lower welding travel speeds to compensate for heat input losses and ensure that the cooling rate remains stable within the optimal range. On the leeward side, where heat dissipation conditions are relatively mild, benchmark parameters balancing performance and operational efficiency are used to achieve uniform weld performance and consistent quality across the entire circumference of the tower. Secondly, specialized control strategies are designed for each process. For the three core processes of multi-layer, multi-pass welding—root pass, fill pass, and cap pass—differentiated optimized parameters are matched. For the root pass, the focus is on controlling the arc penetration and root fusion, matching a higher welding current and a lower welding travel speed to prevent incomplete penetration defects at the source. For the fill pass, the focus is on controlling interpass temperature and heat input. Precise matching is achieved through parameter matching in multi-pass thin-layer welding, refining weld grains and controlling residual welding stress. For the cap weld, the focus is on controlling weld formation and toe transition, matching appropriate welding speed and amplitude to reduce stress concentration at the toe and improve weld fatigue performance. Furthermore, a dynamic environmental closed-loop control strategy is designed, combining the previously mentioned sensor monitoring integration equipment to collect real-time data on ambient temperature, wind speed, substrate temperature, and welding arc parameters during the welding process. When the ambient wind speed increases or the substrate temperature decreases beyond the preset fluctuation range, the welding current and welding speed are automatically adjusted to compensate for heat input loss in real time, ensuring that the effective heat input during the welding process remains stable within the optimal range and avoiding low... The dynamic changes in temperature environment cause fluctuations in weld quality. By integrating the control logic, parameter matching rules, dynamic adjustment trigger conditions, and abnormal operating condition response measures of the entire process, an optimization strategy for tower welding parameters is generated. This strategy covers the entire process, including preheating parameter control before welding, precise matching of parameters by region and process during welding, closed-loop parameter adjustment under dynamic conditions, and interlayer temperature control. This enables intelligent and precise control of tower welding parameters in low-temperature environments, reducing welding structural risks at the source, ensuring that the performance of the entire weld area meets the standards, and taking into account welding efficiency. It perfectly connects with the entire process of data monitoring, thermal field analysis, performance prediction, and risk identification mentioned above, forming a complete closed loop for the quality control of low-temperature tower welding.

[0100] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0101] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for optimizing tower section welding parameters based on cryogenic environmental conditions, characterized in that, The low-temperature environment is a low-temperature working condition environment of not more than 0 DEG C for a tower drum welding scene, and includes the following steps: Step S1: acquiring tower drum welding basic data; performing multi-dimensional perception monitoring processing on the low-temperature environment of tower drum welding through sensor monitoring integrated equipment and tower drum welding basic data, and generating low-temperature welding multi-dimensional perception data; Step S2: performing tower drum welding equivalent transient heat input field analysis based on the low-temperature welding multi-dimensional perception data, and generating tower drum welding equivalent transient heat input field data; Step S3: performing weld structure prediction performance evaluation processing based on the tower drum welding equivalent transient heat input field data, and generating weld structure prediction performance evaluation data; Step S3 includes the following steps: Step S31: performing weld low-temperature phase change mapping path analysis through the tower drum welding equivalent transient heat input field data, and generating weld low-temperature phase change mapping path data; Step S32: performing weld structure distribution feature analysis on the weld low-temperature phase change mapping path data, and generating weld structure distribution feature data; Step S33: performing weld structure structure trend feature analysis according to the weld low-temperature phase change mapping path data, and generating weld structure structure trend feature data; Step S34: performing weld structure mechanical prediction performance distribution feature analysis according to the weld structure distribution feature data and the weld structure structure trend feature data, and generating weld structure mechanical prediction performance distribution feature data; Step S35: performing weld structure prediction performance evaluation processing on the weld structure mechanical prediction performance distribution feature data through a preset low-temperature welding performance evaluation index, and generating weld structure prediction performance evaluation data; Step S4: performing tower drum welding structure risk feature analysis through the weld structure prediction performance evaluation data, and generating tower drum welding structure risk feature data; Step S5: performing intelligent control strategy design for tower drum welding parameter optimization based on the weld structure prediction performance evaluation data and the tower drum welding structure risk feature data, and generating a tower drum welding parameter optimization strategy.

2. The cryogenic environment-based tower weld parameter optimization method of claim 1, wherein, Step S1 includes the following steps: Step S11: acquiring tower drum welding basic data, wherein the tower drum welding basic data includes tower drum welding geometric data and tower drum welding material data; Step S12: performing basic data global feature analysis on the tower drum welding basic data, and generating tower drum welding global basic feature data; Step S13: performing tower drum welding geometric space topology analysis according to the tower drum welding global basic feature data, and generating tower drum welding topology structure data; Step S14: performing tower drum welding base material environment monitoring processing based on the sensor monitoring integrated equipment and the tower drum welding topology structure data, and generating tower drum welding base material environment monitoring data; Step S15: transmitting the tower drum welding base material environment monitoring data to the tower drum welding topology structure data for spatial interpolation processing of base material environment monitoring, and generating low-temperature welding multi-dimensional perception data.

