An old oil painting strength evaluation and intelligent protection management system

Abstract

This invention relates to the field of cultural relic protection and restoration technology, specifically to a quantitative assessment and intelligent protection management system for the tensile stress of an old oil painting, comprising: a multi-source sensor data acquisition module, a tensile stress quantitative assessment module, an intelligent protection decision-making module, a digital management platform for the restoration process, and a tensile stress optimization control module. The multi-source sensor data acquisition module is used to collect multi-dimensional mechanical data of the oil painting in real time during the tensile stress process through a tensile sensor array, strain gauge network, and computer vision monitoring system. This invention, through multi-source sensor data fusion technology, achieves the synchronous acquisition of multi-dimensional mechanical data such as tensile force, strain, and morphology during the tensile stress process of the oil painting, providing a comprehensive data foundation for quantitative assessment. It establishes a coupled mechanical model of the canvas and paint layer, and calculates the stress distribution over the entire canvas area using the finite element method, breaking through the limitations of traditional methods that can only perform local qualitative assessments, and achieving accurate quantitative assessment of damage risk.

Description

Technical Field

[0001] This invention relates to the field of cultural relic protection and restoration technology, specifically to a quantitative assessment and intelligent protection management system for the tensile strength of old oil paintings. Background Technology

[0002] As a non-renewable cultural heritage, antique oil paintings are subject to degradation and loosening of canvas fibers due to factors such as temperature and humidity changes, light aging, and microbial erosion during long-term preservation. This leads to wrinkling, deformation, blistering of the paint layer, and even peeling, severely impacting the artistic value and lifespan of the painting. Canvas flattening is a core process in the restoration of antique oil paintings. By applying even tension to the canvas, it is restored to a flat state, laying the foundation for subsequent paint layer restoration and mounting. Currently, the flattening restoration of antique oil paintings mainly relies on the restorer's manual experience, which presents the following problems:

[0003] Restorers rely solely on feel and experience to adjust the tension, leading to significant variations in technique among restorers and an inability to achieve precise control. This can result in insufficient tension, causing unevenness, or excessive tension, resulting in secondary damage such as canvas tearing and paint peeling. Current restoration techniques use standardized tension safety thresholds, failing to adequately consider individual differences in the age, materials, aging, and size of oil paintings. For ancient, fragile oil paintings, improper threshold settings can easily cause irreversible damage. Therefore, we propose a quantitative assessment and intelligent protection management system for the tensile force required to flatten ancient oil paintings. Summary of the Invention

[0004] The purpose of this invention is to provide a quantitative assessment and intelligent protection management system for the tensile strength of antique oil paintings, so as to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A quantitative assessment and intelligent protection management system for the stress of damaging antique oil paintings includes:

[0007] The multi-source sensor data acquisition module is used to acquire multi-dimensional mechanical data in real time during the oil painting leveling process through a tensile sensor array, strain gauge network and computer vision monitoring system;

[0008] The tensile strength quantitative assessment module is used to calculate the stress distribution across the entire canvas based on the canvas-pigment layer coupled mechanical model, and to quantitatively assess the damage risk index under different tensile forces.

[0009] The intelligent protection decision module is used to dynamically adjust the tensile safety threshold according to the age, material, and condition of the oil painting, so as to realize real-time early warning and protection decision-making for abnormal tensile force.

[0010] The digital management platform for the repair process is used to record sensor data, operating parameters and evaluation results throughout the entire repair process, enabling traceability of the repair process and knowledge accumulation.

[0011] The leveling force optimization control module is used to achieve precise closed-loop control of the leveling process based on the damage risk assessment results and through an adaptive force adjustment algorithm.

[0012] Preferably, in the multi-source sensing data acquisition module, the tensile sensor array is arranged as follows:

[0013] Two to four high-precision S-shaped tension sensors are placed on each of the four sides of the inner frame of the oil painting;

[0014] The tension sensor is connected to the canvas skirt via a special clamp, and the connection point uses a reversible adhesive material to ensure that the canvas body is not damaged.

[0015] The strain gauge network uses resistance strain gauges, with 9 measurement points arranged in a 3×3 grid on the back of the canvas. Each measurement point simultaneously measures transverse and longitudinal strain, and the strain measurement resolution is better than 1 microstrain.

