A method for intelligently adjusting process parameters for manufacturing a tower drum

Abstract

The present application relates to the technical field of intelligent control, and particularly relates to a process parameter intelligent adjustment method for tower drum processing and manufacturing, which comprises the following steps: performing tower drum processing process state feature analysis according to tower drum multi-source processing data to generate tower drum processing process state feature data; performing tower drum process-quality correlation feature analysis based on the tower drum processing process state feature data to generate tower drum process state-quality correlation feature data; analyzing tower drum process parameter deviation influence feature data based on the tower drum process state-quality correlation feature data; performing tower drum optimal process parameter intelligent adjustment strategy analysis based on the tower drum process state-quality correlation feature data and the tower drum process parameter deviation influence feature data to generate a tower drum optimal process parameter intelligent adjustment strategy to perform tower drum processing process parameter intelligent adjustment operation. The present application realizes the comprehensiveness and high efficiency of process parameter intelligent adjustment in tower drum processing and manufacturing.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology, and in particular to a method for intelligent adjustment of process parameters in tower manufacturing. Background Technology

[0002] As the core supporting component of a wind turbine generator, the quality of its processing directly determines the stability and safety of the turbine's operation. The tower processing flow is complex, typically involving multiple consecutive steps such as CNC blanking, thick plate beveling, plate rolling, spot welding positioning, longitudinal and circumferential seam welding, roundness correction, non-destructive testing, sandblasting and painting, etc. The process parameters of each step have a significant impact on the final processing quality. However, existing methods for adjusting process parameters in tower manufacturing largely rely on manual experience. Operators manually adjust process parameters based on on-site observations and their own experience, leading to significant fluctuations and problems such as misalignment, weld defects, and roundness deviations. Tower manufacturing involves multi-source monitoring data, including temperature, pressure, stress, and dimensions, making it difficult to comprehensively and in real-time capture the changing patterns of all data. This makes it impossible to promptly identify minor deviations in process parameters or predict the propagation and evolution of these deviations. Furthermore, it lacks the ability to dynamically optimize process parameters based on quality feedback and is ill-suited for coupled processing scenarios with different specifications and procedures, failing to meet the high-precision processing requirements of megawatt-class towers. In addition, existing tower manufacturing monitoring technologies primarily focus on structural health monitoring, such as monitoring vibration and stress in later operational stages, lacking real-time analysis methods for the correlation between process parameters and quality during manufacturing. They also lack a systematic analysis method for the propagation and evolution of process deviations, resulting in a lack of scientific basis for process parameter adjustments and an inability to fundamentally address the problem of unstable manufacturing quality. Summary of the Invention

[0003] Based on this, the present invention provides a method for intelligent adjustment of process parameters in tower manufacturing to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above objectives, a method for intelligent adjustment of process parameters in tower manufacturing includes the following steps:

[0005] Step S1: Monitor and process multi-source tower processing data using tower processing monitoring sensors to generate multi-source tower processing data; perform tower processing process state characteristic analysis based on the multi-source tower processing data to generate tower processing process state characteristic data.

[0006] Step S2: Perform tower process-quality correlation feature analysis based on tower processing technology status feature data to generate tower process status-quality correlation feature data;

[0007] Step S3: Based on the tower process state-quality correlation characteristic data, perform tower process deviation behavior propagation and evolution characteristic analysis to generate tower process deviation behavior propagation and evolution characteristic data; through the tower process deviation behavior propagation and evolution characteristic data, perform tower process parameter deviation influence characteristic analysis to generate tower process parameter deviation influence characteristic data.

[0008] Step S4: Based on the tower process state-quality correlation characteristic data and the tower process parameter deviation impact characteristic data, analyze the tower optimization process parameter intelligent adjustment strategy and generate the tower optimization process parameter intelligent adjustment strategy; execute the tower processing process parameter intelligent adjustment operation through the tower optimization process parameter intelligent adjustment strategy.

[0009] The beneficial effects of this application are as follows: This invention achieves comprehensive and real-time monitoring of multi-source processing data of the tower through tower processing monitoring sensors, solving the limitations of incomplete and lagging multi-source data capture in traditional processing, and providing complete and accurate data support for subsequent process status analysis. Through a layered and step-by-step analysis approach, it first analyzes the basic process parameters of the multi-source data, then classifies the process stage types and collects the operating condition data for each stage. Combining the basic process parameters and operating condition data, it completes the process status characteristic analysis, progressing step by step with rigorous logic, effectively avoiding the problem of misjudgment of process status caused by single-data-dimensional analysis. In particular, by identifying the behavioral units of the process stages, analyzing behavioral patterns, and analyzing the operating condition response characteristics, it can accurately uncover the core behavioral characteristics and operating condition adaptation rules of different process stages. Combined with the stage type node design, it achieves accurate characterization of the process status of each stage, and can promptly capture minor fluctuations and abnormal tendencies in process parameters, improving the objectivity and efficiency of process status analysis. Based on the characteristic data of tower machining processes, a bridge is constructed between process status and machining quality, effectively addressing the core deficiencies of lacking methods for analyzing the correlation between process parameters and machining quality, and the lack of quality guidance in process adjustment. By acquiring prior data for quality assessment, analyzing quality evaluation indicators, quantifying and fitting quality fluctuations at each stage, it can accurately uncover the regular characteristics of quality fluctuations at different process stages, thereby establishing a comprehensive assessment mapping relationship between tower machining characteristics and quality. The constructed quality assessment model can achieve accurate prediction and evaluation of machining quality. Simultaneously, the quality evaluation indicators clearly cover three core dimensions: geometric quality, thermal behavior, and stress, comprehensively covering the key control points of tower machining quality and avoiding the quality control omissions caused by single quality indicator evaluation. By inputting the machining characterization data corresponding to the process status characteristics into the quality assessment model, accurate quality assessment data is generated. Combined with the process status characteristic data to complete the process-quality correlation analysis, the influence weight and mechanism of different process statuses on machining quality can be clearly defined, clarifying the core direction for process parameter optimization. Based on process status-quality correlation characteristic data, a systematic analysis framework for the propagation and evolution of process deviation behavior was constructed, effectively solving the prominent shortcomings of existing technologies, such as the inability to identify minor deviations in process parameters, the difficulty in predicting the propagation and evolution trend of deviation behavior, and the lack of scientific basis for process adjustment. Through hierarchical analysis of process deviation behavior and quality risk evolution characteristics, it analyzes the evolution law of quality risk not only for single-stage nodes but also considers the risk linkage impact of coupled stage nodes, comprehensively covering the deviation risks of all stages of tower processing and coupled process scenarios, avoiding the problem of incomplete risk prediction caused by single-stage analysis.By analyzing the evolution and propagation characteristics of process deviation behavior layer by layer, and combining multi-factor correlation analysis between process parameters and deviation behavior, the core causes of process parameter deviation can be accurately located. The propagation path, evolutionary pattern, and impact on processing quality at each process stage can be clarified. The generated process parameter deviation impact characteristic data provides precise deviation guidance for subsequent intelligent adjustment of process parameters, enabling early avoidance of quality problems such as misalignment and weld defects caused by the spread of deviation behavior. Integrating process status-quality correlation data with parameter deviation impact data achieves intelligent, precise, and scientific process parameter adjustment, completely solving the technical pain points of traditional manual parameter adjustment, such as large fluctuations, poor adaptability, and inability to dynamically optimize. Through analyzing and optimizing process parameters, defining adjustment boundaries, and allocating adjustment weights, combined with the analysis of the feasible domain of deviation adjustment coupled with multi-source process parameters, the constructed intelligent adjustment strategy not only meets the standard requirements of optimized process parameters but also adapts to the actual scenario of process parameter deviation, balancing the accuracy and feasibility of adjustment, and avoiding process disorder or quality hazards caused by blind adjustment. Meanwhile, this step can adapt to processing scenarios with different specifications and coupled processes. It automatically executes parameter adjustment operations through intelligent adjustment strategies without manual intervention, greatly reducing reliance on human experience and improving the efficiency and consistency of process parameter adjustment. It can dynamically respond to process deviations and quality feedback, realize real-time optimization of process parameters, solve the problem of unstable tower processing quality from the root, effectively support the high-precision processing requirements of large megawatt-level towers, further improve the quality and production efficiency of finished tower products, and realize the intelligent upgrade of tower processing technology.

[0010] Therefore, the intelligent adjustment method for process parameters in tower manufacturing of this invention achieves comprehensive and real-time monitoring of multi-source processing data of the tower through tower processing monitoring sensors. Combined with the acquisition of operating conditions and process state characteristic analysis at each process stage, it realizes intelligent adjustment of process parameters, effectively avoiding the problem of large parameter fluctuations caused by manual adjustment, reducing processing quality hazards such as misalignment, weld defects, and roundness deviations, and improving the consistency and pass rate of tower processing. Addressing the pain points of difficulty in comprehensively capturing the changing patterns of multi-source monitoring data, the inability to identify minor deviations in process parameters, and the inability to predict the propagation and evolution trends of deviations, this invention, through process-quality correlation feature analysis, process deviation behavior propagation and evolution feature analysis, and parameter deviation impact feature analysis, can accurately capture the changing patterns of multi-source data such as temperature, pressure, stress, and dimensions, promptly identify minor deviations in process parameters, and predict the propagation and evolution trends of deviation behavior, providing lead time and scientific basis for process parameter adjustment. By optimizing the process parameter adjustment strategy, process parameters can be dynamically optimized based on processing quality feedback. Furthermore, through the analysis of the feasible domain of deviation adjustment coupled with multi-source process parameters, it can flexibly adapt to coupled processing scenarios of different specifications and processes. In addition, to address the limitations of existing monitoring technologies that focus on monitoring the later stages of tower operation, a real-time correlation analysis system between process parameters and processing quality was constructed, forming a systematic analysis method for the propagation and evolution of process deviation behavior, thus fundamentally solving the problem of unstable processing quality of existing towers. Attached Figure Description

[0011] Figure 1 This is a schematic diagram of the steps of an intelligent adjustment method for process parameters in tower manufacturing according to the present invention;

[0012] Figure 2 for Figure 1 A detailed flowchart illustrating the implementation steps of step S4.

[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 an intelligent adjustment method for process parameters in tower manufacturing. In the embodiments of this invention, please refer to... Figure 1 The diagram shown is a flowchart illustrating the steps of an intelligent adjustment method for process parameters in tower manufacturing according to the present invention. The intelligent adjustment method for process parameters in tower manufacturing includes the following steps:

[0017] Step S1: Monitor and process multi-source tower processing data using tower processing monitoring sensors to generate multi-source tower processing data; perform tower processing process state characteristic analysis based on the multi-source tower processing data to generate tower processing process state characteristic data.

[0018] In this embodiment of the invention, when monitoring and processing multi-source processing data of tower cylinders using tower cylinder processing monitoring sensors, monitoring sensors that are precisely matched to the processing links are deployed throughout the entire tower cylinder processing process. The monitoring sensors are deployed in zones according to multiple core process stages such as rolling, welding, rounding, and flaw detection. Each stage of the sensor corresponds to a specific monitoring object, and the deployment position strictly corresponds to the key processing parts of each process stage. In the rolling stage, sensors are deployed at the rolling rollers and blank clamping points of the rolling equipment; in the welding stage, sensors are deployed at the welding torch, welding area, and weld formation point; in the rounding stage, sensors are deployed at the rounding pressure rollers and tower cylinder positioning points; and in the flaw detection stage, sensors are deployed at the probe head and detection area. Monitoring and processing are carried out continuously according to the processing sequence. Sensors capture the physical signals of each monitored object in real time, converting these signals into quantifiable and analyzable digital signals to form multi-source processing data for the tower. This data covers three main categories: equipment operating status data, real-time status data of processed parts, and processing environment data. Equipment operating status data includes operating parameters, load status, and operational stability of the processing equipment at each process stage. Real-time status data of processed parts includes deformation, dimensional accuracy, and surface condition of the billet and processed parts. Processing environment data includes environmental data affecting processing quality, such as welding zone temperature. After monitoring is completed, the generated multi-source processing data for the tower undergoes layered analysis, categorized and organized according to process stage and data type, and damaged or abnormal data is removed to ensure data integrity and accuracy. After the multi-source processing data is processed, the tower processing process status characteristics are analyzed. The analysis is carried out step by step according to multiple process stages. Combining the processing technology requirements and operation standards of each process stage, a process status analysis system is constructed. Through correlation analysis, trend analysis, and anomaly identification of multi-source data, the process behavior characteristics and operating condition response characteristics of each process stage are extracted. The standardization and stability of process execution at each stage are clarified, the process status is judged to meet the preset standards, abnormal signs appearing in the process are identified, and the process status characteristics and analysis results of each stage are integrated to generate tower processing process status characteristic data.