3. The cryogenic environment-based tower weld parameter optimization method of claim 2, wherein, Step S14 includes the following steps: Step S141: performing tower drum welding topology structure attribute feature analysis on the tower drum welding topology structure data, and generating tower drum welding topology structure attribute data; Step S142: Design tower welding sensing and monitoring node data through tower welding topology attribute data, and use the sensor monitoring integrated equipment configured with tower welding sensing and monitoring node data to perform tower welding substrate environmental monitoring processing to generate tower welding substrate environmental monitoring data.

4. The cryogenic environment-based tower weld parameter optimization method of claim 3, wherein, The environmental monitoring data for the tower welding substrate mentioned in step S142 includes tower welding space environmental data and tower substrate welding arc monitoring characteristic data.

5. The cryogenic environment-based tower weld parameter optimization method of claim 4, wherein, Step S2 includes the following steps: Step S21: Perform three-dimensional heat conduction finite element discrete modeling on the low-temperature welding multi-dimensional sensing data to generate low-temperature welding heat conduction discrete mesh data; Step S22: Perform low-temperature thermal energy boundary condition analysis based on the low-temperature welding heat conduction discrete mesh data to generate low-temperature thermal energy boundary condition data; Step S23: Perform dynamic characteristic analysis of thermal energy of the low-temperature tower based on the discrete mesh data of heat conduction in low-temperature welding and the dynamic boundary condition data of low-temperature thermal energy, and generate dynamic characteristic data of thermal energy of the low-temperature tower. Step S24: Based on the welding arc monitoring characteristic data of the tower substrate corresponding to the low-temperature welding multi-dimensional sensing data, perform welding heat source characteristic analysis to generate tower welding heat source characteristic data; Step S25: Obtain prior data of equivalent heat input for tower welding; Step S26: Establish the correction relationship for equivalent heat input offset by using low-temperature welding multi-dimensional sensing data and tower welding equivalent heat input prior data, and generate the tower welding equivalent heat input correction model; Step S27: Transmit the dynamic characteristic data of the thermal energy of the low-temperature tower and the characteristic data of the heat source of the tower welding to the equivalent heat input correction model of the tower welding to perform equivalent transient heat input field analysis of the tower welding, and generate equivalent transient heat input field data of the tower welding.

6. The cryogenic environment-based tower weld parameter optimization method of claim 1, wherein, Step S31 Includes the following steps: Step S311: Perform tower welding cooling characteristic analysis on the equivalent transient heat input field data of tower welding to generate tower welding cooling characteristic data; Step S312: Analyze the weld microstructure evolution characteristics based on the tower welding cooling characteristic data and the equivalent transient heat input field data of tower welding, and generate weld microstructure evolution characteristic data; Step S313: Perform low-temperature phase transformation mapping path analysis on the weld microstructure evolution characteristic data to generate low-temperature phase transformation mapping path data for the weld.

7. The cryogenic environment-based tower weld parameter optimization method of claim 1, wherein, Step S4 includes the following steps: Step S41: Analyze the characteristics of the predicted performance of weld microstructure based on the predicted performance evaluation data, and generate the predicted performance characteristics data of weld microstructure. Step S42: Perform residual stress characteristic analysis on the predicted performance state characteristic data of weld microstructure to generate residual stress characteristic data of tower welding; Step S43: Perform a risk characteristic analysis on the residual stress characteristic data of the tower welding structure to generate risk characteristic data of the tower welding structure.

8. The cryogenic environment-based tower weld parameter optimization method of claim 7, wherein, Step S43 includes the following steps: Step S431: Obtain prior data on fracture of welding materials for cryogenic towers; Step S432: Perform weld crack sensitivity feature analysis using prior fracture data of low temperature tower welding materials and weld microstructure prediction performance state characteristic data to generate weld crack sensitivity feature data. Step S433: Analyze the risk characteristics of the tower welded structure by using the weld crack sensitive feature data to analyze the residual stress feature data of the tower welded structure, and generate the risk feature data of the tower welded structure.

9. The cryogenic environment-based tower section welding parameter optimization method of claim 1, wherein, Step S5 includes the following steps: Step S51: Based on the risk characteristic data of the tower welding structure, perform tower welding parameter optimization requirement analysis and generate tower welding parameter optimization requirement data; Step S52: Perform multi-objective optimization feature analysis on tower welding quality using weld microstructure prediction performance evaluation data and tower welding parameter optimization requirement data to generate multi-objective optimization feature data for tower welding quality; Step S53: Perform iterative simulation optimization of tower welding parameters based on the multi-objective optimization feature data of tower welding quality to generate iterative optimization data of tower welding parameters; Step S54: Design an intelligent control strategy for optimizing tower welding parameters by using iterative optimization data of tower welding parameters, and generate a tower welding parameter optimization strategy.

Images