A method for identifying damage and predicting life of a harvesting machine chassis for remanufacturing

Abstract

This invention discloses a method for damage identification and life prediction of harvesting machinery chassis for remanufacturing, belonging to the field of agricultural machinery remanufacturing technology. It includes: establishing a finite element model of the chassis frame; identifying weak points through static and modal analysis; performing numerical simulation of damage based on strain modal theory, extracting the strain modal difference as a damage identification index, and verifying it through simply supported beam tests; optimizing the sensor measurement point layout using the effective independent-drive point residual method, and evaluating the scheme using the weighted modal confidence criterion; conducting field tests to collect strain data based on the optimized scheme, and compiling a load spectrum; and predicting the remaining fatigue life by combining the material S-N curve and Miner's cumulative damage theory. This invention can accurately identify the location, extent, and type of damage to the chassis frame, optimize the sensor layout scheme, and scientifically assess the remaining fatigue life, providing an objective basis for judging whether a used chassis frame has remanufacturing and repair value, thereby reducing agricultural production costs.

Description

Technical Field

[0003] This invention relates to the field of agricultural machinery remanufacturing technology, specifically to a method for identifying damage to the chassis frame of harvesting machinery and a method for predicting fatigue life for remanufacturing purposes. This method is applicable to assessing the damage status and predicting the remaining fatigue life of used harvesting machinery chassis frames to determine whether they have remanufacturing and repair value. Background Technology

[0005] Harvesting machinery is crucial agricultural equipment for ensuring the safe production of major grain crops such as rice and wheat. With the continuous improvement of agricultural mechanization in my country, the market stock of harvesting machinery has been steadily increasing, reaching approximately 1.8 million units by 2023. At the same time, a large number of harvesting machines are gradually entering the retirement and scrapping stage, and the number of obsolete harvesting machines is rising year by year. How to scientifically dispose of obsolete harvesting machinery to avoid resource waste and environmental pollution has become an urgent problem for the industry.

[0006] Against the backdrop of global advocacy for sustainable development and a circular economy, remanufacturing engineering, as a green and low-carbon method of waste resource utilization, has received widespread attention. For used harvesting machinery, its core load-bearing component, the chassis frame, although subjected to alternating loads and vibrations in complex field operating environments for extended periods, often does not suffer fatal damage, possessing the potential to be repaired and reused through remanufacturing. However, to achieve the scientific remanufacturing of used chassis frames, two key issues must first be addressed: first, how to accurately identify the existing damage state of the chassis frame (including the location, extent, and type of damage); and second, how to assess its remaining fatigue life to determine whether it has remanufacturing repair value.

[0007] Currently, existing technologies for damage identification and fatigue life prediction of harvesting machinery chassis frames have the following main shortcomings:

[0008] 1. Existing fatigue life prediction methods cannot be directly applied to damage identification in remanufacturing scenarios.

[0009] Existing research indicates that fatigue life prediction technology has achieved relatively mature applications in the automotive and aerospace industries. In recent years, researchers have also attempted to apply finite element analysis and load spectrum compilation methods to the fatigue life analysis of harvesting machinery chassis frames. For example, published patent CN119720602B proposes a fatigue life analysis method for combine harvester chassis frames. This method predicts the fatigue life of the chassis frame under different operating conditions by constructing a finite element model, compiling a load spectrum, and calculating fatigue assessment coefficients. However, the core assumption of this type of method is that the chassis frame is in good or near-good condition, and the fatigue life prediction results are mainly used to guide the design optimization of new products or the maintenance of existing products.

[0010] The situation is entirely different for used harvesting machinery chassis frames. Over long periods of service, these frames have developed various types of damage, including fatigue cracks, plastic deformation, and weld cracking. The combination of these damages in terms of location, extent, and type is extremely complex. Existing fatigue life prediction methods cannot identify the existing damage state of the chassis frame, nor can they determine which damages can be repaired technically or which have rendered the structure unusable for remanufacturing. In other words, current technology addresses the question of how long a new product can be used, rather than whether the used parts are worth repairing.

[0011] 2. Lack of damage identification methods applicable to the chassis and frame of harvesting machinery.

[0012] Structural damage identification technology is a crucial prerequisite for determining whether discarded components have remanufacturing value. Currently, damage identification methods are mainly based on vibration testing and modal analysis, judging structural damage by extracting changes in parameters such as natural frequencies, displacement mode shapes, and strain modes. However, existing research mostly focuses on large-scale civil engineering structures such as bridges and buildings, or high-value components such as aero-engine blades and wind turbine blades, with very few reports on structural damage identification for agricultural machinery, especially the chassis and frames of harvesting machinery.

[0013] Harvesting machinery chassis frames are characterized by complex structures, numerous welded joints, and variable operating loads, making it difficult to directly apply existing damage identification methods. Firstly, existing methods often require a large number of sensors to obtain sufficient modal information, but the limited space and harsh working environment of the chassis frame make dense sensor deployment impractical. Secondly, existing methods lack sufficient sensitivity to damage, making it difficult to effectively extract sensitive features closely related to the location and severity of damage from complex modal parameters. Therefore, there is currently a lack of an effective method suitable for used harvesting machinery chassis frames that can achieve accurate damage identification under limited measurement points.

[0014] 3. The sensor optimization placement method was not specifically designed for the damage identification target.

[0015] In structural damage identification, the arrangement of sensor measurement points directly determines the comprehensiveness of modal parameter acquisition and the accuracy of damage identification. Existing technologies mainly employ sensor optimization methods such as the effective independence method, the improved effective independence method, and intelligent optimization algorithms. The research objectives of these methods are primarily to obtain the most modal information with the fewest sensors or to ensure the spatial independence of modal vectors.

[0016] However, the aforementioned methods primarily consider the information content and independence of modal parameters during design, without prioritizing the sensitivity of damage identification as a core optimization objective. In other words, while the sensor arrangement optimized using existing methods can reflect the overall vibration characteristics of the structure well, it may not be able to most sensitively capture local modal abrupt changes at the damage location. For damage identification of scrap chassis frames, it is precisely necessary for sensors to accurately locate the local anomalies caused by the damage. Therefore, existing sensor optimization methods cannot directly meet the specific requirements of damage identification for harvesting machinery chassis frames.