[0019] Step S2: Perform tower process-quality correlation feature analysis based on tower processing technology status feature data to generate tower process status-quality correlation feature data;

[0020] In this embodiment of the invention, the core analytical basis is the characteristic data of the tower processing technology. Combined with the tower processing quality evaluation standards and the quality control requirements of each process stage, a comprehensive tower process-quality correlation analysis system is constructed. Through a systematic correlation analysis method, the inherent correspondence between the tower processing technology state and processing quality is explored, clarifying the impact mechanism of process state changes on processing quality. The analysis process first identifies the core evaluation dimensions of tower processing quality, covering geometric accuracy quality, structural strength quality, surface and internal defect quality, etc. Each evaluation dimension corresponds to a clear quality judgment standard, with no ambiguity in judgment conditions. Subsequently, correlation analysis is carried out layer by layer according to multiple process stages. Within each stage, the focus is on the process state characteristics of that stage and the corresponding quality state. First, the process state characteristic data of that stage is extracted to clarify core information such as process parameter values, process behavior execution, and operating condition response status. Then, the corresponding processing quality inspection data of that stage is extracted to clarify information such as quality compliance status, quality defect types, and defect severity. By synchronously analyzing and comparing process status data and quality data at the same stage, the system clarifies the quality status corresponding to normal process status, identifies quality defects corresponding to abnormal process status, establishes a one-to-one correspondence between process status characteristics and quality characteristics, defines the quality change patterns caused by different process status changes, and distinguishes between quality changes dominated by process status and those caused by other irrelevant factors. Simultaneously, by integrating the correlation analysis results of each process stage, the system clarifies the influence weight of each stage's process status on the overall processing quality, analyzes the comprehensive impact of cross-stage process status linkages on quality, and generates tower process status-quality correlation characteristic data.

[0021] Step S3: Based on the tower process state-quality correlation characteristic data, perform tower process deviation behavior propagation and evolution characteristic analysis to generate tower process deviation behavior propagation and evolution characteristic data; through the tower process deviation behavior propagation and evolution characteristic data, perform tower process parameter deviation influence characteristic analysis to generate tower process parameter deviation influence characteristic data.

[0022] In this embodiment of the invention, when analyzing the propagation and evolution characteristics of tower process deviation behavior based on tower process state-quality correlation feature data, process deviation behaviors with a clear correlation to quality defects are first identified. The specific manifestations, occurrence scenarios, and initial characteristics of these deviation behaviors are defined, and the criteria for determining deviation behaviors are clarified. Only process state anomalies that lead to quality defects or a decline in quality level are identified, while minor fluctuations with no quality impact are eliminated. Subsequently, the entire process of each identified process deviation behavior, from its occurrence and development to stabilization or termination, is tracked. The propagation path, propagation rate, and propagation range of deviation behaviors at each process stage and processing node are analyzed. The evolution trend, evolution law, and evolution endpoint state of deviation behaviors are clarified. The cross-node and cross-stage propagation mechanisms of deviation behaviors are identified. The propagation differences between single-node deviation behaviors and cross-stage coupled deviation behaviors are distinguished. The linkage relationship between deviation behavior propagation and operating conditions and quality status is analyzed. The propagation starting point, propagation path, propagation rate, evolution trend, and propagation impact of deviation behaviors are recorded in detail, generating tower process deviation behavior propagation and evolution feature data. After generating the propagation and evolution characteristic data of process deviation behavior, the influence characteristics of tower process parameter deviation are analyzed using this data. Taking the generated propagation and evolution characteristic data of process deviation behavior as the core, combined with the process state-quality correlation characteristic data, the analysis focuses on each core process parameter, classifies and analyzes them according to the type of process parameter, clarifies the correspondence between the direction and magnitude of each parameter deviation and the deviation behavior and quality risk, defines the scope, degree, mechanism and controllability of the parameter deviation, clarifies the quality risk level caused by different parameter deviations, distinguishes between reversible and irreversible effects, sorts out the transmission path and superposition effect of the parameter deviation influence, and generates the influence characteristic data of tower process parameter deviation.

[0023] Step S4: Based on the tower process state-quality correlation characteristic data and the tower process parameter deviation impact characteristic data, analyze the tower optimization process parameter intelligent adjustment strategy and generate the tower optimization process parameter intelligent adjustment strategy; execute the tower processing process parameter intelligent adjustment operation through the tower optimization process parameter intelligent adjustment strategy.

[0024] In this embodiment of the invention, based on process state-quality correlation feature data, optimal process parameters that can ensure normal process state and qualified processing quality are selected, and the optimal process parameter standards for each process stage and each processing node are clarified. Combining the process parameter deviation impact feature data, the adjustment boundaries and constraints of each optimal process parameter are clarified, defining the adjustable range and limits of parameter adjustment to avoid over-adjustment leading to process disorder or quality defects. Subsequently, the adjustment weight of each process parameter is analyzed. Based on the degree of impact of parameter deviation on quality, the range of deviation propagation, the irreversibility of risk, and the difficulty of adjustment, the adjustment priority of each parameter is determined to ensure that the adjustment process prioritizes parameters with a significant impact on quality and serious deviation hazards. Combining the process deviation behavior propagation and evolution feature data, targeted adjustment schemes are formulated for different deviation scenarios (single parameter deviation, multi-parameter coupled deviation), clarifying the adjustment parameters, adjustment direction, adjustment amplitude, adjustment sequence, and adjustment verification standards for each deviation scenario. Precautions during the adjustment process are outlined to avoid triggering new process deviations or quality defects during adjustment. All adjustment schemes, adjustment priorities, adjustment boundaries, and verification standards are integrated to form a complete adjustment rule system, generating an intelligent adjustment strategy for the optimal process parameters of the tower. After the optimal intelligent adjustment strategy for process parameters is generated, it is used to execute intelligent adjustment operations for tower processing process parameters. The generated intelligent adjustment strategy serves as the sole execution basis, establishing a linkage mechanism with the tower processing equipment. It receives process status data and multi-source processing data in real time, monitors the real-time values ​​of process parameters, compares them with the optimal process parameter standards, and determines whether there are parameter deviations and deviation scenarios. When a parameter deviation is detected, the corresponding adjustment scheme for the deviation scenario is immediately retrieved, and a precise adjustment command is sent to the processing equipment. The equipment automatically executes the parameter adjustment according to the command. During the adjustment process, changes in process parameters, process status, and quality status are monitored in real time to ensure that the adjusted process parameters return to the optimal range, the process status returns to normal, and quality risks are eliminated. After adjustment, the adjustment effect is verified, and the adjustment process, adjusted parameters, and adjustment effect are recorded. If the parameters deviate again, the corresponding adjustment scheme is immediately executed again to ensure that the process parameters remain within the optimal range throughout the entire processing process, completing the intelligent adjustment operation of tower processing process parameters and achieving intelligent and precise control of process parameters.

[0025] Furthermore, step S1 includes the following steps:

[0026] Step S11: Monitor and process multi-source tower processing data using tower processing monitoring sensors to generate multi-source tower processing data;

[0027] In this embodiment of the invention, tower processing monitoring sensors are deployed throughout the entire tower processing process. These sensors are high-temperature resistant and vibration-resistant industrial sensors, fixedly installed on both sides of the rolling rollers of the tower rolling equipment, at the end of the welding torch and around the weld seam of the welding equipment, and at the contact point of the pressure rollers of the straightening equipment. Simultaneously, corresponding sensors are fixedly installed on both sides of the conveying track at the tower billet feeding end, beside the clamping device in the middle of processing, and above the detection platform at the discharge end, forming a comprehensive, blind-spot-free monitoring layout. The sensors capture various data in real time during the tower processing process. Temperature data is captured by contact temperature sensors to measure the real-time temperature of each processing part; pressure data is captured by pressure sensors to measure the contact pressure and welding gas pressure during rolling and straightening; stress data is captured by strain gauge sensors to measure the stress distribution of the tower billet and the processed part; dimensional data is captured by laser rangefinders to measure the diameter, wall thickness, and length of the processed part; processing speed data is obtained by converting the operating speed of the equipment using speed sensors; and billet material parameters are captured by material sensors to measure the composition and hardness of the billet. The monitoring process is continuous and uninterrupted. The captured raw analog signals are converted into digital signals by the data conversion module built into the sensor. Then, the electromagnetic interference signals generated by the operation of the equipment are removed by low-pass filtering. Finally, all the processed digital signals are integrated according to the processing time sequence to form multi-source processing data of the tower.

[0028] Step S12: Perform analytical processing of tower foundation process parameters based on multi-source tower processing data to generate analytical data of tower foundation process parameters;

[0029] In this embodiment of the invention, based on multi-source processing data of the tower, a data parsing algorithm is used to classify, extract, and analyze the multi-source data. The parsing algorithm, based on preset process parameter association rules, focuses on core data directly related to the tower processing technology, directly eliminating redundant data unrelated to process parameters, such as environmental humidity and equipment shell temperature during billet conveying. Core process parameters such as tower rolling angle, rolling force, welding current, welding voltage, welding temperature, straightening pressure, processing feed, initial billet wall thickness, rolling radius, and welding speed are accurately extracted from the multi-source processing data. The extracted parameters are standardized: rolling angle is uniformly converted to degrees, rolling force and straightening pressure are uniformly converted to kilonewtons, welding current is converted to amperes, welding voltage is converted to volts, and processing feed is converted to millimeters per minute. All parameters are expressed as decimal integers, unifying the parameter expression format and units of measurement. After standardization, the data is verified by the preset parameter threshold range. Abnormal data that exceeds the normal processing range is removed. The core process parameters that pass the verification are sorted according to the processing sequence, and finally the basic process parameter analysis data of the tower is formed. This clearly presents the specific values ​​of the core process parameters in each stage of the tower processing, providing a clear basis for the subsequent process stage division.

[0030] Step S13: Perform process stage type classification on the analytical data of tower foundation process parameters to generate tower foundation process stage type data;

[0031] In this embodiment of the invention, based on the preset process flow of tower processing and combined with the analytical data of the basic process parameters of the tower, the basic process parameters after analysis are clearly divided into stages according to the sequence of processing steps and the differences in process functions. The division is based entirely on the functional attributes of the process parameters and the actual operation scenario, without introducing other additional division standards. The basic process parameters are clearly divided into several core stages, such as the rolling process stage, the welding process stage, the rounding process stage, and the flaw detection process stage. The data corresponding to the rolling process stage includes process parameters directly related to the rolling operation, such as the rolling angle, rolling force, rolling radius, initial value of billet wall thickness, and processing feed rate. The data corresponding to the welding process stage includes process parameters directly related to the welding operation, such as the welding current, welding voltage, welding temperature, welding speed, and weld width. The data corresponding to the rounding process stage includes process parameters directly related to the rounding operation, such as the rounding pressure, the number of rounding times, the measured value of tower roundness, and the rounding speed. The data corresponding to the flaw detection process stage includes process parameters directly related to the flaw detection operation, such as the flaw detection frequency, detection depth, and defect reflection intensity. The basic process parameters corresponding to each stage are classified separately, and the core parameter ranges corresponding to each process stage are clearly defined. The rolling angle is set between 85 degrees and 95 degrees, the welding current is set between 180 amperes and 220 amperes, the rounding pressure is set between 50 kN and 80 kN, and the flaw detection frequency is set between 1 MHz and 5 MHz. The process parameters of each stage are marked with corresponding stage identifiers, generating tower foundation process stage type data, realizing the stage-by-stage management of process parameters, and providing a clear basis for subsequent data collection of operating conditions in each stage.