[0016] Preferably, in the multi-source sensor data acquisition module, the computer vision monitoring system includes:

[0017] High-resolution industrial camera, equipped with a ring-shaped shadowless light source;

[0018] The sub-pixel-level feature point tracking algorithm tracks 100-200 pre-marked feature points on the canvas surface in real time.

[0019] The 3D topography reconstruction unit reconstructs the 3D topography of the canvas surface using binocular stereo vision technology, and calculates the deformation and curvature changes of each region of the canvas.

[0020] Preferably, in the tensile strength quantitative evaluation module, the construction process of the canvas-pigment layer coupled mechanical model is as follows:

[0021] A two-dimensional elastic mechanical model of the canvas is established, and the formula for calculating the tension at the canvas nodes is as follows:

[0022] in, Let be the tension at the i-th node; The canvas elasticity coefficient; Let be the elongation of the i-th node; The damping coefficient;

[0023] Establish a stress transfer model for the pigment layer, taking into account the interfacial bonding strength between the canvas and the pigment layer;

[0024] By solving the coupled model using the finite element method, the two-dimensional plane stress tensor over the entire canvas is obtained:

[0025] in, , These are the normal stresses in the x and y directions, respectively. , This is the shear stress.

[0026] Preferably, in the tensile strength quantitative assessment module, the method for calculating the damage risk index is as follows:

[0027] Calculate von Mises equivalent stress:

[0028] Constructing a damage risk assessment model:

[0029] in, , , Let be the weighting coefficient, satisfying ; The maximum equivalent stress; The yield stress of the material; The rate of change of tensile force; The threshold for the rate of change of safe tensile force; For maximum strain; For the ultimate strain;

[0030] when The time is low risk. The risk level is currently medium. It is a high-risk situation.

[0031] Preferably, in the intelligent protection decision module, the method for dynamically adjusting the tensile safety threshold is as follows:

[0032] Establish oil painting attribute feature vector ;

[0033] Calculate attribute adjustment coefficients using machine learning models:

[0034] in, These are the weight coefficients obtained from model learning;

[0035] Calculate the adaptive safety threshold:

[0036] in, As a baseline safety threshold, It is a dynamic correction function based on real-time tension and risk index.

[0037] Preferably, in the intelligent protection decision-making module, the real-time early warning mechanism for anomalies includes:

[0038] Three-level early warning system:

[0039] Level 1 warning (advice) when A prompt message will be issued at the appropriate time;

[0040] Level 2 warning (warning) An audible and visual warning will be issued and the increase in tension will be suspended.

[0041] Level 3 Warning (Emergency) The tension retraction operation is executed automatically at that time;

[0042] The early warning information includes three dimensions: risk location, risk type, and recommended actions.

[0043] Historical early warning data is automatically stored in a case library for continuous optimization of early warning thresholds and decision-making models.

[0044] Preferably, the digital management platform for the repair process includes:

[0045] The oil painting basic information management unit stores metadata such as the painting's age, artist, materials, size, and historical restoration records.

[0046] The real-time data visualization unit displays the canvas stress distribution, tensile force change curve, and risk index trend in the form of a heat map.

[0047] The intelligent repair solution recommendation unit recommends the optimal repair parameters and operation procedures based on the historical case library;

[0048] The automatic repair report generation unit automatically generates a standardized repair report that includes all sensor data, evaluation results, and operation records.

[0049] The knowledge graph construction unit builds a knowledge graph linking oil painting, materials, defects, and restoration solutions, supporting intelligent retrieval and reasoning.

[0050] Preferably, in the tension optimization control module, the adaptive tension adjustment algorithm is as follows:

[0051] Establish the objective function for tension control:

[0052] in, , , These are weighting coefficients, which respectively control the tension tracking accuracy, damage risk, and tension change stability;

[0053] The optimal control sequence is solved using the Model Predictive Control (MPC) algorithm, with 10 steps in the prediction time domain and 3 steps in the control time domain.

[0054] The optimization solution is executed once per control cycle, with a control cycle of 100ms, to ensure real-time performance and stability.

[0055] A method for optimizing and controlling the tensile stress of an antique oil painting, comprising the following steps:

[0056] S1. Initial assessment of oil painting condition: The initial state data of the oil painting is collected through a multi-source sensor system to assess the elastic coefficient, yield strength and initial damage of the canvas.