[0017] 4. Lack of a complete technical solution from damage identification to remanufacturing decision-making.

[0018] In summary, existing technologies exhibit significant gaps in the remanufacturing assessment of harvesting machinery chassis frames: fatigue life prediction methods fail to identify damage states, damage identification methods lack specificity and sensor optimization support, and there is no effective connection between the two. For scrap chassis frames, the lack of a complete technical solution that can simultaneously achieve damage location, severity quantification, optimized sensor placement, and remaining life prediction makes it difficult for remanufacturing companies to scientifically determine whether scrap chassis frames have repair value. They can only rely on manual experience for rough judgments, making it difficult to guarantee the accuracy and consistency of the assessment results.

[0019] In summary, the development of a method for damage identification and fatigue life prediction of harvesting machinery chassis frames for remanufacturing can accurately identify the damage status of the chassis frame, optimize the arrangement of sensor measuring points, and scientifically assess the remaining fatigue life. This provides an objective basis for determining whether a used chassis frame has remanufacturing and repair value, and has significant practical significance and engineering application value. Summary of the Invention

[0021] The problem this invention aims to solve is that, before remanufacturing and repairing used harvesting machinery chassis frames, existing technologies cannot effectively identify the location, extent, and type of existing damage, lack optimized sensor arrangement methods for damage identification, and fail to consider the impact of existing damage on remaining life in fatigue life prediction. This makes it difficult to scientifically determine whether used chassis frames are worth remanufacturing and repairing. This invention provides a method for damage identification and life prediction of harvesting machinery chassis oriented towards remanufacturing, which can simultaneously achieve accurate damage identification, optimized sensor arrangement, and remaining fatigue life prediction.

[0022] To address the aforementioned problems, this invention provides a method for damage identification and life prediction of harvesting machinery chassis for remanufacturing, comprising the following steps:

[0023] Step S1: Establish a finite element model of the harvesting machinery chassis frame, and perform static and modal analysis on the finite element model to identify the weak points of the chassis frame;

[0024] Step S2: Based on strain modal theory, damage simulation is performed on the weak points in the finite element model. The variation law of modal parameters under different damage conditions is analyzed by numerical simulation. Effective damage identification modal parameters that are sensitive to damage are extracted, and the accuracy of the effective damage identification modal parameters is verified by simply supported beam test.

[0025] Step S3: Based on the effective damage identification modal parameters, the sensor measurement points are optimized using the effective independent driving point residual method to determine the number and location of sensors. The optimized arrangement scheme is then evaluated using the weighted modal confidence criterion to obtain the optimized sensor arrangement scheme.

[0026] Step S4: According to the optimized sensor layout scheme, the sensors are deployed on the actual harvesting machinery, field tests are conducted, strain time history data of weak points in the chassis frame are collected, and load spectrum is compiled based on the strain time history data.

[0027] Step S5: Based on the load spectrum and combined with the material properties of the chassis frame, construct a fatigue life prediction model, calculate the remaining fatigue life of the chassis frame, and determine whether it has remanufacturing and repair value.

[0028] Preferably, identifying the weak points of the chassis frame in step S1 specifically includes:

[0029] Static analysis was performed on the finite element model of the chassis frame to obtain stress distribution cloud maps and identify stress concentration areas;

[0030] Modal analysis was performed on the finite element model of the chassis frame to obtain the location of maximum modal deformation;

[0031] By superimposing the stress concentration area with the location of maximum modal deformation, the weak points of the chassis frame are determined.

[0032] Preferably, the damage simulation in step S2 specifically includes:

[0033] The elastic modulus method is used to simulate damage by reducing the elastic modulus at local locations in the finite element model. The degree of damage is characterized by the following formula:

[0034] ;

[0035] Where Di represents the degree of structural damage, Ei represents the elastic modulus before damage, and Esi represents the elastic modulus after damage.

[0036] Preferably, the effective damage identification modal parameters extracted in step S2 include the strain modal difference, which is calculated by the following formula:

[0037] ;

[0038] The difference in strain modes before damage (order a) It is the strain mode value of order a before damage. It is the value of the a-th strain mode after damage.

[0039] Preferably, step S3 uses the effective independent driving point residual method to optimize the arrangement of sensor measurement points, specifically including:

[0040] Based on the non-destructive finite element model of the chassis frame, strain modal analysis was performed to determine the number of strain modal orders and to construct a candidate measurement point set.

[0041] Calculate the effective independent driving point residual coefficients for each candidate measurement point, and use the driving point residual coefficients to correct the effective independent vectors;

[0042] Candidate points are sorted according to the value of the corrected effective independent vector, and candidate points with small values ​​are eliminated to determine the final location and number of sensor placement points;

[0043] The method for determining the strain mode order is as follows: based on the strain mode matrix corresponding to the candidate point, calculate the rate of change of the second norm of the Fisher information matrix with the increase of the number of modes, plot the rate of change curve, and determine the appropriate mode order when the curve tends to be stable and close to 0.

[0044] Preferably, the driving point residual coefficient is calculated using the following formula:

[0045] ;

[0046] Where CDPRi is the driving point residual coefficient of the i-th measuring point, Φij is the strain mode shape value of the i-th measuring point in the i-th mode, reflecting the deformation characteristics of the measuring point in a specific mode; ωij is the j-th modal frequency, which is the natural frequency of the structure's vibration in that mode, determining the speed of the structure's vibration; N is the modal order; the effective independent vector is corrected using the driving point residual coefficient, and candidate points with small effective independent vector values ​​are eliminated to determine the final sensor placement points and number;

[0047] The method for correcting the effective independent vector using the driving point residual coefficients is as follows:

[0048] ;

[0049] Where Φ is the strain mode vector matrix, Ψ is the eigenvector of the Fisher information matrix Q, λ is the corresponding eigenvalue, and Ik is the sum of all coefficients in the k-th row.

[0050] Preferably, the preparation of the load spectrum in step S4 specifically includes:

[0051] Singularity detection and trend term removal are performed on the strain time history data;

[0052] The rainflow counting method was used to count the processed data, obtain the load amplitude, mean and frequency information, and compile a two-dimensional load spectrum.