[0032] Step S14: Based on the tower foundation process stage type data, collect tower processing conditions for each process stage type from the multi-source processing data of the tower to generate tower processing condition data;

[0033] In this embodiment of the invention, the data of the tower foundation process stage type is used as the clear dividing basis. Corresponding to multiple core process stages such as rolling, welding, rounding, and flaw detection, tower processing condition data of each stage are collected. The collection process is carried out synchronously with the processing operation of each stage to ensure the consistency of the timing of the condition data and process parameters. For the rolling process, the equipment operation status sensor captures the rolling roller speed and motor output power of the rolling equipment, and the deformation sensor captures the real-time deformation degree and deformation rate of the billet during the rolling process. The equipment operation status data and the billet deformation data are then correlated and integrated. For the welding process, the ambient temperature sensor captures the real-time ambient temperature of the welding area, and the visual monitoring module captures the weld forming width, forming height, and surface flatness. The ambient temperature data and the weld forming data are then correlated and integrated. For the rounding process, the load sensor captures the motor operating load and pressure roller contact load of the rounding equipment, and the roundness measurement module captures the real-time roundness change and roundness error of the tower during the rounding process. The equipment operating load data and the tower roundness change data are then correlated and integrated. For the flaw detection process, the surface detection module captures surface conditions such as scratches and dents in the flaw detection area, and the internal detection module captures internal structural conditions such as pores and cracks inside the tower. The surface condition data and the internal structural condition data are then correlated and integrated. The working condition information collected at each stage is matched with the basic process parameters of the corresponding stage according to the processing sequence to generate tower processing working condition data.

[0034] Step S15: Based on the analytical data of the tower foundation process parameters and the tower processing condition data, perform a tower processing process state characteristic analysis to generate tower processing process state characteristic data.

[0035] In this embodiment of the invention, analytical data of tower foundation process parameters and tower processing conditions are integrated to construct a correlation analysis model between process parameters and operating conditions. The model is based on the preset logic of each process stage and comprehensively analyzes the processing conditions of each process stage. Each process stage has preset clear standard parameters and standard operating conditions. For the rolling process stage, the preset standard value for the rolling angle is 90 degrees and the preset standard value for the rolling force is 60 kN. The standard operating condition is that the rolling roller speed is stable and the blank deformation is uniform without wrinkles. For the welding process stage, the preset standard value for the welding current is 200 amperes and the preset standard value for the welding voltage is 28 volts. The standard operating condition is that the ambient temperature of the welding area is stable at about 25 degrees Celsius and the weld is uniform without undercut. For the rounding process stage, the preset standard value for the rounding pressure is 65 kN and the standard value for the tower roundness is that the diameter deviation does not exceed 2 mm. The standard operating condition is that the equipment operating load is stable and the pressure roller contact is uniform. For the flaw detection process stage, the preset standard value for the flaw detection frequency is 3 MHz and the preset standard value for the detection depth is 10 mm. The standard operating condition is that there are no obvious surface defects in the detection area and no excessive pores or cracks inside. The difference between the actual process parameters and the standard parameters of the corresponding stage is calculated and analyzed. The magnitude of the difference is compared and analyzed. At the same time, the specific differences between the actual operating conditions and the standard operating conditions are compared to explore the intrinsic relationship between the changes in process parameter values ​​and the changes in operating conditions. The state characteristics of each process stage are extracted, and the degree of matching between the process parameters and the operating conditions of each stage is marked. If the difference between the actual parameters and the standard parameters is within 5% and the actual operating conditions are consistent with the standard conditions, it is marked as a good match. Otherwise, it is marked as an abnormal match, and the tower processing process state characteristic data is generated.

[0036] Furthermore, step S15 includes the following steps:

[0037] Step S151: Based on the tower foundation process stage type data and tower foundation process parameter analysis data, perform tower process behavior characteristic analysis and processing to generate tower process behavior characteristic data;

[0038] In this embodiment of the invention, data on the process stages of the tower foundation and analytical data on the process parameters of the tower foundation are integrated. Taking the core process stages such as rolling, welding, rounding, and flaw detection as units, the data on the process stages of the tower foundation is used to analyze and process the characteristics of the tower's process behavior one by one. For each process stage, all analytical data on the basic process parameters corresponding to that stage are extracted first. Combined with the preset core parameter ranges for that stage, the value range and change threshold of each process parameter are clarified. Specifically, the rolling process stage focuses on the changes in the values ​​of rolling angle, rolling force, and rolling radius; the welding process stage focuses on the changes in the values ​​of welding current, welding voltage, and welding speed; the rounding process stage focuses on the changes in the values ​​of rounding pressure, rounding times, and measured roundness; and the flaw detection process stage focuses on the changes in the values ​​of flaw detection frequency and detection depth. The process parameters within each stage are continuously tracked according to the processing sequence, capturing the fluctuations and trends of each parameter during processing. Stable and fluctuating segments are divided for the parameters, and the amplitude and rate of change of the parameter values ​​in the fluctuating segments are calculated. Simultaneously, by combining the operational logic of each process stage, the interrelationships between different process parameters are analyzed. The changes in rolling angle and rolling force form a linkage analysis, the changes in welding current and welding temperature form a linkage analysis, the changes in straightening pressure and the changes in measured roundness form a linkage analysis, and the changes in flaw detection frequency and detection depth form a linkage analysis, thereby generating tower process behavior characteristic data.

[0039] Step S152: Perform tower processing condition response characteristic analysis on the tower processing condition data to generate tower processing condition response characteristic data;

[0040] In this embodiment of the invention, tower processing condition data is used as the core, and tower processing condition response characteristic analysis is carried out according to multiple process stages such as rolling, welding, rounding, and flaw detection. This ensures that the condition response analysis accurately corresponds to the process stage and fits the actual processing scenario of each stage. For the condition data of the rolling process stage, the focus is on analyzing the operational stability of the rolling roller speed and motor output power, as well as the changes in the degree and rate of blank deformation. This determines the response of the rolling equipment's operating state and the blank deformation state to the rolling process parameters. If the rolling force increases, the blank deformation rate increases and the motor output power increases. If the rolling angle is adjusted, the rolling roller speed is adjusted synchronously. This detailed characterization of the response relationship between the equipment and the blank is achieved. For the welding process stage, the focus is on analyzing the stability of the ambient temperature in the welding area, as well as the changes in weld width, weld height, and surface smoothness. This captures the impact of ambient temperature changes on the weld formation state, and also analyzes the inverse response of the weld formation state to welding parameters. When the ambient temperature increases, the weld width changes accordingly; when the welding voltage is adjusted, the weld surface smoothness changes synchronously. This bidirectional response correlation is accurately recorded. For the rounding process stage, the focus is on analyzing the fluctuation characteristics of motor operating load and pressure roller contact load, as well as the changes in tower roundness and the adjustment characteristics of roundness error. The response relationship between rounding pressure changes and equipment load and tower roundness is characterized. Increasing the rounding pressure synchronously increases the motor operating load and pressure roller contact load, and the roundness error decreases accordingly. The specific patterns of these response changes are recorded in detail. For the working condition data of the flaw detection process, the focus is on analyzing the surface condition and internal structural condition of the detection area, capturing the detection response of surface scratches, internal pores and cracks when flaw detection parameters are adjusted. The detection sensitivity of internal cracks is correspondingly improved when the flaw detection frequency is increased, and the detection range of internal pores is correspondingly expanded when the detection depth is increased. The working condition response characteristics and laws of each stage are integrated to generate tower processing working condition response characteristic data.

[0041] Step S153: Analyze the tower processing process status characteristics of each stage type node based on the tower process behavior characteristic data and the tower processing condition response characteristic data, and generate tower processing process status characteristic data.

[0042] In this embodiment of the invention, tower process behavior characteristic data and tower processing condition response characteristic data are integrated. First, based on the basic process stage type data of the tower, dedicated stage type nodes are designed for multiple process stages such as rolling, welding, straightening, and flaw detection. Each stage type node corresponds to the core processing operation of that stage. The rolling stage has two nodes: billet feeding and rolling forming; the welding stage has three nodes: arc ignition, welding, and arc extinguishing; the straightening stage has two nodes: initial straightening and fine straightening; and the flaw detection stage has two nodes: surface detection and internal detection. Each node clearly corresponds to a specific range of process behavior characteristics and condition response characteristics. For each stage type node, the corresponding process behavior characteristic data and condition response characteristic data are correlated and matched to explore the inherent correlation between process behavior and condition response. It is determined whether the process behavior characteristics match the condition response characteristics. If the rolling angle and rolling force of the forming node in the rolling stage are within the preset process behavior range, and the corresponding billet deformation degree, roller speed, and other condition response characteristics meet the preset standards, then the process state of that node is determined to be normal. Each stage type node is analyzed comprehensively, focusing on verifying the parameter fluctuations and trends in the process behavior characteristics and their consistency with the equipment operating status, billet and processed part status in the operating condition response characteristics. If parameter fluctuations in the process behavior exceed the preset range, and equipment operation abnormalities and processed part status not meeting requirements occur simultaneously in the operating condition response, the process status of that node is determined to be abnormal, and the cause of the abnormality is identified, i.e., the process parameter fluctuations lead to the abnormal operating condition response. The process status analysis results of all stage type nodes are summarized, and the specific characteristics of normal or abnormal process status of each node are marked. The analysis data of each node are integrated to form a complete process status profile of each stage, generating tower processing process status characteristic data.

[0043] Furthermore, step S151 includes the following steps:

[0044] The data on the process stages of the tower foundation are processed to identify the behavioral units of the tower process stages, thereby generating behavioral unit data for the tower process stages.

[0045] Based on the behavioral unit data of the tower process stage, analyze the behavioral patterns of the tower process stage to generate tower process behavioral pattern data.

[0046] By analyzing the analytical data of the tower foundation process parameters using tower process behavior pattern data, tower process behavior characteristic data is generated.

[0047] In this embodiment of the invention, the data on the stages of the tower foundation process is used as the core basis. This data clearly defines multiple core process stages, such as rolling, welding, rounding, and flaw detection, along with their corresponding basic process parameters. Identification and processing are carried out step-by-step according to multiple process stages, using the smallest functional element of the actual processing operation as the identification standard to clearly define the boundaries of the behavioral units. The rolling process is divided into several behavioral units, including billet positioning, roll adjustment, continuous rolling, and rolling forming inspection, corresponding to parameters such as initial billet wall thickness, rolling angle, rolling force, and rolling radius. The welding process is divided into five behavioral units: welding torch positioning, arc ignition, continuous welding, arc extinguishing, and preliminary weld formation, corresponding to parameters such as welding torch position, arc ignition current and voltage, and welding parameters. The rounding process is divided into several behavioral units, including tower positioning, initial rounding, fine rounding, and roundness inspection, corresponding to parameters such as tower placement position, rounding pressure, and measured roundness. The flaw detection process is divided into several behavioral units, including flaw detection position calibration, surface detection, internal detection, and detection result recording, corresponding to parameters such as flaw detection position, flaw detection frequency, and detection sensitivity. Each behavioral unit is labeled with a unique identifier, corresponding process stage identifier, associated parameter items, and value ranges. These are then integrated to generate tower process stage behavioral unit data, providing a precise unit division basis for subsequent behavioral pattern analysis. Based on the identified behavioral unit data, analysis is conducted according to multiple process stages. The core is to capture the variation patterns of process parameters, execution sequence, and inter-unit linkage logic of each behavioral unit to form a standardized behavioral pattern. For each behavioral unit, corresponding parameter changes are tracked according to the processing sequence, extracting stable segments, periods of change, and patterns. Combined with the operational objectives, a standard execution pattern is summarized: In the rolling stage, the billet positioning parameters are stabilized after a one-time adjustment; the rolling rollers are adjusted gradually to the standard and stabilized; continuous rolling parameters remain constant; and forming detection involves rapid acquisition and comparison. In the welding stage, the arc ignition current and voltage rapidly rise to the preset value and stabilize; continuous welding parameters remain constant; and the arc extinguishing current and voltage gradually decrease to zero. In the rounding stage, the initial straightening pressure gradually increases to the preset value; and the fine straightening pressure remains constant. In the flaw detection stage, the detection parameters are constant, and the speed is uniform. Simultaneously, the linkage sequence of each behavioral unit is clarified, and execution in a fixed order forms an overall stage pattern. The standard patterns, linkage sequences, and parameter patterns of each unit are integrated, the core features of the pattern are marked, and the differences between normal and abnormal patterns are distinguished. Finally, tower process behavioral pattern data is generated, providing a pattern reference for subsequent process behavioral characteristic analysis. This study integrates tower process behavior pattern data and analyzes the basic process parameters of the tower layer by layer, according to process stage and behavior unit. The core is to compare the actual process parameters with the standard behavior pattern, extract the actual behavior characteristics and the degree of conformity with the standard pattern. The analysis is carried out one by one by behavior unit. First, the time series changes, fluctuations and rates of the actual parameters are restored, and then they are compared with the corresponding standard pattern item by item. The focus is on checking whether the initial value of the parameter, the value of the stable section, the change pattern and the execution sequence conform to the standard.During the rolling stage, the actual parameters of the continuous rolling unit are compared with the standard mode to determine whether the parameters are stable and the linkage is smooth. During the welding stage, the parameters of the continuous welding unit are compared to determine whether the parameters are stable and the forming is standardized. All behavioral units are compared one by one, the characteristics of each unit and the linkage between units are extracted, the overall characteristics of each process stage are integrated, the proportion of normal behavioral units, the type of abnormality and the overall stability are marked, and the tower process behavior characteristic data are generated.

