Aluminum plate installation directionality deformation real-time diagnosis system and vector control method

Abstract

The application discloses a kind of aluminum plate installation directionality deformation real-time diagnosis system and vector control method, it is related to the technical field of industrial control, wherein the system includes multi-source data acquisition unit, edge computing unit, deformation analysis unit and dynamic execution unit, method includes data acquisition step, data fusion processing step, vector analysis step, deformation judgment step, vector control step and acceptance step.The application can be adjusted according to real-time diagnosis result aluminum plate deformation state, enhances the dynamic control ability to installation process, improves the consistency of installation flatness, to meet dynamic correction and installation precision demand, can be adjusted installation process in pertinence, and then improve the installation efficiency of aluminum plate.

Description

Technical Field

[0001] This invention relates to the field of industrial control technology, specifically to a real-time diagnostic system and vector control method for directional deformation of aluminum plates. Background Technology

[0002] Aluminum alloy sheets, due to their advantages such as light weight, high strength, corrosion resistance, and good processability, have become a typical structure widely used in modern building curtain walls and aerospace equipment manufacturing. During installation and service, aluminum sheet structures may experience directional micro-deformations due to various factors such as long-term alternating loads, foreign object impacts, uneven fastening forces, and changes in environmental temperature and humidity. If these micro-deformations are not detected and corrected in time, they will not only directly affect installation accuracy and appearance flatness, but also easily lead to structural stress concentration, accelerate fatigue damage, and affect the safety and service life of the structure. Therefore, during the installation of aluminum sheets, especially in high-standard application scenarios such as aerospace vehicle bulkheads and large building curtain walls, real-time monitoring of the shape changes of the sheet structure and effective responses are of great significance for improving product quality and reliability.

[0003] The existing standards for aluminum panel installation mainly include the following: the flatness error of the aluminum panel should not exceed 3mm, there should be no bending or deformation, the gap between the panels should be within 5mm, the gap size should be uniform, and the installation process should meet quality requirements. During installation and adjustment, it often relies on manual judgment and adjustment based on experience. For example, some curtain wall aluminum panel installation devices use mechanical structures such as adjusting balls and arc grooves to manually adjust the angle of the aluminum panels, lacking real-time perception and automatic feedback control of deformation. Furthermore, there are inconsistencies in the skill levels of technicians, resulting in uneven gaps, misaligned joints, and bent gaps even after installation and adjustment. In addition, the aluminum panels may have deformed during transportation, but this may not have been properly checked during installation and adjustment, leading to poor results and overall unevenness after adjustment.

[0004] In the field of aluminum panel installation and deformation monitoring, current quality inspection of aluminum panel installation for building curtain walls mainly relies on three types of technologies: The first type is contact measurement technology, such as Chinese patent with publication number CN110926434A, which uses a mechanical dial indicator or electronic micrometer for single-point measurement, requiring manual measurement operation point by point and panel by panel; the second type is optical non-contact measurement technology, such as Chinese patent with publication number CN112629498A, which uses a laser rangefinder to achieve non-contact measurement and obtain one-dimensional distance data; the third type is image processing technology, such as Chinese patent with publication number CN113763503A, which uses a feature point matching algorithm for surface detection based on computer vision, with high resolution and accuracy, requiring high-contrast marker points.

[0005] The existing technologies have common defects such as lack of directional detection, insufficient real-time performance, and lack of correction methods. They can only measure the magnitude of deformation, but it is difficult to identify the warping direction. The feedback adjustment cycle is long, which cannot meet the requirements of dynamic correction and installation accuracy. As a result, the installation process cannot be adjusted in a targeted manner, thereby reducing the installation efficiency of aluminum plates. Summary of the Invention

[0006] The purpose of this invention is to provide a real-time diagnostic system and vector control method for directional deformation of aluminum plates to solve the problems mentioned in the background art.

[0007] In a first aspect, the present invention provides a real-time diagnostic system for directional deformation of aluminum plates, which achieves the objective of the invention through the following technical solution:

[0008] A real-time diagnostic system for directional deformation of aluminum plate mounting includes a multi-source data acquisition unit, an edge computing unit, a deformation analysis unit, and a dynamic execution unit;

[0009] Multi-source data acquisition unit: used to simultaneously acquire reference point cloud data and attitude data of the aluminum plate after the initial installation and positioning of the aluminum plate, and establish a local coordinate system;

[0010] Edge computing unit: used to preprocess and spatiotemporally align reference point cloud data and attitude data, and output fused high-precision point cloud data and high-precision attitude data;

[0011] Deformation Analysis Unit: Based on high-precision point cloud data and attitude data, it quantifies deformation by calculating the change in the normal vector field of the aluminum plate surface, and predicts the deformation trend by combining a long short-term memory network model. Based on the deformation, it diagnoses the risk level and generates adjustment instructions.

[0012] Dynamic execution unit: Used to receive adjustment commands, dynamically adjust the tightening sequence and installation parameters of the aluminum plate bolts, and adjust the deformation state of the aluminum plate.

[0013] By adopting the above technical solution, a multi-source data acquisition unit is set up to achieve multi-dimensional perception of the aluminum plate's installation status, simultaneously acquiring reference point cloud data and attitude data. This enables the system to obtain the spatial geometric features and motion state information of the aluminum plate, providing a data foundation for subsequent deformation analysis. Compared to monitoring schemes using a single data source, this multi-source data acquisition method enhances the ability to characterize complex deformation patterns. By establishing a local coordinate system, a unified reference framework is provided for the fusion of data from different sensors, improving the consistency of measurement data. The edge computing unit preprocesses and spatiotemporally aligns the reference point cloud data and attitude data at the near end of the data acquisition process, reducing data transmission latency and improving the system's response speed. The fused high-precision point cloud data and high-precision attitude data provide more accurate and reliable input data for deformation analysis. The deformation analysis unit calculates the change in the normal vector field of the aluminum plate surface to... Quantifying deformation enhances the ability to identify subtle deformation features. Combined with a long short-term memory network model to predict deformation trends, the system can not only identify the current deformation state but also anticipate its future development, providing a window for proactive intervention. Based on the deformation, the system diagnoses the risk level and generates adjustment instructions, transforming complex deformation data into actionable decision-making information and improving system processing efficiency. The dynamic execution unit receives adjustment instructions and dynamically adjusts the tightening sequence of aluminum plate bolts and the technical characteristics of installation parameters. This allows the system to adaptively adjust based on real-time diagnostic results, adjusting the aluminum plate deformation state and enhancing the dynamic control capability of the installation process. The system can proactively correct deviations during installation, improving the consistency of installation flatness and meeting the requirements for dynamic correction and installation accuracy. It can also specifically adjust the installation process, thereby improving the installation efficiency of the aluminum plate.

[0014] Optionally, the multi-source data acquisition unit includes an environmental measurement module, and the deformation analysis unit specifically includes a deformation calculation module and a risk diagnosis module;

[0015] Environmental measurement module: used to synchronously collect temperature, humidity and vibration data of the aluminum plate;

[0016] Deformation calculation module: This module constructs the normal vector field of the aluminum plate surface based on high-precision point cloud data and attitude data, calculates the fractional gradient of the normal vector field, calculates the deformation direction angle based on the fractional gradient, and corrects the deformation direction angle using a dynamic compensation algorithm. The calculation model of the dynamic compensation algorithm is as follows: ,in, The corrected angular velocity, γ is the original angular velocity, β is the temperature drift coefficient, T is the temperature, γ is the vibration disturbance amplitude, and f is the vibration frequency.

[0017] Risk diagnosis module: It is used to predict the deformation trend based on the modified deformation direction angle combined with the long short-term memory network model, diagnose the risk level according to the deformation and its rate of change, and output the corresponding adjustment instructions.