[0053] Preferably, the construction of the fatigue life prediction model in step S5 specifically includes:

[0054] Obtain the SN curve of the chassis frame material;

[0055] Fatigue life is calculated using Miner's linear cumulative damage theory. When the cumulative damage reaches a preset threshold, fatigue failure is determined to have occurred. The preset threshold is preferably 0.7.

[0056] Preferably, a damage identification and life prediction system for harvesting machinery chassis for remanufacturing includes:

[0057] The model building and weak point identification module is used to build a finite element model of the chassis frame of the harvesting machinery, and to perform static and modal analysis on the finite element model to identify the weak points of the chassis frame.

[0058] The damage identification parameter determination module is used to simulate damage at weak points in the finite element model based on strain modal theory, analyze the variation law of modal parameters under different damage conditions through numerical simulation, and extract effective damage identification modal parameters that are sensitive to damage.

[0059] The sensor optimization layout module is used to optimize the layout of sensor measurement points based on the effective damage identification modal parameters using the effective independent driving point residual method, determine the number and location of sensors, and evaluate the optimized layout scheme using the weighted modal confidence criterion to obtain the sensor optimization layout scheme.

[0060] The load spectrum acquisition module is used to deploy sensors on actual harvesting machinery according to the optimized sensor layout scheme, conduct field trials, collect strain time history data of weak points in the chassis frame, and compile load spectrum based on the strain time history data.

[0061] The fatigue life prediction module is used to construct a fatigue life prediction model based on the load spectrum and the material properties of the chassis frame, and to calculate the remaining fatigue life of the chassis frame.

[0062] Preferably, an electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor, when executing the program, implements the method for damage identification and fatigue life prediction of the chassis frame of harvesting machinery for remanufacturing.

[0063] Compared with the prior art, the present invention achieves the following beneficial technical effects:

[0064] This invention is the first to construct a complete technical solution for damage identification and life prediction of harvesting machinery chassis for remanufacturing. In existing technologies, fatigue life prediction methods and damage identification methods are fragmented, lacking a complete solution for the remanufacturing assessment of used combined harvester chassis frames. This invention, starting from the actual needs of remanufacturing, organically integrates five stages: weak point identification, damage simulation and sensitive parameter extraction, optimized sensor placement, field load acquisition, and remaining life prediction. This forms a complete technical chain for detection, identification, assessment, and decision-making, filling the technical gap in the field of remanufacturing value judgment of used harvesting machinery chassis frames. It enables remanufacturing companies to scientifically and objectively determine whether used chassis frames have repair value, effectively avoiding the subjectivity and uncertainty caused by relying on rough judgments based on human experience.

[0065] This invention proposes a precise damage identification method for chassis frames based on the synergy of strain modal theory and numerical simulation. Addressing the problem that existing technologies cannot effectively identify the existing damage state of used chassis frames, this invention innovatively applies strain modal theory to damage identification of harvesting machinery chassis frames. By simulating damage conditions at different locations and with varying degrees using the elastic modulus method, it systematically analyzes the sensitivity of various modal parameters, such as natural frequencies, displacement mode shapes, and strain modal differences, to damage. The strain modal difference, which is sensitive to damage, is extracted as a damage identification index, and the numerical simulation results are verified using simply supported beam tests. This method can accurately identify the location, extent, and type of damage, effectively solving the key problem of "where is the damage in the old parts and how severe is it," providing a scientific basis for the formulation of remanufacturing and repair plans.

[0066] This invention proposes a sensor optimization layout method based on the effective independent-driving-point residual method. Addressing the practical engineering constraints of complex chassis and frame structures in harvesting machinery, limited space, and difficulties in densely deploying sensors, this invention employs the effective independent-driving-point residual method to optimize the sensor measurement point layout. The effective independent vector is corrected by the driving-point residual coefficient, taking into account both the spatial independence of modal information and energy distribution characteristics. Simultaneously, a weighted modal confidence criterion is used to evaluate the optimized scheme, ensuring that comprehensive and effective modal parameters of key parts and weak areas are obtained with a smaller number of sensors. This method significantly reduces sensor deployment costs and data acquisition difficulty, improving the economy and engineering feasibility of damage identification solutions.

[0067] This invention achieves a seamless transition from damage identification to remaining life prediction. Existing technologies often assume the structure is in an intact state, failing to consider the impact of existing damage on remaining life. This invention, based on accurate damage identification, utilizes optimized sensors deployed at weak points and damage-sensitive areas to conduct field tests and measure strain-time history data. This data, combined with rainflow counting, is used to compile a load spectrum, and Miner's linear cumulative damage theory is employed to calculate the remaining fatigue life. This approach of first identifying damage and then predicting life ensures that the remaining life prediction results more closely reflect the actual condition of the scrapped chassis frame, providing a more reliable quantitative basis for determining whether repair is worthwhile.

[0068] This invention significantly reduces agricultural production costs and promotes the green development of the agricultural machinery remanufacturing industry. By scientifically identifying damage and assessing the lifespan of discarded harvesting machinery chassis frames, this invention accurately identifies chassis frames with remanufacturing and repair value. After remanufacturing and repair, these frames can be reused, significantly extending their service life and reducing the need for new equipment purchases and energy consumption and resource waste during the manufacturing process. For farmers and agricultural production enterprises, this effectively reduces equipment purchase and replacement costs. From a macro perspective, it helps promote the standardized and scientific development of the agricultural machinery remanufacturing industry, aligning with the national circular economy development strategy and yielding significant economic, social, and environmental benefits. Attached Figure Description

[0070] Figure 1 This is the overall technical roadmap for the method of damage identification and life prediction of harvesting machinery chassis for remanufacturing according to the present invention;

[0071] Figure 2 This is a flowchart of the finite element analysis of the chassis frame of the present invention;

[0072] Figure 3 This is a flowchart illustrating the steps involved in identifying weak points in the chassis frame according to the present invention.

[0073] Figure 4 This is a flowchart illustrating the steps of numerical simulation and simply supported beam test verification for chassis frame damage identification in this invention.