[0057] S11: Collect the initial 3D shape of the canvas under no tension and establish a baseline shape model;

[0058] S12: Apply a small preload of 0.5-1N, measure the load-displacement curve of the canvas, and calculate the elastic coefficient of the canvas through linear regression;

[0059] S13: Gradually increase the tension until microcracks appear in the canvas, and measure the material's yield stress and ultimate strain;

[0060] S14: Detect initial cracks, spalling and other damage through high-resolution imaging, and establish a damage distribution map;

[0061] S2. Restoration plan parameter configuration: Set the baseline tension, target flatness, and maximum allowable risk index according to the properties of the oil painting;

[0062] S3. Real-time acquisition of multi-source sensor data: Simultaneously acquires multi-dimensional data from tension sensors, strain gauges, and vision systems;

[0063] S4. Tensile Strength Evaluation: Stress distribution and damage risk index are calculated based on a coupled mechanical model;

[0064] S41: Preprocess the sensor data, including noise reduction, filtering, and spatiotemporal registration;

[0065] S42: Solve the coupled mechanical model of canvas-pigment layer based on the finite element method to obtain the stress tensor of each node;

[0066] S43: Calculate equivalent stress and principal stress directions, and generate a stress distribution heatmap;

[0067] S44: Calculate the damage risk index of the entire canvas area based on the damage risk assessment model;

[0068] S45: Identify high-risk areas, calculate the area ratio of high-risk areas and the maximum risk value;

[0069] S5. Intelligent protection decision-making: Execute warning, pause or rollback operations according to the risk level;

[0070] S6. Adaptive tension optimization control: Calculates the optimal tension adjustment amount based on the model predictive control algorithm;

[0071] S7. Repair process data recording: Store all sensor data, evaluation results, and control commands into the database;

[0072] S8. Termination condition judgment: When the target flatness is reached and the risk index continues to be lower than the safety threshold, the leveling process ends.

[0073] Compared with the prior art, the beneficial effects of the present invention are:

[0074] This invention utilizes multi-source sensor data fusion technology to simultaneously collect multi-dimensional mechanical data such as tension, strain, and morphology during the leveling process of oil paintings, providing a comprehensive data foundation for quantitative assessment. It establishes a coupled mechanical model of the canvas and paint layer, calculating the stress distribution across the entire canvas using the finite element method, overcoming the limitations of traditional methods that can only perform local qualitative assessments, and achieving precise quantitative assessment of damage risk. It proposes a dynamic safety threshold adjustment mechanism and a three-level early warning system based on the properties of oil paintings, enabling personalized protection strategies to be formulated according to the specific conditions of each painting, effectively avoiding secondary damage caused by excessive tension. It constructs a complete digital management platform for the restoration process, realizing data recording, visualization, and knowledge accumulation throughout the entire restoration process, significantly improving the traceability and standardization of restoration work. It employs a model predictive control algorithm to achieve adaptive optimization control of the leveling tension, minimizing damage risk while ensuring restoration effectiveness, providing strong technical support for the scientific and intelligent restoration of antique oil paintings. Attached Figure Description

[0075] Figure 1 This is a schematic diagram of the modular architecture of the present invention;

[0076] Figure 2 This is a flowchart of the steps of the present invention. Detailed Implementation

[0077] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0078] Please see Figure 1 As shown, the present invention is a quantitative assessment and intelligent protection management system for the tensile stress of old oil paintings, including: a multi-source sensor data acquisition module, a quantitative assessment module for tensile stress, an intelligent protection decision module, a digital management platform for the restoration process, and a tensile stress optimization control module;

[0079] The multi-source sensor data acquisition module is used to acquire multi-dimensional mechanical data in real time during the leveling process of an oil painting through a tensile sensor array, strain gauge network, and computer vision monitoring system.

[0080] In the actual restoration process, the first step is to set up the sensing system. Three high-precision S-shaped tension sensors are placed on each of the four sides of the inner frame of the oil painting, for a total of 12 sensors, to ensure that subtle changes in tension can be captured. The tension sensors are connected to the skirt of the canvas through special aluminum alloy clamps. The connection points use reversible fish glue adhesive material, which can be easily removed with warm water after the restoration is completed, ensuring that the canvas itself is not damaged.