[0018] By adopting the above technical solutions, in the actual working conditions of aluminum plate installation, temperature changes can cause sensor drift, and mechanical vibration can introduce measurement noise. The environmental measurement module synchronously collects temperature, humidity, and vibration data of the aluminum plate, enabling the system to sense these interfering factors and obtain key environmental parameters affecting measurement accuracy, thus creating conditions for improving the accuracy of measurement data. The deformation calculation module constructs a normal vector field on the aluminum plate surface and calculates the fractional gradient of the normal vector field. Compared with the traditional integer gradient operator method, it can better capture the local details and nonlinear characteristics of deformation, improving the sensitivity to subtle deformations. Calculating the deformation direction angle based on the fractional gradient can more accurately identify the main direction of deformation development, improving the accuracy of subsequent correction operations. The dynamic compensation algorithm compensates for the original angular velocity data in real time. The system effectively suppresses the interference of temperature drift and periodic vibration on the measurement data. The parameterized design of the temperature drift coefficient and vibration interference amplitude gives the compensation calculation clear physical meaning and adjustability, enhancing the adaptability of the algorithm in practical applications. The technical solution of the risk diagnosis module improves the system's forward-looking judgment ability by combining a long short-term memory network model to predict deformation trends. This allows the system to learn evolutionary patterns from historical deformation data, which has a positive effect on predicting the future development trend of deformation. The multi-indicator comprehensive judgment strategy based on deformation and its rate of change enhances the ability to distinguish different risk conditions, making the output adjustment instructions more targeted and reasonable. This meets the requirements of dynamic correction and installation accuracy, and allows for targeted adjustment of the installation process, thereby improving the installation efficiency of aluminum plates.

[0019] Optionally, in the deformation calculation module, the dynamic compensation algorithm is further modified, and the calculation model of the modified dynamic compensation algorithm is updated as follows:

[0020] Where β(T) is a temperature-dependent nonlinear temperature drift coefficient, and the calculation model of β(T) is expressed as: ,in To calibrate the reference value, The first calibration coefficient, δ is the second calibration coefficient, H is the ambient humidity, γ(T,H) is the vibration interference amplitude related to temperature and humidity, f(T) is the vibration frequency related to temperature, and φ is the phase compensation angle of the vibration interference.

[0021] By adopting the above technical solution, the modified dynamic compensation algorithm uses a temperature-related nonlinear temperature drift coefficient and performs nonlinear modeling in the form of a quadratic function, which better matches the actual physical characteristics of sensor temperature drift. This temperature-varying coefficient design can more accurately describe the complex behavior of temperature drift, thus maintaining good compensation effect over a wide temperature range. In the actual environment of aluminum plate installation, temperature and humidity changes often occur simultaneously and influence each other. Traditional single-factor compensation models are difficult to accurately describe this coupling effect. By introducing a humidity-temperature coupling coefficient, a deeper consideration of the coupling effect of multiple factors in the environment is reflected. By considering the modulation effect of humidity on temperature drift, the adaptability to complex environmental conditions is enhanced, further improving the accuracy of compensation. The vibration interference amplitude is a variable related to temperature and humidity. In actual working conditions, the vibration characteristics of the structure often... The vibration frequency changes with variations in ambient temperature and humidity. The compensation algorithm can adaptively adjust the suppression intensity of vibration interference, enhancing the tracking ability of time-varying vibration environments. Since the vibration frequency is a temperature-related parameter, this technical solution takes into account the physical laws governing material properties changing with temperature. As temperature changes, the stiffness characteristics of the structure change, thus affecting the vibration frequency. The algorithm can more accurately match actual vibration characteristics, improving the accuracy of vibration interference modeling. In actual vibration environments, interference signals not only have specific amplitudes and frequencies but also phase characteristics. The introduction of the phase compensation angle allows the algorithm to more accurately align the positive and negative peaks of the vibration signal, thereby achieving a more precise cancellation effect in the time domain, enhancing the suppression capability of vibration interference, and thus meeting the requirements of dynamic correction and installation accuracy. It can also specifically adjust the installation process, thereby improving the installation efficiency of aluminum plates.

[0022] Optionally, the dynamic execution unit specifically includes a control module, a bolt fastening tool, and a shape memory alloy actuator;

[0023] Control module: used to parse adjustment commands, generate specific control signals, and transmit the control signals to the bolt fastening tool and shape memory alloy actuator;

[0024] Bolt tightening tools: used to receive control signals and dynamically adjust the tightening sequence and torque of multiple bolts;

[0025] Shape memory alloy actuator: used to be attached to a specific position on an aluminum plate, activated after receiving the control signal, and applying a reverse force to the aluminum plate to perform deformation correction.

[0026] By adopting the above technical solution, the control module parses adjustment instructions and generates specific control signals, which are then transmitted to the bolt tightening tool and shape memory alloy actuator. This allows deformation diagnosis results to be translated into specific equipment operation commands, enhancing the collaborative efficiency between system units. The system can flexibly allocate correction tasks based on different deformation characteristics and severity, improving system resource utilization. The bolt tightening tool receives control signals and dynamically adjusts the tightening sequence and torque of multiple bolts, achieving precise control of the installation process. This changes the traditional fixed-sequence operation mode, allowing the system to determine the optimal tightening path based on real-time deformation status. This adaptive adjustment capability is crucial for controlling installation stress distribution and preventing stress concentration. Simultaneously, precise torque control allows for refined management of the tightening force at each tightening point, contributing to uniform stress distribution on the aluminum plate and reducing stress caused by uneven tightening force. The shape memory alloy actuator, attached to a specific location on the aluminum plate, allows for precise intervention in critical deformation areas. This positioning enhances the correction effect for localized deformation. The actuator can respond quickly when needed, achieving real-time compensation for deformation. By generating an active force opposite to the direction of aluminum plate deformation, it can effectively counteract deformation caused by installation stress or external loads. While bolt fastening tools primarily achieve macroscopic deformation control by adjusting the connection state, the shape memory alloy actuator achieves microscopic deformation correction by applying force to the plate itself. The synergistic effect of these two methods expands the range of handling deformations of different scales and types. This allows the system to select the most suitable correction method based on the actual characteristics of the deformation, or to adopt a combination scheme when necessary. This enhances the comprehensive handling capability for complex deformation problems, thereby meeting the requirements for dynamic correction and installation accuracy. It also allows for targeted adjustments to the installation process, thus improving the installation efficiency of the aluminum plate.

[0027] Optionally, the risk diagnosis module is further updated to: predict the deformation trend based on the modified deformation direction angle combined with a long short-term memory network model, diagnose the risk level according to the deformation amount and its rate of change, and if the magnitude of the deformation direction angle is θ > 5° or the rate of change of the deformation direction angle is... If 2° < θ < 5°, it is judged as a high-risk level; if θ ≤ 2°, it is judged as a medium-risk level; and if θ ≤ 2°, it is judged as a low-risk level. The corresponding adjustment instructions are output according to the different risk levels.

[0028] By adopting the above technical solution and using deformation direction angle and rate of change as indicators for risk assessment, the ability to evaluate deformation state is enhanced. The deformation direction angle reflects the current absolute degree of deformation, while the rate of change characterizes the speed of deformation development. The combination of the two allows the system to simultaneously focus on the static scale and dynamic trend of deformation, reducing the judgment bias that may have been caused by relying on only a single parameter in the past, and improving the comprehensiveness and accuracy of risk diagnosis. In particular, the introduction of the rate of change plays an important role in identifying early and rapidly developing deformation during the deformation development process, providing a clear basis for the system to identify the deformation state. The high-risk level threshold design takes into account the actual tolerance limit in aluminum plate installation projects. When the deformation exceeds this range, it may have a significant impact on structural safety or functionality. The setting of the medium-risk level... This allows the system to monitor minor, but not immediately dangerous, deformations, avoiding overreaction while ensuring proper management and monitoring, such as increasing monitoring frequency or taking preventative adjustments. The low-risk level is determined by considering the unavoidable minor deformations during aluminum plate installation, enabling the system to allocate resources efficiently and focus on more critical deformations, thus improving resource utilization. Corresponding adjustment commands are output based on different risk levels, allowing the system to take appropriate measures according to the severity of the deformation. This tiered response improves the system's ability to react quickly to severe deformations and reduces over-intervention in minor deformations, thereby meeting dynamic correction and installation accuracy requirements. It also allows for targeted adjustments to the installation process, ultimately improving the installation efficiency of the aluminum plates.

[0029] Optionally, the edge computing unit specifically includes a data preprocessing module and a spatiotemporal alignment module;

[0030] Data preprocessing module: used to denoise the collected point cloud data and compensate for temperature drift and vibration interference in the collected attitude data, and output fused high-precision point cloud data and compensated high-precision attitude data.