[0074] Figure 5 This is a schematic diagram of the simply supported beam structure used in the simply supported beam damage test of this invention;

[0075] Figure 6 This is a schematic diagram of the strain gauge bonding method and bridge circuit connection in the damage test of simply supported beam of the present invention, wherein (a) is the strain gauge bonding method and (b) is the strain gauge bridge circuit connection method;

[0076] Figure 7 This is a flowchart illustrating the steps involved in the experimental implementation of the simply supported beam of the present invention.

[0077] Figure 8 This is a flowchart illustrating the steps involved in optimizing the sensor arrangement according to the present invention.

[0078] Figure 9 This is a flowchart illustrating the steps involved in predicting fatigue life according to the present invention.

[0079] Figure 10 This is a schematic diagram of the SN curve of the material used in the fatigue life prediction of this invention;

[0080] Figure 11 This is a schematic diagram of the Miner linear cumulative damage principle used in fatigue life prediction in this invention.

[0081] Figure 12 A schematic diagram of the WMAC matrix for sensor arrangement schemes in different directions. Detailed Implementation

[0083] The present invention will be further explained and described below with reference to the accompanying drawings and embodiments.

[0084] Example 1: This example provides a method for damage identification and life prediction of harvesting machinery chassis for remanufacturing, such as... Figure 1 As shown, it includes the following steps S1 to S5.

[0085] Step S1: Establish a finite element model and identify weak points.

[0086] (1) Establish a three-dimensional model of the harvesting machinery chassis frame.

[0087] Taking a certain model of tracked harvesting machinery as an example, its chassis frame is 2280mm long and 2080mm wide, mainly welded from rectangular tubular profiles. The structure is divided into three layers: the bottom layer consists of two track beams connected to the tracks to support the entire structure; the middle layer bears the loads of components such as the engine, threshing and cleaning frame, and grain tank; the top layer is the cab frame. A three-dimensional model of the harvesting machinery chassis frame was created using SolidWorks software.

[0088] (2) Establishing a finite element analysis model

[0089] The chassis frame material properties were determined, and the 3D model was imported into ANSYS Workbench finite element analysis software, referring to Table 1-1. The chassis frame material was Q235 ordinary carbon structural steel, with the following properties: density 7.85 g / cm³, elastic modulus 200 GPa, Poisson's ratio 0.30, and yield strength 225 MPa. Tetrahedral elements were used for mesh generation, with an element size of 10 mm, generating approximately 450,000 elements. Shell elements were used to simulate weld joints, while Rbe2 elements were used for complex areas. Load application was based on the static equivalence principle, converting the mass of each working component, such as the engine, grain tank, and threshing mechanism, into concentrated forces or surface loads acting on their corresponding locations. Boundary conditions were set as follows: all degrees of freedom were constrained at the bottom of the track beam to simulate its fixed connection with the track. The finite element analysis process is as follows: Figure 2 As shown.

[0090] Determine the loads borne by the chassis frame. By consulting relevant literature, the masses of various working components on the chassis frame, such as the threshing mechanism, grain tank, fuel tank cylinder, engine, battery, and cab, can be obtained. According to the principle of static equivalence, the working components will act on the frame in the form of concentrated or uniformly distributed loads. The static loads applied to the frame by the working components are divided into point loads and surface loads.

[0091] The formula for calculating static point load is as follows:

[0092]

[0093] In the formula: P—the surface load, m—mass, g—gravitational acceleration, n—the number of support points

[0094] The formula for calculating static surface load is as follows:

[0095]

[0096] In the formula: p—the surface load, m—mass, g—gravitational acceleration, s—the area of ​​the force application.

[0097] After the above steps, the finite element model of the chassis frame is established.

[0098] physical quantity Symbols and Units numerical values density ρ / g·cm³ 7.85 elastic modulus GPa 200 Poisson's ratio μ 0.3 Yield strength <![CDATA[σ s / MPa]]> 225 elongation δ / % 24 Allowable shear stress <![CDATA[[τ t ] / MPa]]> 15~25 Allowable bending stress <![CDATA[[σ -1 ] / MPa]]> 40 Extrusion strength <![CDATA[σᵦ s / MPa]]> 120~150 (static load)

[0099] Table 1-1 Material Properties of Harvesting Machinery Chassis Frame

[0100] (3) Static analysis

[0101] Static analysis was performed on the finite element model of the chassis frame under two typical working conditions: full-load bending and full-load torsion. Under the full-load bending condition, the maximum stress on the frame occurred at the connection between the third crossbeam and the longitudinal beam, with a stress value of 156 MPa. Under the full-load torsion condition, the maximum stress occurred at the connection between the left front corner and the first crossbeam, with a stress value of 189 MPa. Both values ​​are lower than the yield strength of Q235 steel (225 MPa), but the stress concentration at these two locations is significant, indicating potential weak points.

[0102] (4) Modal analysis

[0103] Modal analysis was performed on the finite element model of the chassis frame to extract the first six natural frequencies and mode shapes. The calculated first six natural frequencies were 28.5Hz, 35.2Hz, 41.8Hz, 53.6Hz, 67.3Hz, and 79.1Hz. The second mode shape showed significant bending deformation in the middle of the frame, with the maximum deformation located at the midpoint of the central longitudinal beam. The fourth mode shape showed torsional deformation at the left front corner of the frame, which basically coincides with the stress concentration area identified in the static analysis.

[0104] (5) Comprehensively determine the weak points

[0105] By superimposing the stress concentration areas with the locations of maximum modal deformation, the weak points of the chassis frame were determined as follows: ① The connection between the third crossbeam and the middle longitudinal beam (i.e., the stress concentration area and the high deformation zone of the modal vibration mode); ② The connection between the left front corner point and the first crossbeam (i.e., the high stress zone under torsional conditions); ③ The midpoint of the middle longitudinal beam (i.e., the location of maximum modal bending deformation). The weak point identification process is as follows: Figure 3 As shown.

[0106] II. Step S2: Numerical Simulation and Experimental Verification of Damage Identification

[0107] (1) Damage simulation scheme

[0108] Damage was simulated using the elastic modulus method. Single-point and multi-point damage conditions were set at the three weak points, with preset elastic modulus reductions of 0%, 20%, 40%, and 60%, corresponding to no damage, mild damage, moderate damage, and severe damage, respectively. The damage severity calculation formula is as follows:

[0109] ;

[0110] Where Di represents the degree of structural damage, Ei represents the elastic modulus before damage, and Esi represents the elastic modulus after damage.