[0048] Furthermore, step S153 includes the following steps:

[0049] Based on the tower processing condition response characteristic data, tower processing condition status analysis is performed to generate tower processing condition status data.

[0050] The stage type node of the tower processing process state is designed by using the stage type data of the tower foundation process, and the node data of the tower processing process state is obtained.

[0051] Based on the tower process behavior characteristic data and tower processing condition status data, the process behavior and condition response status of tower processing are correlated to generate tower process behavior-condition response status correlation data.

[0052] The data related to the tower process behavior and operating condition response status is transmitted to the tower processing process status node data for tower processing process status feature analysis, generating tower processing process status feature data.

[0053] In this embodiment of the invention, the core is the tower processing condition response characteristic data. Combined with the operational requirements of each stage, a preset standard range for the condition response characteristics is established, without setting fuzzy judgment conditions. During the rolling stage, the focus is on analyzing the stability of the rolling roller speed, motor output power, and billet deformation. Stable parameters and uniform deformation indicate normal operation; fluctuations in speed or uneven deformation indicate abnormal operation, and the type of abnormal equipment operation or billet deformation is marked. During the welding stage, the focus is on analyzing the stability of the ambient temperature and the weld formation state. Constant temperature and uniform weld indicate normal operation; temperature fluctuations or undercut / protrusion in the weld indicate abnormal operation, and the type of abnormal environmental interference or weld formation is marked. During the rounding stage, the focus is on analyzing the stability of the equipment load and changes in the tower's roundness. Stable load and a stable roundness error indicate normal operation; large load fluctuations or increased roundness error indicate abnormal operation, and the type of abnormal load or rounding effect is marked. During the flaw detection stage, the focus is on analyzing the surface and internal state of the detection area. No defects and parameter response meeting standards indicate normal operation; detected defects indicate abnormal operation, and the type of surface or internal defect is marked. By integrating the judgment results, anomaly types, and working condition details from each stage, and sorting them according to the processing sequence, tower processing working condition status data is generated, fully presenting the actual working condition status of each stage. Stage type nodes for tower processing are designed using the tower's basic process stage type data, which clearly defines multiple core process stages and corresponding basic process parameters. Node design follows the work flow of each stage, designing dedicated nodes for each stage according to the sequence of work steps and functional differences. Each node corresponds to a core work step, without crossing stages or duplication, ensuring clear functions and boundaries, and aligning with the intelligent adjustment requirements of process parameters. The rolling stage designs three nodes: billet positioning, rolling forming, and rolling inspection, respectively associated with billet positioning parameters, rolling core parameters, and rolling forming inspection parameters. The welding stage designs multiple nodes, including welding torch positioning, arc ignition, continuous welding, and weld inspection, respectively associated with welding torch positioning, arc ignition, welding core, and weld forming parameters. The rounding stage is designed with three nodes: tower positioning, rounding operation, and roundness detection, which are respectively associated with the core positioning and rounding core parameters and the measured roundness parameters. The flaw detection stage is designed with three nodes: flaw detection calibration, surface detection, and internal detection, which are respectively associated with the calibration, surface detection, and internal detection parameters. Each node is labeled with a unique identifier, corresponding process stage, associated parameters, and operating condition response characteristics, clarifying process requirements and judgment criteria. All node information is integrated to obtain tower processing process status node data, realizing the node-based decomposition of operation links. The process behavior characteristic data and operating condition status data are integrated, and correlation processing is carried out layer by layer according to process stage and node. The core is to explore the inherent correspondence between process behavior and operating condition status within the same node and clarify the impact of process behavior on operating condition status. The correlation processing is carried out in multiple stages one by one. Within each stage, it is carried out sequentially by node. First, the node process behavior characteristics (parameter values, change patterns, etc.) are extracted, and then the corresponding operating condition status data (status type, anomaly details, etc.) are extracted to establish the corresponding correlation between the two.In the rolling stage, at the forming node, normal process behavior (constant rolling angle and force) corresponds to normal operating conditions (stable rotation speed and uniform deformation), while abnormal process behavior (force fluctuations) corresponds to abnormal operating conditions (abnormal deformation rate and power). The reasons for the association are marked. In the welding stage, at the continuous welding node, normal process behavior (constant current and voltage) corresponds to normal operating conditions (stable temperature and uniform weld), while abnormal process behavior (voltage increase) corresponds to abnormal operating conditions (weld protrusion). All nodes are associated one by one, and the corresponding relationships and patterns are recorded. Operating condition changes dominated by process behavior are distinguished, and data is integrated by stage and time sequence to generate associated data, fully presenting the inherent relationship between process behavior and operating condition status. The tower process behavior-operating condition response status associated data is transmitted to the tower processing process status node data for tower processing process status characteristic analysis. This is performed node by node, first clarifying the node process requirements, associated parameters, and judgment criteria, and then combining the associated data to analyze whether the node process status meets the standards and extract status characteristics. If both the process behavior and operating condition of a node are normal and reasonably correlated, it is considered normal, and stable parameters and standardized behavior characteristics are extracted. If both are abnormal and clearly correlated, it is considered abnormal, and abnormal parameters, operating condition details, and correlation patterns are extracted, with the abnormality level and cause marked. A single abnormality is considered suspicious, and a verification step is marked. If the parameters of the rolling forming node are stable and the operating condition is normal, it is considered normal; if the force fluctuates or the deformation is uneven, it is considered abnormal. After analyzing all nodes one by one, the characteristics of each node are integrated, the overall status of each stage is summarized, the proportion of normal nodes, the distribution and type of abnormalities are marked, the mutual influence between stages is analyzed, and tower processing process status characteristic data is generated, accurately presenting the entire process status and abnormal details, meeting the needs of intelligent parameter adjustment.

[0054] Furthermore, step S2 includes the following steps:

[0055] Step S21: Obtain prior data for tower processing quality assessment;

[0056] In this embodiment of the invention, the focus is on the entire tower processing process and the quality control requirements of the finished product. Historical and measured data related to tower processing quality are collected. All data comes from the actual processing of towers of the same specifications and materials, ensuring the relevance and practicality of the data. This provides a reliable data foundation for subsequent quality evaluation index analysis and model establishment, aligning with the core quality-oriented requirements of intelligent adjustment methods for process parameters in tower manufacturing. The collected historical data includes process parameter data for each stage of tower processing under the same past process conditions, corresponding quality inspection data, finished product qualification data, and quality defect handling data, covering complete quality-related records for multiple core process stages such as rolling, welding, rounding, and flaw detection. The measured data were obtained through the quality inspection of the billet before processing the current batch of towers, the phased quality sampling inspection during the processing, and the comprehensive inspection of the finished product after processing. The billet quality inspection data includes data such as billet material composition, wall thickness uniformity, and hardness. The phased sampling inspection data includes data such as the dimensions, stress, and weld formation of the processed parts at each process stage. The comprehensive inspection data of the finished product includes data such as the roundness, wall thickness, weld quality, and overall strength of the finished tower. Finally, the preliminary data for the tower processing quality assessment are formed.

[0057] Step S22: Analyze the tower processing quality evaluation index based on the prior data of tower processing quality assessment, and generate tower processing quality evaluation index data;

[0058] In this embodiment of the invention, prior data for evaluating the quality of tower processing is categorized, analyzed, and refined to clearly define core evaluation indicators that comprehensively reflect the quality of tower processing. Each indicator corresponds to a clear physical meaning and measurement dimension, with no ambiguous indicators, ensuring that the evaluation indicators can accurately depict the quality status of each process stage and the finished product. Based on the quality-related records in the prior data, the quality evaluation indicators are clearly divided into three categories: geometric quality evaluation indicators for tower processing, thermal behavior evaluation indicators for tower processing, and stress evaluation indicators for tower processing. These three categories of indicators complement each other and provide comprehensive coverage, eliminating the omission of key quality influencing factors. Geometric quality evaluation indicators focus on the dimensional and shape accuracy of processed parts and finished products at each process stage. Specific indicators are extracted from the prior data, including billet wall thickness uniformity, rolling roundness, weld width, weld height, weld flatness, roundness error during correction, finished tower wall thickness deviation, finished product length deviation, and surface roughness. Each indicator corresponds to specific measured records in the prior data, clarifying the measurement direction of the indicator's physical meaning. The thermal behavior evaluation indicators focus on the impact of temperature changes during high-temperature processing such as welding on quality. Specific indicators such as the stability of the welding area's ambient temperature, weld cooling rate, weld heat-affected zone (HAZ) range, and HAZ hardness uniformity are extracted. Combined with temperature records and corresponding quality change data from the welding stage in the prior data, the intrinsic relationship between each indicator and welding quality is clarified. The stress evaluation indicators focus on the stress distribution of processed parts and finished products at each process stage. Specific indicators such as the uniformity of blank stress distribution during rolling, weld stress value after welding, residual stress in the tower after rounding, and overall stress distribution uniformity of the finished tower are extracted. Combined with stress detection records from the prior data, the impact of each indicator on the tower structure's stability and service life is clarified. The corresponding data source and measurement dimension are labeled for each specific evaluation indicator, and the value range of the indicator is clarified (determined based on the normal quality value range in the prior data). The three major categories of indicators and specific sub-indicators are integrated to generate tower processing quality evaluation indicator data.

[0059] Step S23: Quantify the tower processing quality fluctuation of each stage type on the tower processing quality evaluation index data to generate stage tower processing quality fluctuation quantitative data;

[0060] In this embodiment of the invention, the tower processing quality evaluation index data is subjected to quantitative processing of tower processing quality fluctuations at each stage. Fluctuation analysis and quantitative calculation are performed on each quality evaluation index at each stage. The quantitative results directly reflect the variation range and stability of the quality index at each stage, providing accurate quantitative data for subsequent stage-by-stage fitting processing and supporting the scientific nature of the quality assessment model. For each process stage, all quality evaluation index data corresponding to that stage are extracted, sorted by processing sequence, and the value changes of each index during the processing of that stage are tracked. Fluctuation quantification is carried out using difference calculation and fluctuation amplitude statistics. For each index, the standard value for that stage is first determined (based on the average value of the index under normal quality conditions in prior data). The difference between the actual value of the index at each time point and the standard value is calculated, and the sign and specific value of the difference are recorded. Positive and negative differences represent that the index value is higher or lower than the standard value, respectively. Simultaneously, the maximum difference, minimum difference, and cumulative sum of differences for each index within that stage are statistically analyzed to calculate the fluctuation amplitude of the index value within that stage. The fluctuation amplitude is determined by the difference between the maximum and minimum differences, directly reflecting the fluctuation range of the index value. The rolling stage focuses on quantifying the fluctuations in billet wall thickness uniformity and rolling roundness; the welding stage focuses on quantifying the fluctuations in weld width, weld height, ambient temperature stability in the welding area, and weld stress value; the rounding stage focuses on quantifying the fluctuations in rounding error and residual stress; the flaw detection stage focuses on quantifying the fluctuations in surface roughness and internal defect detection indicators; and the generation stage generates quantitative data on the fluctuations in tower processing quality.