[0031] Spatiotemporal alignment module: The iterative nearest point algorithm is used to unify high-precision point cloud data and high-precision attitude data collected at different times into the same coordinate system.

[0032] By adopting the above technical solutions, the data preprocessing module performs noise reduction on the acquired point cloud data, effectively suppressing various random noises and outliers introduced during the measurement process, thus improving the signal-to-noise ratio and integrity of the point cloud data. It also compensates for temperature drift and vibration interference in the acquired attitude data, addressing the main error sources faced by the inertial measurement unit in practical applications, significantly improving the accuracy and reliability of the attitude data. The spatiotemporal alignment module uses an iterative nearest-point algorithm to unify data acquired at different times into the same coordinate system, achieving high-precision registration of data acquired at different times, improving the continuity and comparability of time-series data analysis. In practical application scenarios, since the measuring equipment may undergo slight displacement or attitude changes, unifying data from different times into the same coordinate system becomes a necessary condition for accurately analyzing the deformation process. This allows for the simultaneous inclusion of high-precision point cloud data and high-precision attitude data. Spatiotemporal alignment processing not only considers the spatial matching of 3D point clouds but also integrates attitude information provided by the inertial measurement unit, enhancing the ability to describe complex motion patterns. Especially in the case of large rigid body motion, the alignment method combined with attitude data can more accurately separate pure deformation components, improving the accuracy of deformation analysis. By completing computationally intensive data processing tasks at the near end of data acquisition, the transmission requirements to the cloud or central processing unit are reduced, and the overall system latency is reduced. For applications such as aluminum plate installation that require real-time feedback, the low-latency data processing capability enables the system to respond promptly to changes in deformation state, reducing the burden on network transmission and enhancing the system's adaptability in environments with poor network conditions. This meets the requirements for dynamic correction and installation accuracy, allows for targeted adjustments to the installation process, and ultimately improves the installation efficiency of aluminum plates.

[0033] Secondly, the present invention provides a vector control method for directional deformation of aluminum plate installation, which utilizes a real-time diagnostic system for directional deformation of aluminum plate installation as described in the first aspect, and achieves the objective of the invention through the following technical solution:

[0034] A method for controlling the directional deformation vector of an aluminum plate during installation includes the following steps:

[0035] Data acquisition steps: After the aluminum plate is initially installed and positioned, the reference point cloud data and attitude data of the aluminum plate are collected simultaneously to establish a local coordinate system;

[0036] Data fusion processing steps: preprocess and spatiotemporally align the collected reference point cloud data and attitude data, and output the fused high-precision point cloud data and high-precision attitude data;

[0037] Vector analysis steps: Based on high-precision point cloud data and high-precision attitude data, calculate the change of the normal vector field on the aluminum plate surface to quantify the deformation;

[0038] Deformation assessment steps: Based on the deformation variables and combined with a long short-term memory network model, the deformation trend is predicted, and adjustment instructions are generated based on the risk level diagnosed by the deformation variables.

[0039] Vector control steps: Dynamically adjust the tightening sequence and installation parameters of the aluminum plate bolts according to the adjustment instructions, and perform vector control and correction on the deformation state of the aluminum plate;

[0040] Acceptance steps: Generate a 3D deformation distribution map and output an inspection report with historical data comparison.

[0041] By adopting the above technical solution, and simultaneously acquiring reference point cloud data and attitude data to establish a local coordinate system, subsequent deformation monitoring has a reliable benchmark, improving the accuracy and comparability of deformation detection. This facilitates accurate correspondence between measurement data and the actual installation position of the aluminum plate, enhancing the spatial positioning accuracy of the measurement results. Preprocessing and spatiotemporal alignment of multi-source data improves data consistency and helps reduce the risk of misjudgment due to data quality issues. Deformation is quantified by calculating the change in the normal vector field of the aluminum plate surface. Changes in the normal vector field can sensitively reflect subtle changes in surface geometry, providing richer deformation information compared to traditional single-index measurement methods. This facilitates a comprehensive understanding of the deformation state of the aluminum plate and allows for deformation trend analysis using a long short-term memory network model. The system predicts the future development of deformation, providing a window of opportunity for preventative measures. It diagnoses risk levels based on deformation and generates adjustment instructions. By dynamically adjusting bolt tightening sequence and installation parameters, it changes the previous fixed-process operation mode, improving the intelligence level of the installation process. The system performs vector control and correction of aluminum plate deformation, enabling precise intervention in the direction and magnitude of deformation, enhancing its ability to correct complex deformation patterns. By generating a 3D deformation distribution map and a test report with historical data comparison, it facilitates engineers' rapid understanding of the overall deformation status, providing valuable insights for analyzing deformation patterns and optimizing installation processes. This meets the requirements for dynamic correction and installation accuracy, allowing for targeted adjustments to the installation process and ultimately improving the installation efficiency of aluminum plates.

[0042] Optionally, during the data acquisition step, temperature and humidity data and vibration data of the aluminum plate are also acquired simultaneously.

[0043] The vector analysis step is further updated as follows: A normal vector field for the aluminum plate surface is constructed based on high-precision point cloud data and attitude data; the fractional gradient of the normal vector field is calculated; the deformation direction angle is calculated based on the fractional gradient; and the deformation direction angle is corrected using a dynamic compensation algorithm. The calculation model of the dynamic compensation algorithm is as follows: ,in, The corrected angular velocity, γ is the original angular velocity, β is the temperature drift coefficient, T is the temperature, γ is the vibration disturbance amplitude, and f is the vibration frequency.

[0044] By adopting the above technical solution, in the actual environment of aluminum plate installation, temperature changes can cause thermal drift errors in measuring equipment, and mechanical vibration can introduce high-frequency noise. Simultaneous acquisition of temperature, humidity, and vibration data allows for the acquisition of key environmental parameters affecting measurement accuracy, improving the reliability of measurement data. A normal vector field is constructed and a fractional gradient is calculated. Based on the fractional gradient, the deformation direction angle is calculated, providing a means to quantitatively describe the deformation development direction. A dynamic compensation algorithm is introduced to correct the deformation direction angle, effectively suppressing the interference of environmental factors on the measurement data. This algorithm compensates for temperature drift and periodic vibration. The temperature drift coefficient is used to correct the sensor zero-point drift caused by temperature changes, and the vibration interference amplitude and frequency are used to describe and offset the periodic interference caused by mechanical vibration, making the correction process adjustable and enhancing the method's adaptability under different environmental conditions. By monitoring environmental parameters in real time and correcting deformation data accordingly, the method can maintain high measurement accuracy under changing working conditions, improving the reliability of deformation diagnosis results, thus meeting the requirements of dynamic correction and installation accuracy. It also allows for targeted adjustments to the installation process, thereby improving the installation efficiency of aluminum plates.

[0045] Optionally, the vector control step is further updated as follows: establish a digital twin model synchronized with the physical aluminum plate in a virtual space, execute a multi-objective optimization algorithm with the optimization objectives of minimizing deformation, balancing structural stress, and achieving high installation efficiency, simulate various tightening sequences and correction strategies in the digital space, and feed back the optimized optimal adjustment command, or a decision command set containing multiple optional schemes, to the management terminal. The management terminal further dynamically adjusts the tightening sequence and installation parameters of the aluminum plate bolts according to the optimal adjustment command or decision command set, and performs vector control and correction on the deformation state of the aluminum plate.

[0046] By adopting the above technical solution, a digital twin model synchronized with the physical aluminum plate in virtual space is established, providing a high-fidelity simulation environment for control decision-making. The digital twin model receives data from the physical system in real time, accurately reflecting the current state and evolution trend of the aluminum plate. This makes the simulation results in virtual space highly valuable for reference. Engineers can explore different control strategies and parameter settings without interfering with the physical entity, reducing the risk of interference to the actual installation process. By executing multi-objective optimization algorithms, the flatness and dimensional accuracy of the installed appearance are improved, which helps to improve the long-term service performance and fatigue life of the aluminum plate. The final control strategy can achieve a balance among multiple key performance indicators. By systematically exploring different correction schemes, the approximate optimal solution under given constraints can be identified, improving the comprehensiveness and scientific nature of control decision-making. The decision instruction set provides engineers with flexible choices and retains a certain space for manual intervention. The technical solution of dynamically adjusting the installation parameters based on the optimization results at the management end enhances the adaptability to complex engineering scenarios, making the control decision-making forward-looking and globally optimized, thereby meeting the requirements of dynamic correction and installation accuracy. It can also adjust the installation process in a targeted manner, thereby improving the installation efficiency of the aluminum plate.