[0111] There are a total of 7 damage conditions: no damage (0%), condition 1 (single point damage 20%), condition 2 (single point damage 40%), condition 3 (single point damage 60%), condition 4 (multiple point damage 20%), condition 5 (multiple point damage 40%), and condition 6 (multiple point damage 60%).

[0112] (2) Modal parameter extraction and damage sensitivity index analysis

[0113] Modal analyses of different orders were performed on the chassis frame model under each damage condition to obtain its natural frequencies, displacement mode shapes, strain mode shapes, and other modal parameters. The strain modal difference was calculated by subtracting the strain modal values ​​before and after the damage, which serves as an important damage identification indicator.

[0114]

[0115] In the formula For strain mode difference, It is the strain mode value of order a before damage. It is the value of the a-th strain mode after damage.

[0116] Analyze the variation patterns of natural frequencies, displacement mode shapes, and strain mode differences under different damage conditions, observe their abrupt changes at the damage location, and determine their sensitivity to damage.

[0117] Analysis revealed that the natural frequency decreases with increasing damage severity, but is insensitive to the damage location; the displacement mode shape does not change significantly at the damage location; the strain mode difference exhibits a significant abrupt change at the damage location, and the magnitude of this abrupt change is positively correlated with the damage severity. Taking a single-point damage of 40% at the connection between the third crossbeam and the middle longitudinal beam as an example, the third-order strain mode difference at this node increases by 42% compared to the undamaged state, while the change is less than 5% further away from the damage location. Therefore, the strain mode difference is determined to be the most effective modal parameter for damage identification. The numerical simulation process for damage identification is as follows: Figure 4 As shown.

[0118] (3) Verification by simply supported beam test

[0119] To verify the accuracy of the numerical simulation results, a simply supported beam made of the same material as the chassis frame (Q235 steel) was selected for damage identification testing. The simply supported beam was 1200mm long and had a cross-sectional dimension of 50mm × 100mm. Cutting cracks were placed at the mid-span of the beam to simulate damage, with crack depths of 0mm (no damage), 5mm (mild), 10mm (moderate), and 15mm (severe). The structure of the simply supported beam is as follows. Figure 5 As shown.

[0120] A rubber hammer was used to excite the simply supported beam by impact. A PCB triaxial accelerometer (model 356A16) was used to measure the vibration acceleration at the damage location. A BE120-3AAP200 strain gauge was used to measure strain changes, thereby obtaining the strain modes. A Donghua 32-channel DHDAS5902N dynamic signal acquisition instrument was used as the data acquisition instrument. The sensor was connected to the acquisition system and computer to realize the acquisition of strain signals and identification of strain mode parameters. The strain gauge was mounted using a half-bridge compensation method. Figure 6 As shown in (a), the bridge connection is as follows: Figure 6 As shown in (b), to eliminate the influence of temperature and improve measurement sensitivity, two strain gauges, R1 and R2, are perpendicularly attached to the test surface of the simply supported beam along the transverse and longitudinal directions.

[0121] When the measured component is subjected to tension, compression, or bending, the strain of strain gauge R1 is:

[0122]

[0123] The strain of strain gauge R2 caused by the load is -m (where m is Poisson's ratio of the material), the strain of R2 caused by temperature is the same as the strain of R1 caused by temperature. At this point, the total strain of R2 is:

[0124]

[0125] Connect the components according to the half-bridge connection method shown in the image. At this point, the strain test will display the strain. Size:

[0126]

[0127] The magnitude of the strain caused by the force on the measured component is expressed as:

[0128]

[0129] The above-described pasting method can eliminate the influence of temperature on the strain gauge, while simultaneously amplifying the measurement readings and improving the sensitivity of strain measurement.

[0130] Simply supported beam test implementation: First, the simply supported beam must be placed stably, supported at both ends with appropriate supports to ensure it is in a simply supported state. Soft sponge is placed on the beam to provide cushioning and protection. Then, sensors are precisely installed at predetermined positions on the simply supported beam and connected to the data acquisition instrument via a data transmission line. During the test, a hammer is used to strike the simply supported beam. The impact force of the hammer causes the beam to vibrate, and the sensors capture the vibration signals in real time, transmitting them to the data acquisition instrument via the data transmission line. The data acquisition instrument collects these signals, converts them into digital signals, and transmits them to the computer. After receiving the time-domain signals, the computer processes them using enhanced frequency domain decomposition, and further analyzes them to obtain important parameters such as natural frequencies and mode shapes. The simply supported beam test implementation process is as follows: Figure 7 As shown.

[0131] The experimental results are compared with the numerical simulation results as follows: for mild damage, the strain modal difference increases by approximately 18% (20% in numerical simulation); for moderate damage, it increases by approximately 38% (42% in numerical simulation); and for severe damage, it increases by approximately 58% (61% in numerical simulation). The relative error between the two methods is less than 5%, verifying the accuracy of the numerical simulation method and the effectiveness of strain modal difference as a damage identification indicator.

[0132] III. Step S3: Sensor Optimization Layout

[0133] (1) Strain modal analysis and candidate point selection

[0134] Strain modal analysis was performed on the non-destructive finite element model of the chassis frame using ANSYS software, and the first six strain modal matrices were extracted. Based on the structural characteristics of the chassis frame and the analysis results of weak points, 120 nodes were initially selected as candidate points for sensor placement, covering key stress areas, weak points, and symmetrical areas of the structure.

[0135] (2) Determine the number of strain mode orders

[0136] The number of strain mode orders is determined by the Rate of Change of the 2 Norm of the Fisher information matrix based on the strain mode shape as the number of modes increases, using the ROC criterion. The strain mode matrices corresponding to candidate points are imported into the ROC criterion calculation program to calculate ROC values ​​for different mode orders, and ROC curves are plotted. The strain mode order is determined based on the trend of the curves.

[0137]

[0138] Where Q represents the Fisher information matrix, and i is the number of strain mode orders.