[0081] The strain gauge network uses resistance strain gauges, with nine measurement points arranged in a 3×3 uniform grid on the back of the canvas, covering the central and edge areas of the canvas. Two strain gauges, one horizontal and one vertical, are simultaneously attached to each measurement point to form a strain rose structure, which can simultaneously measure the strain in two orthogonal directions. The strain gauges are fixed with a low-modulus adhesive to avoid causing additional stiffness effects on the canvas.

[0082] The computer vision monitoring system uses industrial cameras to form a binocular stereo vision system, equipped with a ring-shaped shadowless LED light source to ensure uniform illumination on the canvas surface and avoid the influence of shadows on the measurement results. 150 circular feature points are pre-marked on the canvas surface. The feature points are marked with reversible white pigment and can be easily removed after repair. Real-time tracking of each feature point is achieved through a sub-pixel-level feature point tracking algorithm. Through binocular matching and 3D reconstruction algorithms, 3D topographic data of the canvas surface is generated in real time, and the deformation and curvature changes of each area are calculated.

[0083] All sensor data are synchronized in time via a synchronous acquisition card to ensure precise alignment of multi-source data on the time axis. The acquired raw data is first preprocessed, including digital filtering to remove noise, temperature compensation to eliminate temperature drift, and coordinate system unification to achieve spatial registration.

[0084] The tensile strength quantitative assessment module is used to calculate the stress distribution across the entire canvas based on the canvas-pigment layer coupled mechanical model, and to quantitatively assess the damage risk index under different tensile forces.

[0085] The process of constructing the canvas-pigment layer coupled mechanical model is as follows:

[0086] First, a two-dimensional elastic mechanical model of the canvas is established, and the canvas is discretized into... The formula for calculating the tensile force at each canvas node in a finite element mesh with 1 node is as follows:

[0087] in, For the first The tension at each node; The canvas elasticity coefficient; For the first The elongation of each node; is the damping coefficient.

[0088] Secondly, a stress transfer model for the paint layer is established, taking into account the interfacial adhesion strength between the canvas and the paint layer. When the interfacial shear stress exceeds the bond strength, the pigment layer will peel off. The pigment layer itself adopts an elastoplastic constitutive model, and when the stress exceeds the yield stress... Then it enters the plastic deformation stage.

[0089] By solving the coupled mechanical model using the finite element method, the two-dimensional plane stress tensor of each node in the global canvas is obtained:

[0090] in, , These are the normal stresses in the x and y directions, respectively. , Shear stress;

[0091] von Mises equivalent stress is calculated based on the stress tensor and used as a comprehensive indicator to measure stress level.

[0092] Construct a multi-factor damage risk assessment model that comprehensively considers three factors: stress level, rate of tensile change, and strain magnitude.

[0093] in, , , Let be the weighting coefficient, satisfying ; The maximum equivalent stress; The yield stress of the material; The rate of change of tensile force; The threshold for the rate of change of safe tensile force; For maximum strain; This is the ultimate strain.

[0094] According to the risk index The values ​​are divided into three risk levels:

[0095] when The time is low risk. The risk level is currently medium. This is a high-risk situation;

[0096] This is a low-risk area; repair operations can proceed normally. This is a medium-risk area, requiring close monitoring and an appropriate reduction in the rate of increase in tension. This is a high-risk area, and protective measures must be taken immediately.

[0097] The intelligent protection decision module is used to dynamically adjust the tensile safety threshold based on the age, material, and condition of the oil painting, so as to realize real-time early warning and protection decision-making for abnormal tensile force.

[0098] First, establish the feature vector of oil painting attributes. ;

[0099] Each attribute is normalized: Year Based on linear normalization from 0 to 500 years, the material The canvas is rated on a scale of 1-10 based on its fiber type and degree of aging. Based on the initial damage severity score (1-10 points), size Normalized to 0-2 square meters, thickness Normalized to 0-2mm.

[0100] Calculate attribute adjustment coefficients using a random forest machine learning model:

[0101] in, These are the weight coefficients obtained from model learning;

[0102] The model was trained using over 1000 historical restoration cases, with typical weight coefficient values ​​of [value missing]. , , It is evident that the older and larger the oil painting, the smaller the safety threshold adjustment coefficient, and the lower the maximum allowable tensile force.