[0047] Optionally, the deformation judgment step is further updated to: predicting the deformation trend based on the corrected deformation direction angle combined with a long short-term memory network model, diagnosing the risk level based on the deformation amount and its rate of change, and determining the risk level if the magnitude of the deformation direction angle is θ > 5° or the rate of change of the deformation direction angle is... If the risk level is high, an emergency correction and adjustment instruction is generated. If 2° < θ < 5°, the risk level is medium, and an instruction to adjust the installation sequence is generated. If θ ≤ 2°, the risk level is low.

[0048] By adopting the above technical solution, the risk level is diagnosed based on the deformation and its rate of change. The deformation reflects the current severity of the deformation, while the rate of change characterizes the speed of deformation development. The combined use of the two makes the risk assessment more comprehensive and reliable, enhances the early warning capability for sudden deformation risks, sets high-risk levels and generates emergency correction and adjustment instructions, providing clear handling specifications for severe deformation states. When the deformation exceeds these limits, it may have a significant impact on structural safety or functionality. Rapid response to high-risk deformation helps prevent further expansion and deterioration of the deformation. The setting of medium-risk levels provides appropriate handling measures for moderate deformation. For this type of deformation, process optimization rather than emergency correction is mainly adopted, making the deformation assessment process more standardized and regulated, reducing reliance on personal experience, improving the consistency and repeatability of the method, optimizing resource allocation, and enabling targeted adjustments to the installation process, thereby improving the installation efficiency of aluminum plates.

[0049] Compared with the prior art, the beneficial effects of the present invention are:

[0050] 1. By synchronously acquiring reference point cloud data and attitude data, the spatial geometric features and motion state information of the aluminum plate are obtained, enhancing the ability to characterize complex deformation patterns. A unified framework for fusing data from different sensors is provided by establishing a local coordinate system. Preprocessing and spatiotemporal alignment of reference point cloud data and attitude data at the near end of data acquisition reduces data transmission latency and improves system response speed. Deformation is quantified by calculating the change in the normal vector field of the aluminum plate surface, enhancing the ability to identify subtle deformation features. Combining a long short-term memory network model to predict deformation trends not only identifies the current deformation state but also anticipates the development trend, providing a time window for proactive intervention. Adjustment instructions are generated based on deformation, transforming complex deformation data into actionable decision information. Dynamically adjusting the tightening sequence and installation parameters of the aluminum plate bolts proactively corrects deviations during installation, improving the consistency of installation flatness and meeting the requirements for dynamic correction and installation accuracy. Targeted adjustments to the installation process further improve the installation efficiency of the aluminum plate.

[0051] 2. In the actual working conditions of aluminum plate installation, temperature changes can cause sensor drift, and mechanical vibration can introduce measurement noise. The environmental measurement module can obtain key environmental parameters affecting measurement accuracy by synchronously collecting temperature, humidity, and vibration data of the aluminum plate. By constructing the normal vector field of the aluminum plate surface and calculating the fractional gradient of the normal vector field, the deformation direction angle can be calculated based on the fractional gradient, which can better capture the local details and nonlinear characteristics of deformation. The dynamic compensation algorithm effectively suppresses the interference of temperature drift and periodic vibration on the measurement data by real-time compensation of the original angular velocity data. Among them, the parameterized design of the temperature drift coefficient and vibration interference amplitude enhances the adaptability of the algorithm in practical applications. By combining the long short-term memory network model to predict the deformation trend, the system can learn the evolution law from historical deformation data, enhance the ability to distinguish different risk conditions, and make the output adjustment instructions more targeted and reasonable, thereby meeting the requirements of dynamic correction and installation accuracy, and can adjust the installation process in a targeted manner, thereby improving the installation efficiency of aluminum plates.

[0052] 3. The revised dynamic compensation algorithm employs a temperature-dependent nonlinear temperature drift coefficient, modeled nonlinearly using a quadratic function. This better reflects the actual physical characteristics of sensor temperature drift. This temperature-dependent coefficient design more accurately describes the complex behavior of temperature drift, thus maintaining good compensation performance over a wide temperature range. Temperature and humidity changes often occur simultaneously and influence each other. By introducing a humidity-temperature coupling coefficient, the algorithm demonstrates a deeper consideration of the coupled effects of multiple factors in the environment. By considering the modulation effect of humidity on temperature drift, the algorithm enhances adaptability to complex environmental conditions and improves compensation accuracy. In practical applications… In this context, the vibration characteristics of a structure often change with variations in ambient temperature and humidity. The compensation algorithm can adaptively adjust the suppression intensity of vibration interference, enhancing its ability to track time-varying vibration environments. As temperature changes, the stiffness characteristics of the structure change, thus affecting the vibration frequency. The algorithm can more accurately match the actual vibration characteristics, improving the accuracy of vibration interference modeling. The introduction of the phase compensation angle enables the algorithm to more precisely align the positive and negative peaks of the vibration signal, enhancing its ability to suppress vibration interference. This meets the requirements for dynamic correction and installation accuracy, allowing for targeted adjustments to the installation process and thus improving the installation efficiency of aluminum plates.

[0053] 4. The bolt tightening tool receives control signals and dynamically adjusts the tightening sequence and torque of multiple bolts, achieving precise control over the installation process. This changes the traditional installation process's reliance on a fixed sequence, allowing the system to determine the optimal tightening path based on real-time deformation. This is crucial for controlling installation stress distribution and preventing stress concentration. Simultaneously, precise torque control enables refined management of the tightening force at each point, helping to achieve uniform stress distribution on the aluminum plate and reducing the risk of deformation due to uneven tightening force. Shape memory alloy actuators are attached to specific locations on the aluminum plate, allowing for precise intervention in key deformation areas, enhancing the correction effect of localized deformation and achieving real-time deformation compensation. The synergistic effect of these two methods expands the processing range for deformations of different scales and types, enabling the system to select the most suitable correction method based on the actual characteristics of the deformation, or, when necessary, employ a combination scheme. This enhances the comprehensive handling capability for complex deformation problems, meeting the requirements for dynamic correction and installation accuracy. It allows for targeted adjustments to the installation process, thereby improving the installation efficiency of the aluminum plate.

[0054] 5. Deformation direction angle and rate of change are used as indicators for risk assessment. The deformation direction angle reflects the current absolute degree of deformation, while the rate of change characterizes the speed of deformation development. The combination of the two allows the system to simultaneously monitor the static scale and dynamic trend of deformation, reducing the judgment bias that may have been caused by relying on a single parameter in the past, and improving the comprehensiveness and accuracy of risk diagnosis. In particular, the introduction of the rate of change plays an important role in identifying early and rapidly developing deformation during the deformation development process, providing a clear basis for the system to identify the deformation state. The threshold design of the risk level takes into account the actual tolerance limit in the aluminum plate installation project, enabling the system to rationally allocate processing resources and focus attention on the deformation states that require more attention, thereby improving the efficiency of system resource utilization. Corresponding adjustment instructions are output according to different risk levels, allowing the system to take appropriate measures according to the severity of deformation. This graded response method improves the system's ability to react quickly to severe deformation and reduces excessive intervention in minor deformation, thereby meeting the requirements of dynamic correction and installation accuracy, enabling targeted adjustments to the installation process, and thus improving the installation efficiency of aluminum plates. Attached Figure Description

[0055] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. Obviously, the drawings described below are merely some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without creative effort. In the drawings:

[0056] Figure 1 This is a block diagram of a real-time diagnostic system for directional deformation of an aluminum plate according to an embodiment of the present invention.