[0139] As the modal order increases from 1 to 5, the ROC value gradually decreases from 0.32 to 0.08; upon reaching 6, the ROC value is 0.03, stabilizing and approaching 0. When the ROC curve stabilizes and approaches 0 after a certain modal order, this modal order is considered suitable for ensuring the integrity of modal information. Therefore, determining a 6th-order strain mode ensures the integrity of modal information.

[0140] (3) Optimization using the effective independent-driving point residual method

[0141] The sixth-order strain mode matrix was substituted into the effective independent driving point residual method calculation program.

[0142] The effective independent driving point residual method is adopted as the sensor optimization algorithm. Considering the influence of energy distribution, the effective independent method avoids placing the sensor at a low-energy location, thus obtaining more comprehensive modal parameters.

[0143] The strain mode matrix after selecting the modal order is substituted into the effective independent-driving point residual method calculation program for iteration. The program first calculates the effective independence vector according to the principle of the effective independent-driving point residual method. This vector reflects the contribution of each candidate point to the modal information and its spatial independence.

[0144] First, calculate the driving point residual coefficients for each candidate measurement point:

[0145]

[0146] N: Modal order;

[0147] The strain mode shape value of the i-th measuring point in the j-th mode reflects the deformation characteristics of the measuring point in a specific mode.

[0148] The j-th modal frequency is the natural frequency of the structure's vibration in that mode, which determines the speed of the structure's vibration.

[0149] The effective independent assignment vector of the effective independent method, corrected using the driving point residual coefficients, is:

[0150]

[0151] The strain mode vector matrix, composed of strain mode shapes of various orders, reflects the deformation mode of the structure under different strain modes.

[0152] : The eigenvectors of Q, used to describe the characteristics of the strain mode vector matrix.

[0153] The corresponding eigenvalues, together with the eigenvectors, determine the characteristic properties of the matrix.

[0154] The sum of all coefficients in the k-th row, used for normalization or weighting during the calculation process.

[0155] The 120 candidate points were sorted in descending order according to the values ​​of the corrected effective independent vectors, and points with smaller values ​​were removed. When 32 measurement points were retained, the cumulative contribution rate of the effective independent vectors reached 95%; when 20 measurement points were retained, the cumulative contribution rate was 92%; and when 12 measurement points were retained, the cumulative contribution rate was 85%. Considering both recognition accuracy and sensor cost, 20 sensor measurement points were selected as the final optimized arrangement. The sensor optimization arrangement process is as follows: Figure 8 As shown.

[0156] (4) Weighted Modal Confidence Criterion Assessment

[0157] Through iterative calculations, the final placement points and number of sensors were determined. The node number or location coordinates corresponding to each measuring point were recorded, clarifying the specific installation position of the sensor on the chassis frame. The weighted modal confidence criterion (WMAC) was used to evaluate the final measuring point placement scheme, the WMAC matrix was calculated, and the magnitude of the off-diagonal elements in the matrix was analyzed.

[0158] Weighted Modal Confidence Criterion (WMAC):

[0159]

[0160] : The i-th strain mode vector;

[0161] : The j-th strain mode vector;

[0162] M: Weighted matrix.

[0163] The WMAC matrix of a sensor arrangement scheme in different directions of a certain structure is as follows: Figure 12As shown. If the off-diagonal elements in the WMAC matrix are small, it indicates good independence between sensor measurement points, and the measurement point layout scheme can effectively reflect the structural vibration characteristics of the chassis frame, possessing high reliability and accuracy; conversely, further optimization or adjustment of the measurement point layout scheme is required. By plotting the fitting curve, the number of sensors corresponding to the minimum fitting parameters is selected, and this number is used as the final number of sensors employed in the layout scheme.

[0164] This embodiment uses the Weighted Modal Confidence Criterion (WMAC) to evaluate the arrangement of 20 measuring points. The WMAC matrix is ​​calculated; the maximum value of the off-diagonal elements is 0.13, and the average value is less than 0.08, indicating good spatial independence between the measuring points and effectively reflecting the structural vibration characteristics of the chassis frame. If the off-diagonal elements of the WMAC matrix exceed 0.2, the measuring point positions need to be adjusted.

[0165] IV. Step S4: Load Spectrum Compilation

[0166] (1) Field trial data collection

[0167] Based on the optimized sensor layout, BE120-3AAP200 strain gauges were placed at 20 measuring points on the chassis frame of the actual harvesting machinery, connected using a half-bridge configuration. Strain time history data were collected using a DHDAS5902N dynamic signal acquisition instrument. Typical operating conditions were selected: field harvesting (forward speed 0.5 m / s, 1.0 m / s), field transfer (speed 1.5 m / s), road transport (speed 3.0 m / s, 5.0 m / s), and turning. Data was continuously collected for at least 10 minutes under each condition, with a sampling frequency of 500 Hz.

[0168] (2) Data processing

[0169] Singularity detection was performed on the collected strain time history data using wavelet analysis. The principle behind this method is to identify and remove outlier data points caused by environmental interference, system errors, etc., thus avoiding the influence of singularities on subsequent analysis results. Least squares and other techniques were used to remove trend terms from the strain data to prevent distortion of low-frequency signal components and ensure that the data accurately reflects the true stress state of the chassis frame.

[0170] Signals containing a large number of singularities It can be represented as:

[0171]

[0172] : Real signal :noise, Noise standard deviation.

[0173] Under constant loading, the strain gauge output exhibits a gradual trend. The presence of this trend term distorts the analysis of low-frequency components of the signal; therefore, the least squares method is used to remove the trend term. Furthermore, the maximum, minimum, mean, standard deviation, and root mean square values ​​of strain data at each measuring point are calculated under different operating conditions to analyze the variation of stress amplitude with vehicle speed and operating conditions. Based on the stress amplitude variation, the load characteristics of each measuring point are determined, distinguishing between dynamic and static loads, and identifying the locations of measuring points requiring focused attention for fatigue life calculation, thus providing data support for fatigue life calculation.

[0174] This embodiment performs singularity detection on the collected strain time history data and uses wavelet analysis to remove outlier data points; it also uses the least squares method to remove trend terms and eliminate low-frequency drift. Taking field harvesting operations as an example, the root mean square value of the preprocessed strain data is 45 με, and the peak value is 128 με.