[0061] Step S24: Perform staged fitting processing of tower processing quality fluctuation quantification based on the stage tower processing quality fluctuation quantification data to generate fitted tower processing quality fluctuation quantification data;

[0062] In this embodiment of the invention, based on the quantitative data of quality fluctuations in the staged tower processing, fitting processing is carried out separately for multiple process stages such as rolling, welding, rounding, and flaw detection. The core of the fitting processing is to explore the changing patterns of quality index fluctuations in each stage, ensuring that the fitted data can truly reproduce the inherent patterns of quality fluctuations. The fitting processing is carried out one by one for each process stage and each quality evaluation index. For the stage fluctuation quantitative data of each index, the fluctuation difference and fluctuation amplitude data are arranged in an orderly manner according to the processing time sequence. The fitting method is used to analyze the arranged data. The fitting method is determined based on the changing characteristics of index fluctuations. For indices with stable fluctuation trends and no drastic changes, a linear fitting method is used. For indices with non-linear fluctuation trends and regular fluctuations, a non-linear fitting method is used. The fitting process strictly follows the data change patterns and does not arbitrarily modify the fluctuation trends. During the fitting processing, random outliers (i.e., single data points that deviate too much from the overall fluctuation trend and are irregular) in the fluctuation quantitative data are removed to ensure that the fitting results can reflect the overall fluctuation patterns. The removal criterion is that the difference between a single data point and its adjacent data exceeds a preset multiple of the overall average fluctuation amplitude. In the rolling stage, the fluctuation data of rolling roundness and blank wall thickness uniformity are fitted to reconstruct the fluctuation trend of geometric quality indicators during the rolling process. In the welding stage, the fluctuation data of weld formation-related indicators, thermal behavior indicators, and stress indicators are fitted to reconstruct the fluctuation law of welding temperature and weld stress. In the rounding stage, the fluctuation data of roundness error and residual stress are fitted to reconstruct the stabilization process of quality indicators during the rounding process. In the flaw detection stage, the fluctuation data of defect detection-related indicators and surface quality indicators are fitted to reconstruct the fluctuation characteristics of quality detection indicators during the flaw detection process. The fitted tower processing quality fluctuation quantitative data are generated to clearly present the regular characteristics of quality indicator fluctuations in each process stage.

[0063] Step S25: Based on the fitted tower processing quality fluctuation quantification data, characterize the tower processing and establish the mapping relationship for the global quality assessment to obtain the tower processing quality assessment model;

[0064] In this embodiment of the invention, a mapping relationship for the characterization of tower processing and the overall quality assessment is established based on the fitted quantitative data of tower processing quality fluctuations. Tower processing characterization refers to the quantifiable representation of the process behavior characteristics and operating condition response characteristics at each process stage, including process parameter values, parameter fluctuation characteristics, equipment operating status, and the real-time status of the processed parts. These characterizations can all be described using the data generated in the preceding steps. The mapping relationship is established stage by stage, with each stage establishing a correlation between the processing characterization and the corresponding stage's quality evaluation indicators. Simultaneously, a correlation is established between the processing characterizations and stage quality indicators at each stage and the finished product quality indicators, forming a comprehensive mapping system. For the rolling stage, a mapping relationship is established between process characteristics such as rolling angle and rolling force and quality indicators such as rolling roundness and billet wall thickness uniformity. The corresponding proportions of rolling angle fluctuations and roundness error fluctuations, and the correspondence between changes in rolling force and changes in wall thickness uniformity are clarified. For the welding stage, a mapping relationship is established between process characteristics such as welding current, voltage, and temperature and weld formation indicators, thermal behavior indicators, and stress indicators. The correspondence between changes in welding current and changes in weld width, and the correspondence between welding temperature fluctuations and the range of the heat-affected zone are clarified. Simultaneously, the mapping relationships of each stage are integrated to establish a comprehensive mapping between overall processing characteristics and finished product quality indicators, clarifying the influence weight of each stage's process characteristics on finished product quality. All mapping relationships are standardized and organized, clarifying the corresponding rules for input (processing characteristic data) and output (quality assessment results), defining the quality grades and quality defect types corresponding to different processing characteristics, and forming a complete tower processing quality assessment model to ensure that the model can accurately output the corresponding quality assessment results based on the input processing characteristic data.

[0065] Step S26: Collect tower processing characterization data based on tower processing process state characteristic data to obtain tower processing characterization data;

[0066] In this embodiment of the invention, tower processing characterization data is collected based on the tower processing technology state characteristic data, including multiple dimensions such as process parameter values, parameter fluctuation characteristics, process behavior execution, and working condition response status at each process stage. Each dimension collects specific and quantifiable characterization information, without any fuzzy or abstract content. During the rolling stage, process parameter values ​​such as billet positioning parameters, rolling angle, rolling force, and rolling roller speed are collected, along with fluctuation characteristics such as the fluctuation amplitude and rate of the rolling angle and force, and working condition response status such as the billet deformation state and the rolling roller running state. During the welding stage, process parameter values ​​such as welding current, welding voltage, welding speed, and welding temperature are collected, along with the fluctuation characteristics of each parameter, and working condition response status such as the weld formation state and the ambient temperature state of the welding area. During the rounding stage, process parameter values ​​such as rounding pressure and rounding times are collected, along with pressure fluctuation characteristics, and working condition response status such as tower roundness changes and equipment load status. During the flaw detection stage, process parameter values ​​such as flaw detection frequency and detection depth are collected, along with parameter fluctuation characteristics, and working condition response status such as the surface and internal structure state of the detection area. During the data collection process, each characterization information is checked against the corresponding stage and node process status characteristic data to ensure that the collected characterization data is consistent with the process status characteristic data. Redundant information unrelated to the process status is removed, and tower processing characterization data is generated.

[0067] Step S27: Transmit the tower processing characterization data to the tower processing quality assessment model for tower processing quality assessment and generate tower processing quality assessment data;

[0068] In this embodiment of the invention, the evaluation process is performed step by step according to the process stage. First, the processing characterization data of each process stage is split according to the index type. The corresponding processing characterization input items in the model for that stage are matched one by one according to the pre-set mapping relationship of the model to determine the corresponding quality evaluation index value and quality level. In the rolling stage, the collected characterization data such as rolling angle and rolling force are input into the model. The model outputs the evaluation values ​​and quality levels of quality indicators such as rolling roundness and billet wall thickness uniformity according to the pre-set mapping relationship, and marks whether the quality of this stage meets the standard. In the welding stage, the characterization data such as welding current, voltage, and temperature are input into the model, and the model outputs the evaluation values ​​and quality levels of quality indicators such as weld formation, thermal behavior, and stress, and marks whether there are forming defects in the weld and whether the stress exceeds the standard. After the evaluation of each stage is completed, the model combines the quality evaluation results of each stage to comprehensively evaluate the overall quality of the finished tower, and outputs the finished product quality level, overall quality score, potential quality defects (if any) and the process characterization reasons corresponding to the defects. During the evaluation process, the evaluation results and mapping logic corresponding to each characterization data are recorded, the evaluation results are verified, and tower processing quality evaluation data are generated.

[0069] Step S28: Perform tower process state-quality correlation characteristic analysis based on tower processing technology status characteristic data and tower processing quality assessment data to generate tower process state-quality correlation characteristic data.

[0070] In this embodiment of the invention, the tower process state-quality correlation feature analysis is performed sequentially across multiple process stages. Within each stage, the analysis is conducted according to stage type nodes. First, the process state feature data corresponding to the node is extracted to clarify the process behavior state, operating condition response state, presence of process anomalies, and anomaly details of the node. Then, the quality assessment data corresponding to the node is extracted to clarify the quality index assessment results, quality grade, presence of quality defects, and defect types of the node. For the rolling forming node in the rolling process stage, the process state of the node (stability of rolling angle, rolling force, and blank deformation state) is correlated with the quality assessment results (assessment values ​​of rolling roundness, wall thickness uniformity, and quality grade). If the process state of the node is normal (stable parameters, uniform deformation), the corresponding quality assessment result meets the standard and the quality grade is high, establishing a positive correlation between the two. If the process state of the node is abnormal (fluctuation of rolling force, uneven blank deformation), the corresponding quality assessment result does not meet the standard and there are defects such as excessive roundness error, establishing an abnormal correlation between the two and marking the reason for the correlation (fluctuation of process parameters leading to quality defects). For continuous welding nodes in the welding process, the process status of the node (stability of welding current and voltage, weld formation status) is correlated with the quality assessment results (assessment results of weld formation index and stress index). If the process status is normal, the quality assessment result meets the standard and there are no weld defects; if the process status is abnormal (welding voltage fluctuation), the quality assessment result does not meet the standard and there are defects such as weld protrusion and excessive stress. The abnormal correlation between the two is clarified, and tower process status-quality correlation characteristic data is generated.

[0071] Furthermore, the tower processing quality evaluation index data mentioned in step S22 includes tower processing geometric quality evaluation index data, tower processing thermal behavior evaluation index data, and tower processing stress evaluation index data.

[0072] Furthermore, step S3 includes the following steps:

[0073] Step S31: Analyze the tower process deviation behavior for quality risk based on the tower process status-quality correlation characteristic data, and generate tower process deviation behavior data;

[0074] In this embodiment of the invention, the process is carried out sequentially through multiple process stages. Within each stage, nodes are investigated according to stage type. The core objective is to identify process behaviors that deviate from the standard range and are clearly associated with quality defects. This involves accurately defining the specific manifestations, timing, and corresponding quality risks of these deviations, providing a clear basis for subsequent risk evolution management. This aligns with the core requirement of proactively preventing quality risks in intelligent adjustment methods for tower manufacturing process parameters. For each node, the standard range of the process state is first retrieved (based on the process parameters and operating condition standards preset in S1 and S2). The actual process state characteristic data of that node is compared to identify abnormal process behaviors whose process parameter values, fluctuation characteristics, and operating condition response exceed the standard range. Simultaneously, combined with the quality assessment data of that node, it is determined whether the abnormal process behaviors lead to quality defects or a decline in quality level. Only abnormal process behaviors directly related to quality risks are retained, while minor fluctuations in process parameters (behaviors not exceeding the normal fluctuation range in the associated characteristic data) that have no quality impact are eliminated. The analysis focuses on the following process deviations: In the rolling stage, deviations in rolling angle and rolling force from standard values ​​lead to excessive roundness and uneven blank deformation. In the welding stage, deviations in welding current and voltage from standard values ​​lead to weld formation defects and excessive stress. In the rounding stage, deviations in rounding pressure from standard values ​​lead to excessive roundness error and excessive residual stress. In the flaw detection stage, deviations in flaw detection frequency and depth from standard values ​​lead to missed or false defects. For each identified process deviation, the corresponding process stage, node identifier, deviation parameter name, deviation direction (higher or lower than standard), specific timing of the deviation, corresponding quality defect type, and risk level are labeled. All deviation information is integrated according to the processing sequence to generate tower process deviation data, comprehensively presenting all process deviation details related to quality risk throughout the entire processing flow.