[0057] Figure 2 This is a flowchart illustrating a method for controlling the directional deformation vector of an aluminum plate according to an embodiment of the present invention. Detailed Implementation

[0058] The following will be based on embodiments of the present invention. Figure 1 and Figure 2 The technical solutions in the embodiments of the present invention are clearly and completely described herein. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0059] Example 1: This example discloses a real-time diagnostic system for directional deformation of aluminum plates during installation, referring to... Figure 1It includes a multi-source data acquisition unit, an edge computing unit, a deformation analysis unit, and a dynamic execution unit.

[0060] System deployment plan: including hardware installation and network architecture.

[0061] Hardware installation: The lidar is installed on the construction scaffold, 1.5-3m away from the aluminum plate. The IMU sensor is directly pasted on the surface of the aluminum plate, with one sensor per square meter. The edge computing box is fixed to the construction platform with IP67 protection and is designed to be waterproof and dustproof.

[0062] Network architecture: Sensor nodes are connected to the edge gateway via LoRaWAN with a latency of 50-100ms and a bandwidth of 50kbps. The edge gateway is connected to the cloud platform via 5G NR or 5G private network + TSN with a latency of 30ms and a bandwidth of 100Mbps. The cloud platform is connected to the supervision terminal and construction tablet via wireless networks.

[0063] Multi-source data acquisition unit: After the aluminum plate is initially installed and positioned, it is used to simultaneously acquire the reference point cloud data, attitude data, temperature and humidity data and vibration data of the aluminum plate, and to perform data verification, check the integrity of the point cloud, and establish a local coordinate system. It includes a laser scanning module, an inertial measurement module and an environmental measurement module.

[0064] Laser scanning module: Used to acquire reference point cloud data of aluminum plates using laser scanning.

[0065] The laser scanning module uses a 1550nm wavelength lidar with a ranging accuracy of ±0.1mm@10m, a field of view of 270°×90°, and a scanning frequency of 10Hz. Alternative option: RGB-D camera.

[0066] Inertial measurement module: Used to acquire attitude data of aluminum plate using inertial sensors.

[0067] The inertial measurement module uses a 6-axis MEMS sensor, model: BMI088, with an angular velocity range of ±2000dps, an acceleration range of ±16g, and a zero-bias stability of 0.5° / hr.

[0068] Environmental measurement module: Used to synchronously collect temperature, humidity and vibration data of aluminum plates using environmental sensors.

[0069] Edge computing unit: used to preprocess and spatiotemporally align reference point cloud data and attitude data, and output fused high-precision point cloud data and high-precision attitude data, including data preprocessing module and spatiotemporal alignment module.

[0070] The specific hardware implementation of the edge computing unit is an edge computing gateway. An alternative is a 4G module plus cloud computing.

[0071] Edge computing unit hardware configuration: main control chip: NVIDIA Jetson AGX Orin, computing power: 200TOPS (INT8), memory: 32GB LPDDR5.

[0072] Data preprocessing module: used to denoise the acquired point cloud data and compensate for temperature drift and vibration interference in the acquired attitude data, outputting fused high-precision point cloud data and compensated high-precision attitude data.

[0073] Spatiotemporal alignment module: The iterative nearest point algorithm is used to unify high-precision point cloud data and high-precision attitude data collected at different times into the same coordinate system.

[0074] Deformation Analysis Unit: Based on high-precision point cloud data and attitude data, it quantifies deformation by calculating the change in the normal vector field of the aluminum plate surface, and predicts the deformation trend by combining a long short-term memory network model. Based on the risk level of deformation, it generates adjustment instructions, including a deformation calculation module and a risk diagnosis module.

[0075] Construction of normal vector field: Local surface analysis based on PCA: ,in, For point The unit normal vector, representing the point The local surface direction, Let be the eigenvector corresponding to the smallest eigenvalue of the matrix, and k be the number of nearest neighbors, k=30. Let be the set of k-nearest neighbors of a point, and j be the number of points in the neighborhood. The mean center of the neighborhood points, Let j be the j-th point in the neighborhood, representing a three-dimensional coordinate vector.

[0076] Deformation calculation module: It is used to construct the normal vector field of the aluminum plate surface based on high-precision point cloud data and attitude data, calculate the fractional gradient of the normal vector field, calculate the deformation direction angle based on the fractional gradient, and correct the deformation direction angle using a dynamic compensation algorithm.

[0077] Fractional gradient calculation: using the Grünwald-Letnikov definition: Where α is the fractional order, α = 0.8, and h is the step size, representing the discretization parameter. denoted as the generalized binomial coefficient, f(x-mh) is the value of the function at the lag point, M is the cutoff order, and x is the sampling point.

[0078] The computational model of the dynamic compensation algorithm is as follows: ,in, This is the corrected angular velocity, which is used to calculate the quantized value of the deformation direction angle. Let T be the original angular velocity, β be the temperature, and β(T) be the temperature drift coefficient. The calculation model for β(T) is expressed as follows: ,in To calibrate the reference value, The first calibration coefficient, δ is the second calibration coefficient, H is the ambient humidity, γ is the vibration interference amplitude, γ(T,H) is the vibration interference amplitude related to temperature and humidity, f is the vibration frequency, f(T) is the vibration frequency related to temperature, and φ is the phase compensation angle of the vibration interference.

[0079] Risk diagnosis module: Used to predict deformation trends based on the modified deformation direction angle combined with a long short-term memory network model, and diagnose the risk level according to the deformation amount and its rate of change. If the magnitude of the deformation direction angle is θ > 5° or the rate of change of the deformation direction angle is... If the risk level is 2° / s, it is judged as high risk level; if 2° < θ < 5°, it is judged as medium risk level; if θ ≤ 2°, it is judged as low risk level, and corresponding adjustment instructions are output according to different risk levels.

[0080] Deformation trend prediction: Using an LSTM network structure, the prediction accuracy (test set) is: MAE: 0.12°, prediction step size: 15 minutes, and accuracy rate: 92.3%.

[0081] Dynamic execution unit: Used to receive adjustment instructions, dynamically adjust the tightening sequence and installation parameters of aluminum plate bolts, and adjust the deformation state of aluminum plate. It includes a control module, bolt tightening tools, and shape memory alloy actuators.

[0082] Control module: used to parse adjustment instructions, generate specific control signals, and transmit the control signals to bolt fastening tools and shape memory alloy actuators.

[0083] Bolt tightening tools: used to receive control signals and dynamically adjust the tightening sequence and torque of multiple bolts.

[0084] Shape memory alloy actuator: This actuator is attached to a specific location on an aluminum plate, activates upon receiving the control signal, and applies a reverse force to the aluminum plate for deformation correction. Alternative: Micro servo motor.

[0085] Workflow example: Taking the installation of aluminum panels for on-site signboards as an example, the first stage is the initial installation stage. After the workers initially position the aluminum panels, the system automatically scans and obtains the reference point cloud, establishes a local coordinate system, and the accuracy is ±0.3mm.

[0086] Phase 2: Monitoring the tightening process, detecting micro-deformation caused by bolt tightening in real time, alarming when >0.5°, and dynamically adjusting the tightening sequence;

[0087] Phase 3: Environmental compensation. When the temperature change is greater than 10℃, the material expansion coefficient is automatically corrected. When the wind speed is greater than 8m / s, the wind vibration resistance mode is activated.

[0088] Phase 4: Acceptance phase, generating a 3D deformation distribution map and outputting an inspection report with historical data comparison.

[0089] The implementation principle of the real-time diagnostic system for directional deformation of aluminum plate mounting in this embodiment is as follows:

[0090] By setting up multi-source data acquisition units, the system achieves multi-dimensional perception of the installation status of the aluminum plate, and simultaneously collects reference point cloud data and attitude data. This enables the system to acquire the spatial geometric features and motion state information of the aluminum plate, providing a data foundation for subsequent deformation analysis. Compared with monitoring schemes based on a single data source, this multi-source data acquisition method enhances the ability to characterize complex deformation patterns. By establishing a local coordinate system, it provides a unified reference framework for the fusion of data from different sensors, improving the consistency of measurement data.

[0091] In actual working conditions of aluminum plate installation, temperature changes can cause sensor drift, and mechanical vibration can introduce measurement noise. The environmental measurement module synchronously collects temperature, humidity and vibration data of the aluminum plate, enabling the system to sense these interference factors and obtain key environmental parameters that affect measurement accuracy, thereby creating conditions for improving the accuracy of measurement data.