[0103] Calculate the adaptive safety threshold:

[0104] Among them, the benchmark security threshold For a standard-sized oil painting, set to 50N, dynamic correction function. The system adjusts online based on real-time tension and risk index, automatically lowering the safety threshold when the risk index increases.

[0105] Establish a three-tiered early warning and response mechanism: Level 1 early warning... When this occurs, a yellow warning message will be displayed on the operating interface, suggesting a reduction in the rate of increase in tension; a level two warning will be issued when... When the risk level drops, an audible and visual warning will be issued and the increase in tension will be automatically halted, waiting for the risk index to decrease; a level three warning will be issued when... When the risk index drops below 0.5, the system automatically performs a tension reduction operation, decreasing the tension at a rate of 0.2 N / s until the risk index drops to below 0.5. The entire early warning response time is controlled within 500 ms, ensuring timely and effective protection.

[0106] The digital management platform for the repair process is used to record sensor data, operating parameters, and evaluation results throughout the entire repair process, enabling traceability and knowledge accumulation of the repair process.

[0107] The digital management platform for the restoration process includes a basic information management unit for oil paintings, a real-time data visualization unit, an intelligent recommendation unit for restoration plans, an automatic generation unit for restoration reports, and a knowledge graph construction unit.

[0108] The oil painting basic information management unit establishes a digital archive for each oil painting, including metadata such as dating, artist information, material analysis, size measurement, and historical restoration records, and supports quick retrieval by scanning QR codes.

[0109] The real-time data visualization unit uses WebGL technology to render the canvas stress distribution heat map in real time, supports clicking and querying any area, and displays key indicators such as the tensile change curves of each sensor and the risk index trend chart.

[0110] The intelligent restoration plan recommendation unit is based on case reasoning (CBR) technology. It retrieves the five cases most similar to the current oil painting from the historical case library, extracts their restoration parameters and operation procedures, and combines them with the current status assessment results to generate personalized restoration plan suggestions, including recommended tensile loading curves, dwell time at key nodes, risk control strategies, etc.

[0111] The automatic restoration report generation unit automatically integrates all data from the restoration process, including basic information, sensor data curves, stress distribution cloud maps, risk assessment results, and operation logs, according to the standard format of cultural relic restoration archives, and generates a standardized PDF restoration report that supports printing and electronic archiving.

[0112] The knowledge graph construction unit uses the Neo4j graph database to build a knowledge graph linking oil painting, materials, defects, and restoration solutions. It supports intelligent retrieval using natural language, and the system can automatically infer and return relevant restoration cases and technical parameters.

[0113] The leveling force optimization control module is used to achieve precise closed-loop control of the leveling process based on the damage risk assessment results and through an adaptive force adjustment algorithm.

[0114] Establish a multi-objective optimization objective function for tension control:

[0115] in, , , These are weighting coefficients, which respectively control the tension tracking accuracy, damage risk, and tension change stability.

[0116] Model predictive control (MPC) algorithm is used to solve for the optimal control sequence. The prediction time domain is set to 10 steps, and the control time domain is set to 3 steps. The optimization is performed once every 100ms in each control cycle. By using rolling optimization, the optimal tension adjustment is recalculated based on the latest state at each step to achieve real-time compensation for disturbances.

[0117] like Figure 2 As shown, a method for optimizing and controlling the tensile stress of an antique oil painting includes the following steps:

[0118] S1. Initial assessment of oil painting condition: The initial state data of the oil painting is collected through a multi-source sensor system to assess the elastic coefficient, yield strength and initial damage of the canvas.

[0119] S11: Collect the initial 3D shape of the canvas under no tension and establish a baseline shape model;

[0120] S12: Apply a small preload (0.5-1N), measure the load-displacement curve of the canvas, and calculate the elastic coefficient through linear regression. ;

[0121] S13: Gradually increase the tensile force until microcracks appear, and determine the material's yield stress. and ultimate strain ;

[0122] S14: Detect initial cracks, spalling and other damage through high-resolution imaging, and establish a damage distribution map.