[0175] (3) Load spectrum compilation

[0176] Rainflow counting was used to process the pre-processed strain data, and Ncode software was used to process the pre-processed strain time history data to obtain information on load amplitude, mean, and corresponding frequency, thus compiling a two-dimensional load spectrum. During the counting process, parameters for rainflow counting, such as the number of stress amplitude and mean groups, were appropriately set to ensure accurate description of load cycle characteristics. The compiled load spectrum was visualized, and its distribution pattern was analyzed to provide accurate input data for subsequent fatigue life calculations. The stress amplitude was divided into 16 levels, and the mean was divided into 8 levels, thus compiling the two-dimensional load spectrum. The fatigue life prediction process is as follows: Figure 9 As shown.

[0177] V. Step S5: Prediction of Remaining Fatigue Life

[0178] (1) Obtaining the SN curve

[0179] The chassis frame is made of Q235 steel. Referring to the material handbook, its SN curve equation is: σN=C, where m=3.2 and C=1.32×10¹². The SN curve is as follows: Figure 10 As shown.

[0180] (2) Miner linear cumulative damage calculation

[0181] Fatigue life is calculated using Miner's linear cumulative damage theory. The total cumulative damage D is the sum of damage caused by each stress cycle, the principle of which is as follows: Figure 11 As shown:

[0182] In the figure: σ1, σ2, σi, and σj are the load stress values;

[0183] N1, N2, and Ni represent the number of cycles corresponding to different stress values;

[0184] n1, n2, ni, and nj represent the number of stresses applied for different stress values.

[0185] σ−1 is the fatigue limit of the material.

[0186] Assuming D is the critical value for fatigue failure, according to Miner's theory, the damage caused by each cycle of stress σ1 is D / N1. After n1 cycles, the total damage caused by stress σ1 is n1D / N1. Similarly, the total damage caused by stress σ2 is n2D / N2. Since the stress σj in the figure is less than the fatigue limit of the material, it can be ignored in the calculation. And so on, the damage caused by each stress level is niD / Ni. When fatigue failure occurs, we can obtain:

[0187]

[0188] Dividing both sides by D, we get:

[0189]

[0190] Or written as:

[0191]

[0192] According to Miner's cumulative damage theory, fatigue failure occurs when D=1. However, experiments have shown that the fatigue life calculated directly based on Miner's cumulative damage theory needs to be corrected. Therefore, under normal circumstances, the value on the right side of the above equation is not equal to 1, but is replaced by the value 'a', which makes the fatigue life estimation safer and more accurate.

[0193] By calculating the fatigue life of weak points under various typical working conditions, the fatigue performance of the chassis frame in its current state is evaluated, and it is determined whether its remaining life meets the requirements for reuse after technical reinforcement and repair, thus providing a basis for subsequent maintenance and repair decisions.

[0194] When the cumulative damage D reaches the threshold a (a=0.7 in this embodiment), fatigue failure is determined to have occurred. Taking the connection between the third crossbeam and the middle longitudinal beam (weak point 1) as an example, its fatigue life under typical working conditions is calculated to be 1.2×10⁻⁶. 7 Each cycle, equivalent to approximately 2800 hours of work time.

[0195] (3) Remanufacturing value judgment

[0196] Based on the above calculations, the remaining fatigue life of the weak point 1 of the scrapped chassis frame is approximately 2800 hours, which can still meet the usage requirements for 1-2 operating seasons. Furthermore, the damage is mainly repairable surface fatigue cracks, with no penetrating fractures or plastic instability observed. Therefore, the chassis frame is deemed to have remanufacturing and repair value, and a remanufacturing scheme combining welding repair, stress relief treatment, and surface strengthening is recommended.

[0197] Conversely, if the remaining fatigue life at a certain measuring point is less than 500 hours, or if there are multiple severe damages (elastic modulus reduced by more than 60%), it is determined that it is not worth remanufacturing and repair, and scrapping is recommended.

[0198] Example 2: This example provides a damage identification and life prediction system for harvesting machinery chassis for remanufacturing, including:

[0199] Model building and weak point identification module: used to build a finite element model of the harvesting machinery chassis frame, and to perform static and modal analysis on the finite element model to identify the weak points of the chassis frame.

[0200] Damage identification parameter determination module: Based on strain modal theory, it is used to simulate damage at weak points in the finite element model, analyze the variation law of modal parameters under different damage conditions through numerical simulation, and extract effective damage identification modal parameters that are sensitive to damage.

[0201] Sensor Optimization Layout Module: Based on effective damage identification modal parameters, this module uses the effective independent-driving point residual method to optimize the layout of sensor measurement points, determine the number and location of sensors, and evaluate the optimized layout scheme using the weighted modal confidence criterion to obtain the optimal sensor layout scheme.

[0202] Load spectrum acquisition module: It is used to deploy sensors on actual harvesting machinery according to the sensor optimization layout scheme, conduct field tests, collect strain time history data of weak points in the chassis frame, and compile load spectrum based on strain time history data.

[0203] Fatigue life prediction module: Based on the load spectrum and combined with the material properties of the chassis frame, it constructs a fatigue life prediction model and calculates the remaining fatigue life of the chassis frame.

[0204] The specific implementation methods of each module correspond to the steps described in Example 1, and will not be repeated here.

[0205] Example 3: This example provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the method for damage identification and life prediction of harvesting machinery chassis for remanufacturing as described in Example 1.

[0206] Example 4: This example provides a computer-readable storage medium on which a computer program is stored. When executed by a processor, the program implements the method for damage identification and life prediction of harvesting machinery chassis for remanufacturing as described in Example 1.

[0207] In the description of this invention, it should be noted that the terms "upper," "lower," etc., indicating the orientation or positional relationship are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Unless otherwise expressly specified and limited, the terms "installed," "connected," and "linked" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication between two elements. For those skilled in the art, the specific meaning of the above terms in this invention can be understood according to the specific circumstances.

[0208] It should be noted that in this invention, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0209] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention described herein.