[0092] The edge computing unit preprocesses and spatiotemporally aligns the reference point cloud data and attitude data near the data acquisition point, reducing data transmission latency and improving system response speed. The fused high-precision point cloud data and high-precision attitude data provide more accurate and reliable input data for deformation analysis. The deformation analysis unit quantifies deformation by calculating the change in the normal vector field of the aluminum plate surface. This method enhances the ability to identify subtle deformation features. Combined with a long short-term memory network model, the deformation trend is predicted, enabling the system not only to identify the current deformation state but also to predict the development trend of deformation, providing a time window for proactive intervention.

[0093] The data preprocessing module performs noise reduction on the acquired point cloud data, which can effectively suppress various random noises and outliers introduced during the measurement process, improve the signal-to-noise ratio and integrity of the point cloud data, compensate for temperature drift and vibration interference in the acquired attitude data, and process the main error sources faced by the inertial measurement unit in practical applications, which significantly improves the accuracy and reliability of attitude data.

[0094] The spatiotemporal alignment module uses an iterative nearest point algorithm to unify data collected at different times into the same coordinate system, achieving high-precision registration of data collected at different times and improving the continuity and comparability of time series data analysis. In practical application scenarios, since the measuring equipment may undergo slight displacement or attitude changes, unifying data at different times into the same coordinate system becomes a necessary condition for accurately analyzing the deformation process.

[0095] Incorporating both high-precision point cloud data and high-precision attitude data into spatiotemporal alignment processing not only considers the spatial matching of 3D point clouds but also integrates attitude information provided by the inertial measurement unit, enhancing the ability to describe complex motion patterns. Especially in the case of large rigid body motion, the alignment method combined with attitude data can more accurately separate pure deformation components, improving the accuracy of deformation analysis.

[0096] By completing computationally intensive data processing tasks near the data acquisition point, the transmission requirements to the cloud or central processing unit are reduced, thus reducing the overall system latency. For applications such as aluminum plate installation that require real-time feedback, the low-latency data processing capability enables the system to respond promptly to changes in deformation state, reducing the burden on network transmission and enhancing the system's adaptability in environments with poor network conditions. This meets the requirements for dynamic correction and installation accuracy, allows for targeted adjustments to the installation process, and ultimately improves the installation efficiency of aluminum plates.

[0097] The deformation calculation module constructs a normal vector field on the surface of the aluminum plate and calculates the fractional gradient of the normal vector field. Compared with the traditional integer gradient operator method, it can better capture the local details and nonlinear characteristics of deformation, improve the sensitivity to subtle deformation, and calculate the deformation direction angle based on the fractional gradient. This can more accurately identify the main direction of deformation development and improve the accuracy of subsequent correction operations.

[0098] The dynamic compensation algorithm effectively suppresses the interference of temperature drift and periodic vibration on the measurement data by real-time compensation of the original angular velocity data. The parameterized design of the temperature drift coefficient and vibration interference amplitude gives the compensation calculation clear physical meaning and adjustability, enhancing the adaptability of the algorithm in practical applications.

[0099] The revised dynamic compensation algorithm uses a temperature-dependent nonlinear temperature drift coefficient and performs nonlinear modeling in the form of a quadratic function, which better matches the actual physical characteristics of sensor temperature drift. This temperature-dependent coefficient design can more accurately describe the complex behavior of temperature drift, thus maintaining good compensation performance over a wide temperature range.

[0100] In the actual environment of aluminum plate installation, temperature and humidity changes often occur simultaneously and influence each other. Traditional single-factor compensation models are difficult to accurately describe this coupling effect. By introducing a humidity-temperature coupling coefficient, we reflect a deeper consideration of the coupling effects of multiple factors in the environment. By considering the modulation effect of humidity on temperature drift, we enhance the adaptability to complex environmental conditions and further improve the accuracy of compensation.

[0101] The amplitude of vibration disturbance is a variable related to temperature and humidity. In actual working conditions, the vibration characteristics of a structure often change with the changes in ambient temperature and humidity. The compensation algorithm can adaptively adjust the suppression intensity of vibration disturbance, thereby enhancing the ability to track time-varying vibration environments.

[0102] Vibration frequency is a temperature-related parameter. This technical solution takes into account the physical law of material properties changing with temperature. As temperature changes, the stiffness characteristics of the structure will change, which in turn affects the vibration frequency. The algorithm can more accurately match the actual vibration characteristics and improve the accuracy of vibration disturbance modeling.

[0103] In actual vibration environments, interference signals not only have specific amplitudes and frequencies, but also phase characteristics. The introduction of phase compensation angles enables the algorithm to more accurately align the positive and negative peaks of vibration signals, thereby achieving a more precise cancellation effect in the time domain, enhancing the ability to suppress vibration interference, thus meeting the requirements for dynamic correction and installation accuracy, enabling targeted adjustments to the installation process, and thus improving the installation efficiency of aluminum plates.

[0104] The risk diagnosis module enhances the system's forward-looking judgment ability by combining a long short-term memory network model to predict deformation trends. This allows the system to learn evolutionary patterns from historical deformation data, which has a positive effect on predicting the future development of deformation. The multi-indicator comprehensive judgment strategy based on deformation and its rate of change enhances the ability to distinguish different risk conditions, making the output adjustment instructions more targeted and reasonable. This meets the requirements of dynamic correction and installation accuracy, and allows for targeted adjustments to the installation process, thereby improving the installation efficiency of aluminum plates.

[0105] The system diagnoses risk levels based on deformation and generates adjustment instructions, thereby transforming complex deformation data into actionable decision-making information and improving system processing efficiency. The dynamic execution unit receives adjustment instructions and dynamically adjusts the tightening sequence of aluminum plate bolts and the technical characteristics of installation parameters. This allows the system to adaptively adjust based on real-time diagnostic results, adjusting the deformation state of the aluminum plate and enhancing the dynamic control capability of the installation process. The system can proactively correct deviations that occur during installation, improving the consistency of installation flatness, thus meeting the requirements for dynamic correction and installation accuracy. It can also make targeted adjustments to the installation process, thereby improving the installation efficiency of the aluminum plate.

[0106] The control module parses adjustment instructions and generates specific control signals, which are then transmitted to bolt fastening tools and shape memory alloy actuators. This allows deformation diagnosis results to be translated into specific equipment operation commands, enhancing the collaborative efficiency between system units. The system can flexibly allocate correction tasks according to the different characteristics and severity of deformation, thereby improving the utilization efficiency of system resources.

[0107] The bolt tightening tool receives control signals and dynamically adjusts the tightening sequence and torque of multiple bolts, achieving precise control over the installation process. This changes the traditional installation process that relies on a fixed sequence, allowing the system to determine the optimal tightening path based on real-time deformation. This adaptive adjustment capability is crucial for controlling installation stress distribution and preventing stress concentration. At the same time, precise torque control enables refined management of the tightening force at each fastening point, helping to achieve a uniform distribution of stress on the aluminum plate and reducing the risk of deformation caused by uneven tightening force.

[0108] Shape memory alloy actuators are attached to specific locations on the aluminum plate, enabling the actuators to precisely intervene in key deformation areas. This positioning arrangement enhances the correction effect of local deformation. The actuators can respond quickly when needed to achieve real-time compensation for deformation. By generating an active force opposite to the direction of deformation of the aluminum plate, it can effectively counteract deformation caused by installation stress or external load.

[0109] Bolt fastening tools primarily achieve macroscopic deformation control by adjusting the connection state, while shape memory alloy actuators achieve microscopic deformation correction by applying force to the plate itself. The synergistic effect of these two methods expands the range of processing for deformations of different scales and types, enabling the system to select the most suitable correction method based on the actual characteristics of the deformation, or to adopt a combination scheme when necessary. This enhances the comprehensive processing capability for complex deformation problems, thereby meeting the requirements for dynamic correction and installation accuracy, allowing for targeted adjustments to the installation process, and ultimately improving the installation efficiency of aluminum plates.