[0123] S2. Restoration plan parameter configuration: Based on the age, material, size and other attributes of the oil painting, set the benchmark tensile force range to 20-60N, the target flatness error to be less than 0.1mm, and the maximum allowable risk index to be 0.6.

[0124] S3. Real-time acquisition of multi-source sensor data: Simultaneously acquires 12-channel tension sensor data, 18-channel strain signal data, and binocular vision image data, and performs time synchronization and preprocessing.

[0125] S4. Stress-force quantitative assessment: Solve the coupled mechanical model to obtain the stress distribution, calculate the global risk index, and identify high-risk areas;

[0126] S41: Preprocess the sensor data, including noise reduction, filtering, and spatiotemporal registration;

[0127] S42: Solve the coupled mechanical model of canvas-pigment layer based on the finite element method to obtain the stress tensor of each node;

[0128] S43: Calculate the von Mises equivalent stress and principal stress directions, and generate a stress distribution heatmap;

[0129] S44: Calculate the global risk index based on the damage risk assessment model. ;

[0130] S45: Identify high-risk areas, calculate the area ratio of high-risk areas and the maximum risk value.

[0131] S5. Intelligent protection decision-making: Execute corresponding early warning and protection operations based on the risk level.

[0132] S6. Adaptive tension optimization control: The MPC algorithm calculates the optimal tension adjustment amount and outputs control commands to the electric actuator.

[0133] S7. Repair process data recording: All data is written to the database in real time, supporting breakpoint resume and abnormal recovery.

[0134] S8. Termination condition judgment: When the canvas flatness reaches the target value and the risk index is below 0.3 for 30 consecutive seconds, the flattening process will automatically end.

[0135] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A quantitative assessment and intelligent protection management system for the tensile strength of antique oil paintings, characterized in that, include: The multi-source sensor data acquisition module is used to acquire multi-dimensional mechanical data in real time during the oil painting leveling process through a tensile sensor array, strain gauge network and computer vision monitoring system; The tensile strength quantitative assessment module is used to calculate the stress distribution across the entire canvas based on the canvas-pigment layer coupled mechanical model, and to quantitatively assess the damage risk index under different tensile forces. The intelligent protection decision module is used to dynamically adjust the tensile safety threshold according to the age, material, and condition of the oil painting, so as to realize real-time early warning and protection decision-making for abnormal tensile force. The digital management platform for the repair process is used to record sensor data, operating parameters and evaluation results throughout the entire repair process, enabling traceability of the repair process and knowledge accumulation. The leveling force optimization control module is used to achieve precise closed-loop control of the leveling process based on the damage risk assessment results and through an adaptive force adjustment algorithm.

2. The methodological assessment and intelligent protection management system for the tensile strength of antique oil paintings according to claim 1, characterized in that, In the multi-source sensing data acquisition module, the tensile sensor array is arranged as follows: Two to four high-precision S-shaped tension sensors are placed on each of the four sides of the inner frame of the oil painting; The tension sensor is connected to the canvas skirt via a special clamp, and the connection point uses a reversible adhesive material to ensure that the canvas body is not damaged. The strain gauge network uses resistance strain gauges, with 9 measurement points arranged in a 3×3 grid on the back of the canvas. Each measurement point simultaneously measures the transverse and longitudinal strain.

3. The methodological assessment and intelligent protection management system for the tensile strength of antique oil paintings according to claim 1, characterized in that, The computer vision monitoring system in the multi-source sensor data acquisition module includes: High-resolution industrial camera, equipped with a ring-shaped shadowless light source; The sub-pixel-level feature point tracking algorithm tracks 100-200 pre-marked feature points on the canvas surface in real time. The 3D topography reconstruction unit reconstructs the 3D topography of the canvas surface using binocular stereo vision technology, and calculates the deformation and curvature changes of each region of the canvas.

4. The methodological assessment and intelligent protection management system for the tensile strength of antique oil paintings according to claim 1, characterized in that, In the tensile strength quantitative evaluation module, the construction process of the canvas-pigment layer coupled mechanical model is as follows: A two-dimensional elastic mechanical model of the canvas is established, and the formula for calculating the tension at the canvas nodes is as follows: in, For the first The tension at each node; The canvas elasticity coefficient; For the first The elongation of each node; The damping coefficient; Establish a stress transfer model for the pigment layer, taking into account the interfacial bonding strength between the canvas and the pigment layer; By solving the coupled model using the finite element method, the two-dimensional plane stress tensor over the entire canvas is obtained: in, , These are the normal stresses in the x and y directions, respectively. , This is the shear stress.