[0075] Step S32: Based on the tower process deviation behavior data, perform tower processing quality risk evolution processing for single-stage nodes and coupled-stage nodes, and generate tower single quality risk evolution data and tower coupled quality risk evolution data respectively;

[0076] In this embodiment of the invention, the quality risk evolution processing of a single-stage node is carried out sequentially across multiple process stages. For the process deviation behavior occurring at a single node within each stage, the entire process of the deviation behavior from its occurrence and development to stabilization (or termination) is tracked. The changes in the degree of deviation behavior at different time sequences and the corresponding trends in quality defects are recorded, clarifying the quality risk evolution law when the deviation behavior exists alone. If the rolling force deviates from the standard at the rolling forming node in the rolling stage, the change in the magnitude of the force deviation with the processing sequence is tracked, and the corresponding changes in the rolling roundness error are recorded, clarifying the evolution process of a single deviation behavior leading to the gradual expansion or stabilization of quality risk. If the welding voltage deviates from the standard at a continuous welding node in the welding stage, the evolutionary correlation between the voltage deviation and weld defects (such as undercut and protrusion) is tracked, and the specific time sequence and degree of defect expansion from its occurrence are recorded. The quality risk evolution processing of coupled-stage nodes focuses on the linkage effect between nodes in different process stages, focusing on analyzing the transmission effect of the process deviation behavior of the previous stage node on the process state and quality of the next stage node, and identifying cross-stage coupled deviation behaviors. Deviating from the standard roundness of the rolling forming node during the rolling stage will lead to positioning deviation of the welding torch positioning node during the welding stage, thereby causing weld formation defects. This study traces the transmission process of such cross-stage deviations, recording the correlation between the degree of deviation in the previous stage and the degree of deviation in the next stage, as well as the quality risk. Similarly, deviations from the standard roundness error during the rounding stage will affect the detection accuracy during the flaw detection stage, leading to an increased risk of missed defects. The study comprehensively records the risk evolution process under this coupled correlation. The risk evolution records of single-stage nodes and coupled-stage nodes are integrated separately, annotating the evolution sequence, changes in deviation degree, changes in quality risk, and the evolution endpoint state. Finally, single-stage quality risk evolution data and coupled-stage quality risk evolution data for the tower are generated respectively.

[0077] Step S33: Based on the single quality risk evolution data of the tower, perform tower process deviation behavior evolution characteristic analysis on the tower process deviation behavior data to generate tower process deviation behavior evolution characteristic data;

[0078] In this embodiment of the invention, process deviations are categorized by deviation parameter type. The rolling stage is divided into rolling angle deviation and rolling force deviation, while the welding stage is divided into welding current deviation and welding voltage deviation. Each deviation type is analyzed separately. For each deviation, its corresponding single quality risk evolution data is retrieved, tracking the entire process from occurrence to stabilization (or termination), and extracting evolution characteristics: the evolution rate is calculated by the change in deviation amplitude at different time intervals to determine whether the deviation is rapid or slow; the evolution trend is divided into three types: continuous expansion, gradual reduction, and stabilization, determined in conjunction with the quality risk evolution; the evolution endpoint characteristics include whether the deviation has terminated, the deviation amplitude at termination, and the corresponding quality risk state (whether irreversible defects have formed). Simultaneously, the linkage between the evolution of deviation behavior and the evolution of single quality risks is analyzed, clarifying the corresponding ratio between changes in deviation amplitude and changes in quality risk level. When the rolling force deviates rapidly, the quality risk (roundness exceeding the standard) also expands rapidly; when the deviation is slow, the quality risk gradually increases. The evolution rate, trend, endpoint characteristics, and correlation with quality risk of each process deviation behavior are recorded in detail. Corresponding process stages, node identifiers, and deviation parameters are labeled to distinguish the evolutionary differences of different deviation types. For example, rolling angle deviations tend to exhibit a slow evolutionary trend, while welding current deviations tend to exhibit a rapid evolutionary trend. All evolutionary characteristic information of deviation behaviors is integrated and sorted by process stage and deviation type to finally generate tower process deviation behavior evolution characteristic data, fully presenting the evolutionary pattern of each process deviation behavior and providing clear support for subsequent propagation and evolution analysis.

[0079] Step S34: Analyze the evolution characteristics of tower processing quality risk based on the single quality risk evolution data and the coupled quality risk evolution data of the tower, and generate tower processing quality risk evolution characteristic data;

[0080] In this embodiment of the invention, the evolutionary characteristics of single quality risks and coupled quality risks are extracted separately and then compared and analyzed. The evolutionary characteristics of single quality risks focus on analyzing the independence of risk occurrence, evolution rate, risk peak, and attenuation law. Quality risks at a single node often occur independently, with an evolution rate consistent with the evolution rate of the corresponding process deviation behavior. The risk peak appears in the sequence where the deviation behavior is most severe, and some risks can attenuate as the deviation behavior ends. The evolutionary characteristics of coupled quality risks focus on analyzing the transmissibility, superposition, and cross-stage impact range of risks. Coupled risks are transmitted from the previous stage to the next stage, superimposed with the risks of the next stage itself, leading to a significant increase in risk level. The impact range covers multiple process stages, with a faster evolution rate, higher risk peak, and greater difficulty in attenuation. Furthermore, according to the classification and analysis of quality risk types, the evolutionary characteristics of geometric quality risks (roundness exceeding standards, weld formation defects), thermal behavior risks (excessive heat-affected zone), and stress risks (excessive residual stress) differ. Geometric quality risks often increase with the continuous expansion of process deviation behavior, while stress risks often remain stable after the deviation behavior ends, making rapid attenuation difficult. The evolution rate, trend, peak value, decay pattern, transmission mechanism, and scope of influence of all quality risks are recorded in detail to generate tower processing quality risk evolution characteristic data.

[0081] Step S35: Analyze the propagation and evolution characteristics of tower process deviation behavior using tower processing quality risk evolution characteristic data, and generate tower process deviation behavior propagation and evolution characteristic data.

[0082] In this embodiment of the invention, the process is carried out sequentially according to the linkage of each stage, focusing on the linkage relationship of the process from rolling to welding to rounding to flaw detection. The propagation possibility and process of each process deviation are analyzed one by one. For propagable deviations, the propagation starting point (the node where the initial deviation occurs), the propagation path (which subsequent nodes are propagated from the starting point), the propagation rate (the time span from the starting point to subsequent nodes), and the degree of propagation attenuation (the change in the deviation amplitude during propagation) are identified. Simultaneously, combined with quality risk evolution characteristic data, the impact of deviation propagation on the quality risk of subsequent nodes is analyzed, clarifying the changing relationship between deviation and quality risk after propagation. For example, in the rolling stage, deviation in the roundness of the rolling forming node propagates along the path from the rolling forming node to the welding torch positioning node to the continuous welding node. The propagation rate is synchronously transmitted with the processing sequence, and the deviation amplitude does not significantly attenuate during propagation, leading to a significant increase in the risk of weld formation defects in subsequent welding nodes. Similarly, in the welding stage, deviation in the weld stress of the continuous welding node propagates along the path from the continuous welding node to the initial rounding node. The deviation amplitude slightly attenuates during propagation, but it still leads to an increase in the risk of excessive residual stress during the rounding stage. For non-propagating deviations (affecting only the node itself and not spreading to other nodes), their limitations and non-propagation characteristics are marked. Detailed records are kept of the propagation origin, path, rate, attenuation, and impact of each deviation, along with the corresponding process stage, node identifier, and associated quality risks. All propagation evolution information is integrated to generate tower process deviation propagation evolution characteristic data.

[0083] Step S36: Perform correlation analysis on the propagation and evolution characteristics of the tower process deviation behavior, and generate correlation data of process parameters and multi-factor characteristics of process deviation behavior.

[0084] In this embodiment of the invention, the process is categorized by process parameter type, focusing on core process parameters such as rolling angle, rolling force, welding current, welding voltage, rounding pressure, and flaw detection frequency. Correlation analysis is performed on each parameter individually. For each process parameter, the parameter's value change data and fluctuation characteristic data are extracted, and compared with the corresponding process deviation behavior data and propagation evolution data to determine the correlation between the parameter and the deviation behavior: if the parameter value exceeds the standard range, the corresponding deviation behavior will inevitably occur, and the greater the parameter deviation, the greater the process deviation and the wider the propagation range, then the parameter is determined to be highly correlated with the deviation behavior; if the parameter value change and the occurrence of deviation behavior are not significantly related, then there is no correlation. The rolling force parameter is highly correlated with the rolling roundness deviation behavior; the greater the deviation of the rolling force value from the standard, the greater the deviation in rolling roundness, and the higher the possibility of propagation to the welding stage. The welding voltage parameter is highly correlated with the weld formation deviation behavior; voltage deviation from the standard will directly lead to deviations in weld width and height, and is easily propagated to the rounding stage. Simultaneously, the correlation between multi-parameter coupling and deviation behavior was analyzed. When two or more parameters deviate from the standard simultaneously, the propagation rate and scope of the deviation behavior significantly expand. When welding current and voltage deviate simultaneously, the propagation speed of weld defect deviation behavior is faster, and the superposition of quality risks is more obvious. Detailed records were kept of the correlation degree, correlation patterns, and correspondence between parameter deviation magnitude and deviation impact for each process parameter and deviation behavior, generating multi-factor characteristic correlation data of process parameters and deviation behavior.

[0085] Step S37: Based on the multi-factor characteristic correlation data of tower process parameters and deviation behavior, perform a characteristic analysis of the influence of tower process parameter deviation, and generate characteristic data of tower process parameter deviation influence.

[0086] In this embodiment of the invention, the process parameters are categorized by type, and each parameter is analyzed according to its deviation direction (above standard, below standard) and deviation magnitude to clarify the specific impact of different deviation scenarios. When the rolling force parameter is above the standard, it leads to an increased deviation in rolling roundness, which propagates rapidly to the welding stage, causing weld formation defects and increasing the quality risk level. Furthermore, the impact of this type of deviation on subsequent processes is difficult to eliminate quickly. When the rolling force is below the standard, the deviation in rolling roundness is smaller, and the propagation is limited to the rolling stage, resulting in a lower quality risk level, which can be compensated for by subsequent rounding processes. When the welding current parameter is above the standard, it leads to excessive weld stress, which propagates to the rounding stage, resulting in excessive residual stress, affecting the stability of the tower structure, and causing irreversible quality risks. When the welding current is below the standard, it leads to incomplete weld penetration defects, with no significant propagation of the deviation, which can be eliminated through repair welding. When the rounding pressure parameter deviates from the standard, it directly leads to excessive roundness error, which propagates to the flaw detection stage, affecting the accuracy of defect detection. The greater the deviation, the greater the detection error. For each deviation of a process parameter, detailed records are made of the scope of impact (single node, single stage, cross-stage), degree of impact (slight, moderate, severe), mechanism of impact (direct impact, indirect propagation impact), and degree of controllability (compensable, irreversible). The corresponding quality risk type and associated deviation behavior are marked, and all parameter deviation impact information is integrated, sorted by process stage and parameter type, and finally generated into tower process parameter deviation impact characteristic data. This fully presents the specific impact of each process parameter deviation, providing clear support for the formulation of subsequent intelligent adjustment strategies.

[0087] Furthermore, step S32 includes the following steps:

[0088] A single-stage node tower quality risk characteristic analysis is performed on the tower process deviation behavior data to generate single-stage tower quality risk characteristic data. Based on the single-stage tower quality risk characteristic data, the tower processing quality risk evolution is processed at the single-stage node to generate single-stage tower quality risk evolution data.

[0089] The tower quality risk characteristics of the coupled stage nodes are analyzed by the tower process deviation behavior data to generate tower coupled quality risk characteristic data. Based on the tower coupled quality risk characteristic data, the tower processing quality risk evolution of the coupled stage nodes is processed to generate tower coupled quality risk evolution data.