[0110] Using deformation direction angle and rate of change as indicators for risk assessment enhances the ability to evaluate deformation state. Deformation direction angle reflects the current absolute degree of deformation, while rate of change characterizes the speed of deformation development. The combination of the two allows the system to simultaneously focus on the static scale and dynamic trend of deformation, reducing the judgment bias that may have been caused by relying on a single parameter in the past, and improving the comprehensiveness and accuracy of risk diagnosis. In particular, the introduction of rate of change plays an important role in identifying early and rapidly developing deformation during the deformation development process, providing a clear basis for the system to identify deformation state.

[0111] The high-risk threshold design takes into account the actual tolerance limits in aluminum plate installation projects. When the deformation exceeds this range, it may have a significant impact on structural safety or functionality. The medium-risk level setting allows the system to monitor minor deformation states that are not immediately dangerous, avoiding overreaction while ensuring appropriate management and monitoring of these deformations, such as increasing monitoring frequency or taking preventive adjustment measures. The low-risk level is determined to take into account the unavoidable minor deformations during aluminum plate installation, enabling the system to allocate processing resources rationally and focus attention on deformation states that require more attention, thereby improving the efficiency of system resource utilization.

[0112] By outputting corresponding adjustment instructions based on different risk levels, the system can take appropriate measures according to the severity of deformation. This graded response method improves the system's ability to react quickly to severe deformation and reduces excessive intervention in minor deformation, thereby meeting the requirements for dynamic correction and installation accuracy. It can also adjust the installation process in a targeted manner, thereby improving the installation efficiency of aluminum plates.

[0113] Example 2: This example discloses a method for controlling the directional deformation vector of an aluminum plate during installation, referring to... Figure 2 It includes data acquisition steps, data fusion and processing steps, vector analysis steps, deformation judgment steps, vector control steps, and acceptance steps.

[0114] Data acquisition steps: After the aluminum plate is initially installed and positioned, the reference point cloud data, attitude data, temperature and humidity data and vibration data of the aluminum plate are collected simultaneously to establish a local coordinate system.

[0115] Data fusion processing steps: The collected point cloud data is denoised, and the collected attitude data is compensated for temperature drift and vibration interference. The iterative nearest point algorithm is used to unify the high-precision point cloud data and high-precision attitude data collected at different times into the same coordinate system, and the fused high-precision point cloud data and the compensated high-precision attitude data are output.

[0116] Vector analysis steps: Based on high-precision point cloud data and compensated high-precision attitude data, a normal vector field is constructed on the surface of the aluminum plate. The fractional gradient of the normal vector field is calculated. Based on the fractional gradient, the deformation direction angle is calculated. The deformation direction angle is corrected using a dynamic compensation algorithm. The calculation model of the dynamic compensation algorithm is as follows: ,in, This is the corrected angular velocity, which is used to calculate the quantized value of the deformation direction angle. γ is the original angular velocity, β is the temperature drift coefficient, T is the temperature, γ is the vibration disturbance amplitude, and f is the vibration frequency.

[0117] Deformation assessment steps: Based on the corrected deformation direction angle and combined with the long short-term memory network model, the deformation trend is predicted. The risk level is diagnosed according to the deformation amount and its rate of change. If the magnitude of the deformation direction angle is θ≤2°, it is judged as a low risk level.

[0118] If 2° < θ < 5°, it is judged as a medium risk level, and an instruction to adjust the installation sequence is generated.

[0119] If θ > 5° or the rate of change of the deformation direction angle is... If the risk level is determined to be high, an emergency correction and adjustment instruction will be generated.

[0120] Vector control steps: Establish a digital twin model synchronized with the physical aluminum plate in virtual space, execute a multi-objective optimization algorithm with the optimization objectives of minimizing deformation, balancing structural stress, and achieving high installation efficiency, simulate various tightening sequences and correction strategies in digital space, and feed back the optimized adjustment command, or a decision command set containing multiple optional schemes, to the management end. The management end further dynamically adjusts the tightening sequence and installation parameters of the aluminum plate bolts according to the optimal adjustment command or decision command set, and performs vector control and correction on the deformation state of the aluminum plate.

[0121] Acceptance steps: Generate a 3D deformation distribution map and output an inspection report with historical data comparison.

[0122] The implementation principle of the directional deformation vector control method for aluminum plate installation in this embodiment is as follows:

[0123] By synchronously collecting reference point cloud data and attitude data and establishing a local coordinate system, subsequent deformation monitoring has a reliable benchmark, which improves the accuracy and comparability of deformation detection. This facilitates the accurate correspondence between measurement data and the actual installation position of the aluminum plate, enhances the spatial positioning accuracy of the measurement results, and improves data consistency by preprocessing and spatiotemporal alignment of multi-source data, which helps reduce the risk of misjudgment due to data quality issues.

[0124] In the actual environment of aluminum plate installation, temperature changes can cause thermal drift errors in measuring equipment, and mechanical vibration can introduce high-frequency noise. By simultaneously collecting temperature, humidity and vibration data, key environmental parameters that affect measurement accuracy can be obtained, improving the reliability of measurement data. By constructing a normal vector field and calculating the fractional gradient, and calculating the deformation direction angle based on the fractional gradient, a means of quantitatively describing the deformation development direction can be provided.

[0125] A dynamic compensation algorithm is introduced to correct the deformation direction angle, effectively suppressing the interference of environmental factors on the measurement data. This algorithm compensates for temperature drift and periodic vibration. The temperature drift coefficient is used to correct the sensor zero-point drift caused by temperature changes, and the vibration interference amplitude and frequency are used to describe and cancel the periodic interference caused by mechanical vibration, making the correction process adjustable and enhancing the adaptability of the method under different environmental conditions. By monitoring environmental parameters in real time and correcting the deformation data accordingly, the method can maintain high measurement accuracy under changing working conditions, improve the reliability of deformation diagnosis results, and thus meet the requirements of dynamic correction and installation accuracy. It can also adjust the installation process in a targeted manner, thereby improving the installation efficiency of aluminum plates.

[0126] Deformation is quantified by calculating the change in the normal vector field of the aluminum plate surface. The change in the normal vector field can sensitively reflect the subtle changes in the surface geometry. Compared with the traditional single index measurement method, it provides richer deformation information, which is conducive to a comprehensive understanding of the deformation state of the aluminum plate. Combined with the long short-term memory network model for deformation trend prediction, the system can predict the future development trend of deformation and provide a time window for taking preventive measures.

[0127] Based on the risk level diagnosis of deformation, the system generates adjustment instructions and dynamically adjusts the bolt tightening sequence and installation parameters. This changes the previous operation mode that relied on fixed procedures, improves the intelligence level of the installation process, and performs vector control and correction of the deformation state of the aluminum plate. The system can accurately intervene in the direction and magnitude of deformation, and improves the ability to correct complex deformation patterns.

[0128] By generating a three-dimensional deformation distribution map and a test report with historical data comparison, engineers can quickly grasp the overall deformation status, which is of reference value for analyzing deformation patterns and optimizing installation processes. This meets the requirements for dynamic correction and installation accuracy, and allows for targeted adjustments to the installation process, thereby improving the installation efficiency of aluminum panels.

[0129] Establishing a digital twin model synchronized with the physical aluminum plate in a virtual space provides a high-fidelity simulation environment for control decisions. The digital twin model receives data from the physical system in real time and can accurately reflect the current state and evolution trend of the aluminum plate, making the simulation results in the virtual space highly valuable. Engineers can explore different control strategies and parameter settings without interfering with the physical entity, reducing the risk of interference to the actual installation process.

[0130] By executing a multi-objective optimization algorithm, the flatness and dimensional accuracy of the installed appearance are improved, which helps to improve the long-term service performance and fatigue life of the aluminum plate. This enables the final control strategy to achieve a balance among multiple key performance indicators. By systematically exploring different correction schemes, the approximate optimal solution under given constraints can be identified, thus improving the comprehensiveness and scientific nature of control decisions.

[0131] The decision instruction set provides engineers with flexible options and retains some room for manual intervention. The management end further dynamically adjusts the installation parameters based on the optimization results, which enhances the adaptability to complex engineering scenarios and makes the control decision have forward-looking and global optimization characteristics. This meets the requirements of dynamic correction and installation accuracy, and can make targeted adjustments to the installation process, thereby improving the installation efficiency of aluminum plates.