5. The methodological assessment and intelligent protection management system for the tensile strength of antique oil paintings according to claim 4, characterized in that, In the tensile strength quantitative assessment module, the method for calculating the damage risk index is as follows: Calculate von Mises equivalent stress: Constructing a damage risk assessment model: in, , , For the weighting coefficients, satisfying ; The maximum equivalent stress; The yield stress of the material; The rate of change of tensile force; The threshold for the rate of change of safe tensile force; For maximum strain; For the ultimate strain; when The time is low risk. The risk level is currently medium. It is a high-risk situation.

6. The methodological assessment and intelligent protection management system for the tensile strength of antique oil paintings according to claim 1, characterized in that, In the intelligent protection decision module, the method for dynamically adjusting the tensile safety threshold is as follows: Establish oil painting attribute feature vector ; Calculate attribute adjustment coefficients using machine learning models: in, These are the weight coefficients obtained from model learning; Calculate the adaptive safety threshold: in, As a baseline safety threshold, It is a dynamic correction function based on real-time tension and risk index.

7. The methodological assessment and intelligent protection management system for the tensile strength of antique oil paintings according to claim 1, characterized in that, In the intelligent protection decision-making module, the real-time early warning mechanism for anomalies includes: Three-level early warning system: Level 1 warning: When A prompt message will be issued at the appropriate time; Level 2 warning: When An audible and visual warning will be issued and the increase in tension will be suspended. Level 3 Warning: When The tension retraction operation is executed automatically at that time; The early warning information includes three dimensions: risk location, risk type, and recommended actions. Historical early warning data is automatically stored in a case library for continuous optimization of early warning thresholds and decision-making models.

8. The methodological assessment and intelligent protection management system for the tensile strength of antique oil paintings according to claim 1, characterized in that, The digital management platform for the repair process includes: The oil painting basic information management unit stores metadata such as the painting's age, artist, materials, size, and historical restoration records. The real-time data visualization unit displays the canvas stress distribution, tensile force change curve, and risk index trend in the form of a heat map. The intelligent repair solution recommendation unit recommends the optimal repair parameters and operation procedures based on the historical case library; The automatic repair report generation unit automatically generates a standardized repair report that includes all sensor data, evaluation results, and operation records. The knowledge graph construction unit builds a knowledge graph linking oil painting, materials, defects, and restoration solutions, supporting intelligent retrieval and reasoning.

9. The methodological assessment and intelligent protection management system for the tensile strength of antique oil paintings according to claim 1, characterized in that, In the aforementioned tension optimization control module, the adaptive tension adjustment algorithm is as follows: Establish the objective function for tension control: in, , , These are weighting coefficients, which respectively control the tension tracking accuracy, damage risk, and tension change stability; The optimal control sequence is solved using a model predictive control algorithm, with several steps in the prediction time domain and several steps in the control time domain. The optimization solution is executed once per control cycle, with a control cycle of 100ms, to ensure real-time performance and stability.

10. A method for optimizing and controlling the tensile stress of an antique oil painting, characterized in that, Includes the following steps: S1. Initial assessment of oil painting condition: The initial state data of the oil painting is collected through a multi-source sensor system to assess the elastic coefficient, yield strength and initial damage of the canvas. S2. Restoration plan parameter configuration: Set the baseline tension, target flatness, and maximum allowable risk index according to the properties of the oil painting; S3. Real-time acquisition of multi-source sensor data: Simultaneously acquires multi-dimensional data from tension sensors, strain gauges, and vision systems; S4. Tensile Strength Evaluation: Calculation of stress distribution and damage risk index based on coupled mechanical model; S5. Intelligent protection decision-making: Execute warning, pause or rollback operations according to the risk level; S6. Adaptive tension optimization control: Calculates the optimal tension adjustment amount based on the model predictive control algorithm; S7. Repair process data recording: Store all sensor data, evaluation results, and control commands into the database; S8. Termination condition judgment: When the target flatness is reached and the risk index continues to be lower than the safety threshold, the leveling process ends.

Images