[0090] In this embodiment of the invention, the quality risk characteristic analysis of a single-stage node is carried out step by step according to multiple process stages such as rolling, welding, rounding, and flaw detection. It focuses on the independent process deviation behavior of a single node, without involving cross-node or cross-stage linkages, ensuring accurate analysis. For each single node in each stage, the corresponding process deviation behavior data is extracted, clarifying the deviation parameters, direction, magnitude, and timing. The quality risk characteristics caused by the deviation behavior are analyzed by comparing them with the quality evaluation index standards for that node. In the rolling stage, at the rolling forming node, deviations in rolling force from the standard will cause excessive roundness error, manifesting as irregular shape and uneven wall thickness of the rolled part, with the impact limited to the rolling stage. In the welding stage, at the continuous welding node, deviations in welding voltage from the standard will cause weld formation defects, manifesting as uneven weld width, undercut, or protrusions on the surface, with the impact limited to the welding stage. The quality risk characteristics of each single node are recorded in detail, annotating the corresponding process stage, node identifier, deviation details, risk manifestations, and impact range, and integrated to generate single-quality risk characteristic data for the tower. Subsequently, risk evolution processing was conducted. Based on single quality risk characteristic data, the entire process of deviation behavior at each node was tracked from occurrence and development to stabilization (or termination). Changes in risk characteristics were recorded chronologically, clarifying the evolution rate, trend, and endpoint state: the risk of roundness exceeding the standard caused by deviation in rolling force gradually increased as the deviation continued, and tended to stabilize after the deviation terminated; the risk of weld formation defects caused by deviation in welding voltage rapidly emerged, and the risk level increased with the deviation magnitude. All evolution records were integrated to generate single quality risk evolution data for the tower. The analysis of quality risk characteristics at coupled stage nodes focused on the linkage relationship between nodes in different process stages, analyzing the transmission effect of deviation behavior in the previous stage on nodes in the next stage. Coupled risks have transmission and superposition characteristics, distinguishing them from the independence of single node risks. The analysis was conducted according to the linkage sequence of process stages, focusing on the coupling relationship between rolling and welding, welding and rounding, and rounding and flaw detection. Deviation behavior and risk characteristics of the previous stage were extracted, and their impact on related nodes in the next stage was analyzed to clarify the characteristics of coupled risks. If the roundness of the forming node deviates from the standard during the rolling stage, it will be transmitted to the welding torch positioning node during the welding stage, causing welding torch positioning deviation and weld formation defects. The coupled risk is the superposition of roundness exceeding the standard and weld formation defects, affecting the coverage of nodes in both stages. If the roundness of the roundness detection node deviates during the roundness correction stage, it will be transmitted to the surface detection node during the flaw detection stage, causing detection deviation and missed defect detection. The coupled risk is the superposition of roundness not meeting the standard and missed defect detection, affecting the coverage of the roundness correction and flaw detection stages. The risk characteristics of each coupled node are recorded in detail, and the combination of coupled nodes, transmission path, deviations and risk characteristics of the preceding and following stages, and superposition effects are marked. These are integrated to generate tower coupling quality risk characteristic data. Subsequently, risk evolution processing is carried out to track the entire process of coupled risk from occurrence and transmission to stabilization (or termination), recording the temporal changes, deviation magnitude and risk degree, clarifying the transmission rate, evolution trend, superposition law and endpoint state, and generating tower coupling quality risk evolution data.

[0091] 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 S4 is shown in this embodiment. Step S4 includes:

[0092] Step S41: Analyze the optimal process parameters for tower processing based on the tower process status-quality correlation characteristic data, and generate optimal process parameter data for tower processing;

[0093] In this embodiment of the invention, the process is carried out layer by layer according to process stages and node types. Each stage uses the correlation pattern of "normal process status - quality meets standards" in the associated feature data as the core screening criterion, extracting the value range and specific values ​​of the process parameters corresponding to this correlation pattern. For the rolling stage, the focus is on analyzing the correlation data between the process status and quality of the rolling forming node. The value ranges of parameters such as rolling angle, rolling force, and rolling speed are extracted when the rolling roundness meets the standard and the blank deformation is uniform. Combined with the optimal parameter records under the same working conditions in the prior data, the optimal values ​​of the rolling angle, rolling force, and rolling speed are determined. It is clarified that the optimal parameters must ensure that the blank wall thickness is uniform after rolling, the roundness meets the standard, and there are no obvious deformation defects. For the welding stage, focusing on continuous welding nodes, the optimal values ​​for parameters such as welding current, welding voltage, and welding temperature are extracted to ensure weld formation meets standards, stress distribution is uniform, and the heat-affected zone meets requirements. Combined with the parameter records showing optimal welding quality from prior data, the optimal values ​​for each welding parameter are determined to ensure that the welds corresponding to the optimal parameters are free of undercut and protrusions, have uniform width and height, and stress values ​​are controlled within the standard range. In the rounding stage, the focus is on analyzing rounding operation nodes, extracting the optimal values ​​for parameters such as rounding pressure and the number of rounding cycles to ensure roundness error meets standards and stress residue is minimized, thus determining the optimal parameters. In the flaw detection stage, surface and internal detection nodes are analyzed, extracting the optimal values ​​for flaw detection frequency and depth to ensure accurate defect detection with no missed or false detections. Detailed records of the optimal process parameters for each stage and node are kept, labeling the corresponding process stage, node identifier, parameter name, optimal value, and corresponding quality assurance requirements, generating optimal process parameter data for tower processing.

[0094] Step S42: Based on the tower processing optimization process parameter data, perform adjustment boundary feature analysis of the tower optimization process parameters to generate tower optimization process parameter adjustment boundary feature data;

[0095] In this embodiment of the invention, the process is categorized by process parameter type, and each parameter undergoes individual boundary analysis. Combining parameter characteristics, equipment operating limits, and quality constraints, the upper and lower limits and constraints of the adjustment boundary are clearly defined. For the optimal rolling angle parameter, the upper and lower adjustment boundaries are determined based on the mechanical adjustment limits of the rolling equipment and the rolling roundness quality evaluation indicators. The adjustment boundaries must satisfy the following: after parameter adjustment, the rolling roundness still meets the standard and does not exceed the mechanical adjustment range of the equipment. Furthermore, based on the parameter deviation impact characteristic data, it is clear that when the rolling angle exceeds the adjustment boundary, it will lead to severely excessive rolling roundness that cannot be compensated for by subsequent processes; therefore, the adjustment boundary cannot be exceeded. For the optimal welding current parameter, the upper and lower adjustment boundaries are determined based on the rated current range of the welding equipment, weld stress, and forming quality evaluation indicators. The adjustment boundaries must satisfy the following: after welding current adjustment, the weld stress does not exceed the standard, the forming is defect-free, and it does not exceed the rated operating range of the equipment. Furthermore, based on the parameter deviation impact characteristic data, it is clear that exceeding the adjustment boundary will cause incomplete weld penetration or burn-through defects, and these defects are irreversible. For each optimized process parameter, detailed records are kept of the upper and lower adjustment boundaries, boundary constraints (equipment constraints, quality constraints), and quality risks exceeding the boundaries. The corresponding process stage, node identifier, and parameter name are labeled, distinguishing the boundary differences between different parameters. For example, the rolling force adjustment boundary is more strictly constrained by equipment load, while the welding voltage adjustment boundary is more strictly constrained by quality. All parameter adjustment boundary information is integrated to generate characteristic data of the tower's optimized process parameter adjustment boundaries.

[0096] Step S43: Perform a weight analysis of tower process parameter adjustment based on the influence characteristic data of tower process parameter deviation, and generate tower process parameter adjustment weight data;

[0097] In this embodiment of the invention, the importance of adjusting each process parameter is determined, and the priority of different parameters in the adjustment process is clarified. This ensures that subsequent adjustment strategies prioritize parameters that have a significant impact on quality and whose deviations pose serious risks, thereby improving adjustment efficiency and accuracy. This aligns with the core requirements of precise adjustment and key control in intelligent adjustment methods for process parameters in tower manufacturing. The criteria for determining the adjustment weight are the degree of impact of parameter deviation on quality, the scope of deviation propagation, the irreversibility of risk, and the difficulty of adjustment. The greater the impact, the wider the propagation, the irreversible risk, and the lower the adjustment difficulty, the higher the adjustment weight; conversely, the lower the adjustment weight. The analysis process is carried out step by step according to multiple process stages such as rolling, welding, rounding, and flaw detection. Within each stage, all core process parameters are ranked by weight. For the rolling stage, the analysis focuses on three core parameters: rolling angle, rolling force, and rolling speed. Based on the characteristic data of parameter deviations, deviations in rolling force lead to excessive roundness, and this deviation easily propagates to the welding stage. The risk is wide-ranging and irreversible, and adjustment is relatively easy; therefore, it is assigned the highest adjustment weight. Deviations in rolling angle have a relatively small impact on quality and are limited to the rolling stage; adjustment is of medium difficulty, and it is assigned a medium adjustment weight. Deviations in rolling speed have the least impact on quality and are relatively easy to adjust, and are assigned the lowest adjustment weight. For the welding stage, the analysis focuses on three core parameters: welding current, welding voltage, and welding temperature. Deviations in welding current cause excessive weld stress and forming defects; the risk is irreversible and propagates widely; adjustment is of medium difficulty, and it is assigned the highest adjustment weight. Deviations in welding voltage have a significant impact on weld formation, but the propagation range is limited to the welding stage; adjustment is relatively easy, and it is assigned a medium adjustment weight. Deviations in welding temperature have a relatively small impact and can be adjusted by auxiliary cooling or heating; adjustment is relatively easy, and it is assigned a low adjustment weight. For each process parameter, the adjustment weight value and the basis for weight determination (degree of impact, scope of propagation, irreversibility of risk, difficulty of adjustment) are calculated in detail. The corresponding process stage, node identifier, and parameter name are marked. The weight ranking and priority are clarified, and the weight differences of parameters in different stages are distinguished. Finally, the adjustment weight data of tower process parameters is generated.

[0098] Step S44: Perform a feasible domain analysis of the deviation adjustment of the tower's optimal process parameters by analyzing the deviation influence feature data of the tower process parameters, and generate feasible domain data of the deviation adjustment of the tower process parameters.

[0099] In this embodiment of the invention, the focus is on multi-parameter coupling deviation scenarios, conducted step by step according to the process stage. Within each stage, the analysis focuses on parameter combinations with interrelationships. In the rolling stage, the focus is on the coupling between rolling angle and rolling force; in the welding stage, the focus is on the coupling between welding current and welding voltage; and in the rounding stage, the focus is on the coupling between rounding pressure and the number of rounding cycles. For each parameter combination, the adjustment boundary characteristic data of each parameter in the combination is first extracted to clarify the adjustable range of individual parameters. Then, combined with the parameter deviation influence characteristic data, the linkage influence between parameters is analyzed to clarify the influence law of one parameter adjustment on another, thereby determining the feasible domain for coupling adjustment of the parameter combination. In the rolling stage, the coupling analysis of rolling angle and rolling force is performed. Adjusting the rolling force affects the actual value of the rolling angle. When the rolling force increases, the rolling angle needs to be simultaneously fine-tuned to decrease to ensure roundness meets the standard. Combining the adjustment boundaries and linkage law of both, the feasible domain for coupling adjustment of the two is determined, clarifying the corresponding adjustable range of the rolling angle when the rolling force changes within the adjustment boundary, ensuring that neither exceeds its respective boundary after adjustment, and that roundness meets the standard. The welding current and welding voltage coupling analysis during the welding stage reveals that as the welding current increases, the welding voltage needs to be increased simultaneously with slight adjustments to prevent burn-through or incomplete penetration of the weld. Combining the adjustment boundaries and linkage effects of both, the feasible region for coupling adjustment is determined, and the corresponding adjustment ranges for current and voltage are clarified to ensure that both meet the boundary requirements after adjustment and that the weld quality meets the standards. For each process stage's parameter coupling combination, detailed records are kept of the coupled parameter names, individual parameter adjustment boundaries, linkage patterns between parameters, feasible region range for coupling adjustment, and feasible region constraints. Corresponding process stages and node identifiers are marked, and all feasible region information for coupling adjustment is integrated to generate multi-source process parameter coupling deviation adjustment feasible region data.

[0100] Step S45: Based on the influence characteristic data of the deviation of the tower process parameters and the feasible domain data of the multi-source process parameter coupling deviation adjustment, analyze the intelligent adjustment strategy of the tower optimization process parameters and generate the intelligent adjustment strategy of the tower optimization process parameters.