[0132] Risk levels are diagnosed based on deformation and its rate of change. Deformation reflects the current severity of deformation, while the rate of change characterizes the speed of deformation development. The combined use of both makes risk assessment more comprehensive and reliable, enhances the early warning capability for sudden deformation risks, sets high-risk levels and generates emergency correction and adjustment instructions, providing clear handling standards for severe deformation states. When deformation exceeds these limits, it may have a significant impact on structural safety or functionality. Rapid response to high-risk deformation helps prevent further expansion and deterioration of deformation. Setting medium-risk levels provides appropriate handling measures for moderate deformation. For this type of deformation, process optimization rather than emergency correction is mainly adopted, making the deformation assessment process more standardized and regulated, reducing reliance on personal experience, improving the consistency and repeatability of the method, optimizing resource allocation, and enabling targeted adjustments to the installation process, thereby improving the installation efficiency of aluminum panels.

[0133] The above are all preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Therefore, all equivalent changes made in accordance with the structure, shape and principle of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A real-time diagnostic system for directional deformation of an aluminum plate, characterized in that, It includes a multi-source data acquisition unit, an edge computing unit, a deformation analysis unit, and a dynamic execution unit; Multi-source data acquisition unit: used to simultaneously acquire reference point cloud data and attitude data of the aluminum plate after the initial installation and positioning of the aluminum plate, and to establish a local coordinate system, including an environmental measurement module; Environmental measurement module: used to synchronously collect temperature, humidity and vibration data of the aluminum plate; Edge computing unit: used to preprocess and spatiotemporally align reference point cloud data and attitude data, and output fused high-precision point cloud data and high-precision attitude data; Deformation Analysis Unit: Based on high-precision point cloud data and attitude data, it quantifies deformation by calculating the change of the normal vector field on the surface of the aluminum plate, and predicts the deformation trend by combining a long short-term memory network model. Based on the risk level of the deformation, it generates adjustment instructions, including a deformation calculation module and a risk diagnosis module. Deformation calculation module: This module constructs the normal vector field of the aluminum plate surface based on high-precision point cloud data and attitude data, calculates the fractional gradient of the normal vector field, calculates the deformation direction angle based on the fractional gradient, and corrects the deformation direction angle using a dynamic compensation algorithm. The calculation model of the dynamic compensation algorithm is as follows: ,in, The corrected angular velocity, Let γ be the original angular velocity, β be the temperature drift coefficient, T be the temperature, γ be the vibration disturbance amplitude, and f be the vibration frequency; β(T) is the temperature-dependent nonlinear temperature drift coefficient, and the calculation model of β(T) is expressed as: ,in To calibrate the reference value, The first calibration coefficient, δ is the second calibration coefficient, H is the ambient humidity, γ(T,H) is the vibration interference amplitude related to temperature and humidity, f(T) is the vibration frequency related to temperature, and φ is the phase compensation angle of the vibration interference. Risk diagnosis module: It is used to predict the deformation trend based on the modified deformation direction angle combined with the long short-term memory network model, diagnose the risk level according to the deformation and its rate of change, and output the corresponding adjustment instructions. Dynamic execution unit: Used to receive adjustment commands, dynamically adjust the tightening sequence and installation parameters of the aluminum plate bolts, and adjust the deformation state of the aluminum plate.

2. The real-time diagnostic system for directional deformation of aluminum plate mounting according to claim 1, characterized in that: The dynamic execution unit specifically includes a control module, a bolt fastening tool, and a shape memory alloy actuator; Control module: used to parse adjustment commands, generate specific control signals, and transmit the control signals to the bolt fastening tool and shape memory alloy actuator; Bolt tightening tools: used to receive control signals and dynamically adjust the tightening sequence and torque of multiple bolts; Shape memory alloy actuator: used to be attached to a designated position on an aluminum plate, activated after receiving the control signal, and applying a reverse force to the aluminum plate to perform deformation correction.

3. The real-time diagnostic system for directional deformation of aluminum plate mounting according to claim 1, characterized in that: The risk diagnosis module is further updated to: predict deformation trends based on the modified deformation direction angle combined with a long short-term memory network model; diagnose risk levels based on the deformation amount and its rate of change; and determine risk levels if the magnitude of the deformation direction angle is θ > 5° or the rate of change of the deformation direction angle is... If 2° < θ < 5°, it is judged as a high-risk level; if θ ≤ 2°, it is judged as a medium-risk level; and if θ ≤ 2°, it is judged as a low-risk level. The corresponding adjustment instructions are output according to the different risk levels.

4. The real-time diagnostic system for directional deformation of aluminum plate mounting according to claim 1, characterized in that: The edge computing unit specifically includes a data preprocessing module and a spatiotemporal alignment module; Data preprocessing module: used to denoise the collected point cloud data and compensate for temperature drift and vibration interference in the collected attitude data, and output fused high-precision point cloud data and compensated high-precision attitude data. Spatiotemporal alignment module: The iterative nearest point algorithm is used to unify high-precision point cloud data and high-precision attitude data collected at different times into the same coordinate system.

5. A method for vector control of directional deformation during aluminum plate mounting, employing a real-time diagnostic system for directional deformation during aluminum plate mounting as described in any one of claims 1-4, characterized in that, Includes the following steps: Data acquisition steps: After the aluminum plate is initially installed and positioned, the reference point cloud data and attitude data of the aluminum plate are collected simultaneously to establish a local coordinate system; Data fusion processing steps: preprocess and spatiotemporally align the collected reference point cloud data and attitude data, and output the fused high-precision point cloud data and high-precision attitude data; Vector analysis steps: Based on high-precision point cloud data and high-precision attitude data, calculate the change of the normal vector field on the aluminum plate surface to quantify the deformation; Deformation assessment steps: Based on the deformation variables and combined with a long short-term memory network model, the deformation trend is predicted, and adjustment instructions are generated based on the risk level diagnosed by the deformation variables. Vector control steps: Dynamically adjust the tightening sequence and installation parameters of the aluminum plate bolts according to the adjustment instructions, and perform vector control and correction on the deformation state of the aluminum plate; Acceptance steps: Generate a 3D deformation distribution map and output an inspection report with historical data comparison.

6. The method for controlling the directional deformation vector of an aluminum plate according to claim 5, characterized in that: During the data acquisition step, temperature, humidity, and vibration data of the aluminum plate are also collected simultaneously. The vector analysis step is further updated as follows: A normal vector field for the aluminum plate surface is constructed based on high-precision point cloud data and attitude data; the fractional gradient of the normal vector field is calculated; the deformation direction angle is calculated based on the fractional gradient; and the deformation direction angle is corrected using a dynamic compensation algorithm. The calculation model of the dynamic compensation algorithm is as follows: ,in, The corrected angular velocity, γ is the original angular velocity, β is the temperature drift coefficient, T is the temperature, γ is the vibration disturbance amplitude, and f is the vibration frequency.

7. The method for controlling the directional deformation vector of an aluminum plate according to claim 5, characterized in that: The vector control step is further updated as follows: a digital twin model synchronized with the physical aluminum plate in a virtual space is established, a multi-objective optimization algorithm is executed, with the optimization objectives of minimizing deformation, balancing structural stress, and high installation efficiency. Multiple fastening sequences and correction strategies are simulated in the digital space, and the optimized optimal adjustment command, or a decision command set containing multiple optional schemes, is fed back to the management terminal. The management terminal further dynamically adjusts the fastening sequence and installation parameters of the aluminum plate bolts according to the optimal adjustment command or decision command set, and performs vector control and correction on the deformation state of the aluminum plate.

8. The method for controlling the directional deformation vector of an aluminum plate according to claim 5, characterized in that: The deformation judgment step is further updated as follows: Based on the corrected deformation direction angle combined with a long short-term memory network model, the deformation trend is predicted; the risk level is diagnosed based on the deformation amount and its rate of change; if the magnitude of the deformation direction angle is θ > 5° or the rate of change of the deformation direction angle is... If the risk level is high, an emergency correction and adjustment instruction is generated. If 2° < θ < 5°, the risk level is medium, and an instruction to adjust the installation sequence is generated. If θ ≤ 2°, the risk level is low.

Images