Claims

1. A method for damage identification and life prediction of harvesting machinery chassis for remanufacturing, characterized in that, Includes the following steps: Step S1: Establish a finite element model of the harvesting machinery chassis frame, and perform static and modal analysis on the finite element model to identify the weak points of the chassis frame; Step S2: Based on strain modal theory, damage simulation is performed on the weak points in the finite element model. The variation law of modal parameters under different damage conditions is analyzed by numerical simulation. Effective damage identification modal parameters that are sensitive to damage are extracted, and the accuracy of the effective damage identification modal parameters is verified by simply supported beam test. Step S3: Based on the effective damage identification modal parameters, the sensor measurement points are optimized using the effective independent driving point residual method to determine the number and location of sensors. The optimized arrangement scheme is then evaluated using the weighted modal confidence criterion to obtain the optimized sensor arrangement scheme. Step S4: According to the optimized sensor layout scheme, the sensors are deployed on the actual harvesting machinery, field tests are conducted, strain time history data of weak points in the chassis frame are collected, and load spectrum is compiled based on the strain time history data. Step S5: Based on the load spectrum and combined with the material properties of the chassis frame, construct a fatigue life prediction model, calculate the remaining fatigue life of the chassis frame, and determine whether it has remanufacturing and repair value.

2. The method for damage identification and life prediction of harvesting machinery chassis for remanufacturing according to claim 1, characterized in that, Step S1, which identifies the weak points of the chassis frame, specifically includes: Static analysis was performed on the finite element model of the chassis frame to obtain stress distribution cloud maps and identify stress concentration areas; Modal analysis was performed on the finite element model of the chassis frame to obtain the location of maximum modal deformation; By superimposing the stress concentration area with the location of maximum modal deformation, the weak points of the chassis frame are determined.

3. The method for damage identification and life prediction of harvesting machinery chassis for remanufacturing according to claim 1, characterized in that, The damage simulation in step S2 specifically includes: The elastic modulus method is used to simulate damage by reducing the elastic modulus at local locations in the finite element model. The degree of damage is characterized by the following formula: ； Where Di represents the degree of structural damage, Ei represents the elastic modulus before damage, and Esi represents the elastic modulus after damage.

4. The method for damage identification and life prediction of harvesting machinery chassis for remanufacturing according to claim 1, characterized in that, The effective damage identification modal parameters extracted in step S2 include the strain modal difference, which is calculated by the following formula: ； The difference in strain modes before damage (order a) It is the strain mode value of order a before damage. It is the value of the a-th strain mode after damage.

5. The method for damage identification and life prediction of harvesting machinery chassis for remanufacturing according to claim 1, characterized in that, Step S3 uses the effective independent driving point residual method to optimize the arrangement of sensor measurement points, specifically including: Based on the non-destructive finite element model of the chassis frame, strain modal analysis was performed to determine the number of strain modal orders and to construct a candidate measurement point set. Calculate the effective independent driving point residual coefficients for each candidate measurement point, and use the driving point residual coefficients to correct the effective independent vectors; Candidate points are sorted according to the value of the corrected effective independent vector, and candidate points with small values ​​are eliminated to determine the final location and number of sensor placement points; The method for determining the strain mode order is as follows: based on the strain mode matrix corresponding to the candidate point, calculate the rate of change of the second norm of the Fisher information matrix with the increase of the number of modes, plot the rate of change curve, and determine the appropriate mode order when the curve tends to be stable and close to 0.

6. The method for damage identification and life prediction of harvesting machinery chassis for remanufacturing according to claim 5, characterized in that, The driving point residual coefficient is calculated using the following formula: ； Where CDPRi is the driving point residual coefficient of the i-th measuring point, Φij is the strain mode shape value of the i-th measuring point in the i-th mode, reflecting the deformation characteristics of the measuring point in a specific mode; ωij is the j-th modal frequency, which is the natural frequency of the structure's vibration in that mode, determining the speed of the structure's vibration; N is the modal order; the effective independent vector is corrected using the driving point residual coefficient, and candidate points with small effective independent vector values ​​are eliminated to determine the final sensor placement points and number; The method for correcting the effective independent vector using the driving point residual coefficients is as follows: ； Where Φ is the strain mode vector matrix, Ψ is the eigenvector of the Fisher information matrix Q, λ is the corresponding eigenvalue, and Ik is the sum of all coefficients in the k-th row.

7. The method for damage identification and life prediction of harvesting machinery chassis for remanufacturing according to claim 1, characterized in that, Step S4, which involves compiling the load spectrum, specifically includes: Singularity detection and trend term removal are performed on the strain time history data; The rainflow counting method was used to count the processed data, obtain the load amplitude, mean and frequency information, and compile a two-dimensional load spectrum.

8. The method for damage identification and life prediction of harvesting machinery chassis for remanufacturing according to claim 1, characterized in that, Step S5, which involves constructing a fatigue life prediction model, specifically includes: Obtain the SN curve of the chassis frame material; Fatigue life is calculated using Miner's linear cumulative damage theory. When the cumulative damage reaches a preset threshold, fatigue failure is determined to have occurred. The preset threshold is preferably 0.

7.

9. A damage identification and life prediction system for harvesting machinery chassis for remanufacturing, characterized in that, include: The model building and weak point identification module is used to build a finite element model of the chassis frame of the harvesting machinery, and to perform static and modal analysis on the finite element model to identify the weak points of the chassis frame. The damage identification parameter determination module is used to simulate damage at weak points in the finite element model based on strain modal theory, analyze the variation law of modal parameters under different damage conditions through numerical simulation, and extract effective damage identification modal parameters that are sensitive to damage. The sensor optimization layout module is used to optimize the layout of sensor measurement points based on the effective damage identification modal parameters, using the effective independent driving point residual method, to determine the number and location of sensors, and to evaluate the optimized layout scheme using the weighted modal confidence criterion, thereby obtaining the sensor optimization layout scheme. The load spectrum acquisition module is used to deploy sensors on actual harvesting machinery according to the optimized sensor layout scheme, conduct field trials, collect strain time history data of weak points in the chassis frame, and compile load spectrum based on the strain time history data. The fatigue life prediction module is used to construct a fatigue life prediction model based on the load spectrum and the material properties of the chassis frame, and to calculate the remaining fatigue life of the chassis frame.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method for damage identification and fatigue life prediction of harvesting machinery chassis frame for remanufacturing as described in any one of claims 1 to 8.

Images