[0101] In this embodiment of the invention, the process is carried out step by step according to the process stages. Within each stage, adjustment strategies are formulated for different parameter deviation scenarios (single parameter deviation, multi-parameter coupled deviation). For single parameter deviation scenarios, based on adjustment weight data, parameters with higher weights are adjusted first. According to the parameter deviation impact characteristic data, the direction of parameter deviation (above or below the standard) and the deviation magnitude are determined. Combined with adjustment boundary characteristic data, the adjustment direction (opposite to the deviation direction) and adjustment magnitude are determined to ensure that the adjusted parameter returns to the optimal range and does not exceed the adjustment boundary. In the winding stage, if the winding force deviates from the standard (above the standard), based on the adjustment weight (highest), the adjustment direction is determined to be reducing the winding force. Combined with the deviation magnitude and adjustment boundary, the adjustment magnitude is determined, and the adjustment sequence is determined to be real-time adjustment. After adjustment, the winding force returns to the optimal value and is not lower than the lower adjustment boundary. Simultaneously, roundness changes are tracked to ensure quality meets standards. For multi-parameter coupled deviation scenarios, based on the coupled adjustment feasible region data, the adjustment order (sorted by weight), adjustment direction, and adjustment magnitude of the coupled parameters are determined to ensure that each parameter changes within the feasible region during the adjustment process and synchronously returns to the optimal range. During the welding stage, both welding current and voltage deviate simultaneously (both exceeding the standard). Based on adjustment weights, the welding current is adjusted first, decreasing it, followed by simultaneous adjustment of the welding voltage, also decreasing it. Combining the coupling feasible region and the deviation magnitude, the adjustment range for both is determined, and the adjustment sequence is clarified as current adjustment first, followed by simultaneous fine-tuning of the voltage. This ensures that both return to their optimal range after adjustment, and that the weld quality meets the standards. For each process stage and each deviation scenario, detailed records are kept of the adjustment parameters, adjustment sequence, adjustment direction, adjustment magnitude, adjustment timing, and post-adjustment quality verification standards. Corresponding node identifiers and deviation scenario details are marked, and precautions during the adjustment process are clearly defined to avoid over- or under-adjustment. The adjustment schemes for all adjustment scenarios are integrated, sorted by process stage and deviation type, forming a complete adjustment rule system. This generates an intelligent adjustment strategy for the optimal process parameters of the tower, ensuring that the strategy can directly guide subsequent process parameter adjustment operations, balancing adjustment accuracy, feasibility, and efficiency.

[0102] Step S46: Execute intelligent adjustment of tower processing parameters using the intelligent adjustment strategy for tower optimization process parameters.

[0103] In this embodiment of the invention, the adjustment operation is carried out simultaneously in multiple process stages such as rolling, welding, rounding, and flaw detection. Each stage has a dedicated adjustment module that works in conjunction with the processing equipment to receive real-time process status characteristic data, monitor real-time values ​​of process parameters, compare them with optimized process parameter data, and determine whether there are parameter deviations and deviation scenarios (single parameter deviation or multi-parameter coupled deviation). When a parameter deviation is detected, the adjustment scheme corresponding to the deviation scenario in the intelligent adjustment strategy is immediately retrieved, specifying the adjustment parameters, direction, magnitude, and timing. The adjustment module sends an adjustment command to the processing equipment, which automatically executes the parameter adjustment according to the command without manual intervention. In the rolling stage, the adjustment module monitors the real-time values ​​of the rolling angle, rolling force, and rolling roller speed. When the rolling force is detected to be higher than the optimized parameter and exceeds the allowable fluctuation range, the corresponding adjustment scheme is immediately retrieved, and a command to reduce the rolling force is sent to the rolling equipment. The adjustment magnitude is executed according to the strategy requirements. During the adjustment process, changes in rolling force and rolling roundness are monitored in real-time until the rolling force returns to the optimized range and the roundness meets the standard. After adjustment, the adjustment time, magnitude, and effect are recorded. The welding stage adjustment module monitors welding current, welding voltage, and welding temperature in real time. When both welding current and voltage deviate simultaneously, it sends adjustment commands to the welding equipment according to the required adjustment sequence and magnitude, synchronously fine-tuning the current and voltage. During the adjustment process, it monitors the weld formation state and stress changes in real time to ensure that the weld quality meets the standards after adjustment. During the adjustment operation, the adjustment parameters, adjustment commands, adjustment magnitude, adjustment time, and adjustment effect of each adjustment link are recorded in real time. The stability of the process parameters after adjustment is monitored. If the parameters deviate again, the corresponding adjustment plan is immediately executed again to ensure that the process parameters remain within the optimal range. At the same time, the entire adjustment operation is recorded, marking the corresponding process stage, node identification, deviation, adjustment plan, and adjustment effect. All adjustment records are integrated to form an adjustment operation report, completing the intelligent adjustment of the tower processing process parameters and ensuring process stability and quality compliance during the processing.

[0104] 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.

[0105] 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 intelligent adjustment of process parameters in tower manufacturing, characterized in that, Includes the following steps: Step S1: Monitor and process multi-source tower processing data using tower processing monitoring sensors to generate multi-source tower processing data; perform tower processing process state characteristic analysis based on the multi-source tower processing data to generate tower processing process state characteristic data. Step S2: Perform tower process-quality correlation feature analysis based on tower processing technology status feature data to generate tower process status-quality correlation feature data; Step S2 includes the following steps: Step S21: Obtain prior data for tower processing quality assessment; Step S22: Analyze the tower processing quality evaluation index on the prior data of tower processing quality assessment, and generate tower processing quality evaluation index data, wherein the tower processing quality evaluation index data includes tower processing geometric quality evaluation index data, tower processing thermal behavior evaluation index data, and tower processing stress evaluation index data. Step S23: Quantify the tower processing quality fluctuation of each stage type on the tower processing quality evaluation index data to generate stage tower processing quality fluctuation quantitative data; Step S24: Perform staged fitting processing of tower processing quality fluctuation quantification based on the stage tower processing quality fluctuation quantification data to generate fitted tower processing quality fluctuation quantification data; Step S25: Based on the fitted tower processing quality fluctuation quantification data, characterize the tower processing and establish the mapping relationship for the global quality assessment to obtain the tower processing quality assessment model; Step S26: Collect tower processing characterization data based on tower processing process state characteristic data to obtain tower processing characterization data; Step S27: Transmit the tower processing characterization data to the tower processing quality assessment model for tower processing quality assessment and generate tower processing quality assessment data; Step S28: Based on the tower processing technology status characteristic data and tower processing quality assessment data, perform tower process status-quality correlation characteristic analysis to generate tower process status-quality correlation characteristic data; Step S3: Based on the tower process state-quality correlation characteristic data, perform tower process deviation behavior propagation and evolution characteristic analysis to generate tower process deviation behavior propagation and evolution characteristic data; through the tower process deviation behavior propagation and evolution characteristic data, perform tower process parameter deviation influence characteristic analysis to generate tower process parameter deviation influence characteristic data. Step S3 includes the following steps: Step S31: Analyze the tower process deviation behavior for quality risk based on the tower process status-quality correlation characteristic data, and generate tower process deviation behavior data; Step S32: Based on the tower process deviation behavior data, perform tower processing quality risk evolution processing for single-stage nodes and coupled-stage nodes, and generate tower single quality risk evolution data and tower coupled quality risk evolution data respectively; Step S33: Based on the single quality risk evolution data of the tower, perform tower process deviation behavior evolution characteristic analysis on the tower process deviation behavior data to generate tower process deviation behavior evolution characteristic data; Step S34: Analyze the evolution characteristics of tower processing quality risk based on the single quality risk evolution data and the coupled quality risk evolution data of the tower, and generate tower processing quality risk evolution characteristic data; Step S35: Analyze the propagation and evolution characteristics of tower process deviation behavior using tower processing quality risk evolution characteristic data, and generate tower process deviation behavior propagation and evolution characteristic data. Step S36: Perform correlation analysis on the propagation and evolution characteristics of the tower process deviation behavior, and generate correlation data of process parameters and multi-factor characteristics of process deviation behavior. Step S37: Based on the multi-factor characteristic correlation data of tower process parameters and deviation behavior, perform a characteristic analysis of the influence of tower process parameter deviation, and generate characteristic data of tower process parameter deviation influence. Step S4: Based on the tower process state-quality correlation characteristic data and the tower process parameter deviation impact characteristic data, analyze the tower optimization process parameter intelligent adjustment strategy and generate the tower optimization process parameter intelligent adjustment strategy; execute the tower processing process parameter intelligent adjustment operation through the tower optimization process parameter intelligent adjustment strategy.

2. The intelligent adjustment method for process parameters in tower manufacturing according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: Monitor and process multi-source tower processing data using tower processing monitoring sensors to generate multi-source tower processing data; Step S12: Perform analytical processing of tower foundation process parameters based on multi-source tower processing data to generate analytical data of tower foundation process parameters; Step S13: Perform process stage type classification on the analytical data of tower foundation process parameters to generate tower foundation process stage type data; Step S14: Based on the tower foundation process stage type data, collect tower processing conditions for each process stage type from the multi-source processing data of the tower to generate tower processing condition data; Step S15: Based on the analytical data of the tower foundation process parameters and the tower processing condition data, perform a tower processing process state characteristic analysis to generate tower processing process state characteristic data.

3. The intelligent adjustment method for process parameters in tower manufacturing according to claim 2, characterized in that, Step S15 includes the following steps: Step S151: Based on the tower foundation process stage type data and tower foundation process parameter analysis data, perform tower process behavior characteristic analysis and processing to generate tower process behavior characteristic data; Step S152: Perform tower processing condition response characteristic analysis on the tower processing condition data to generate tower processing condition response characteristic data; Step S153: Analyze the tower processing process status characteristics of each stage type node based on the tower process behavior characteristic data and the tower processing condition response characteristic data, and generate tower processing process status characteristic data.

4. The intelligent adjustment method for process parameters in tower manufacturing according to claim 3, characterized in that, Step S151 includes the following steps: The data on the process stages of the tower foundation are processed to identify the behavioral units of the tower process stages, thereby generating behavioral unit data for the tower process stages. Based on the behavioral unit data of the tower process stage, analyze the behavioral patterns of the tower process stage to generate tower process behavioral pattern data. By analyzing the analytical data of the tower foundation process parameters using tower process behavior pattern data, tower process behavior characteristic data is generated.

5. The intelligent adjustment method for process parameters in tower manufacturing according to claim 3, characterized in that, Step S153 includes the following steps: Based on the tower processing condition response characteristic data, tower processing condition status analysis is performed to generate tower processing condition status data. The stage type node of the tower processing process state is designed by using the stage type data of the tower foundation process, and the node data of the tower processing process state is obtained. Based on the tower process behavior characteristic data and tower processing condition status data, the process behavior and condition response status of tower processing are correlated to generate tower process behavior-condition response status correlation data. The data related to the tower process behavior and operating condition response status is transmitted to the tower processing process status node data for tower processing process status feature analysis, generating tower processing process status feature data.

6. The intelligent adjustment method for process parameters in tower manufacturing according to claim 1, characterized in that, Step S32 includes the following steps: A single-stage node tower quality risk characteristic analysis is performed on the tower process deviation behavior data to generate single-stage tower quality risk characteristic data. Based on the single-stage tower quality risk characteristic data, the tower processing quality risk evolution is processed at the single-stage node to generate single-stage tower quality risk evolution data. The tower quality risk characteristics of the coupled stage nodes are analyzed by the tower process deviation behavior data to generate tower coupled quality risk characteristic data. Based on the tower coupled quality risk characteristic data, the tower processing quality risk evolution of the coupled stage nodes is processed to generate tower coupled quality risk evolution data.

7. The intelligent adjustment method for process parameters in tower manufacturing according to claim 1, characterized in that, Step S4 includes the following steps: Step S41: Analyze the optimal process parameters for tower processing based on the tower process status-quality correlation characteristic data, and generate optimal process parameter data for tower processing; Step S42: Based on the tower processing optimization process parameter data, perform adjustment boundary feature analysis of the tower optimization process parameters to generate tower optimization process parameter adjustment boundary feature data; Step S43: Perform a weight analysis of tower process parameter adjustment based on the influence characteristic data of tower process parameter deviation, and generate tower process parameter adjustment weight data; Step S44: Perform a feasible domain analysis of the deviation adjustment of the tower's optimal process parameters by analyzing the deviation influence feature data of the tower process parameters, and generate feasible domain data of the deviation adjustment of the tower process parameters. Step S45: Based on the influence characteristic data of the deviation of the tower process parameters and the feasible domain data of the multi-source process parameter coupling deviation adjustment, analyze the intelligent adjustment strategy of the tower optimization process parameters and generate the intelligent adjustment strategy of the tower optimization process parameters. Step S46: Execute intelligent adjustment of tower processing parameters using the intelligent adjustment strategy for tower optimization process parameters.

Images