Non-extreme value evaluation and prediction method
The non-extreme value evaluation and prediction method addresses the limitations of numerical simulation by incorporating 3D reconstruction and machine learning to comprehensively assess product performance, improving accuracy and efficiency.
Patent Information
- Application Number
- CN202411138252.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-19
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-08-19
AI Technical Summary
Numerical simulation in manufacturing often relies on extreme value situations, ignoring nearby influence areas, leading to increased error and inadequate product performance evaluation, and existing prediction methods are costly and time-consuming.
A non-extreme value evaluation and prediction method that involves 3D reconstruction, breadth evaluation using shape factor and gradient mean, and machine learning to accurately assess product performance under various conditions.
Enables accurate and efficient evaluation and prediction of product performance by considering both height and breadth of extreme value regions, reducing errors and costs associated with traditional methods.
Smart Images

Figure CN119066966A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of intelligent manufacturing, and in particular to a non-extreme value evaluation and prediction method. Background Art
[0002] As a supplement to experiments, numerical simulation is of great significance in the manufacturing field, especially when the experimental cost is high and the time is long. However, numerical simulation is often based on extreme value situations and ignores the nearby associated influence areas, which increases the risk of error to a certain extent and is not conducive to the true evaluation of product performance. Therefore, a comprehensive evaluation method is urgently needed to solve the problem of extreme value evaluation in the existing technology.
[0003] In addition to real performance evaluation, the prediction of unknown product performance is also extremely critical. Existing technologies are mostly based on artificial intelligence model training under experimental conditions, which greatly increases the experimental sample data, is costly, and takes a long time. In contrast, numerical simulation can provide more samples and is more popular. Therefore, based on the proposed comprehensive evaluation method, it is extremely important to conduct sample data training based on numerical simulation, which provides a new method for developing new products and predicting excellent performance.
[0004] In view of this, it is urgent and necessary to study non-extreme comprehensive evaluation and prediction methods. Summary of the invention
[0005] The embodiment of the present invention provides a non-extreme value evaluation and prediction method to solve the problem that the numerical simulation data cannot be accurately evaluated and predicted in the prior art.
[0006] In order to solve the above-mentioned technical problems, the present invention is achieved as follows:
[0007] The embodiment of the present invention provides a non-extreme value evaluation and prediction method for truly evaluating and predicting product performance, comprising the following steps:
[0008] S1: Establish a model based on product performance requirements and conduct at least one numerical simulation, and collect sample data of simulation results;
[0009] S2: 3D reconstruction or 2D marking of the sample extreme value region data, and determining breadth evaluation data, wherein a calculation method of the breadth evaluation includes at least one of a shape factor and a gradient mean;
[0010] S3: comprehensively evaluating the performance based on the height and breadth of the extreme value region;
[0011] S4: Train the sample data with a machine learning algorithm and predict the product performance under different conditions in real time.
[0012] Optionally, the at least one numerical simulation includes under the same operating condition or under different operating conditions; the acquisition method includes under the same operating condition or under different operating conditions; the acquisition method includes under the same operating condition or under different operating conditions;
[0013] The acquisition method needs to acquire sample data at different angles under the same color band condition.
[0014] Optionally, the sample data includes at least one of an image, text or video; and the marking includes obtaining an area within a specific color range boundary of the image.
[0015] Optionally, the height is the maximum product performance requirement obtained through the numerical simulation, and the requirement includes at least one of stress, displacement, temperature, pressure, speed, and concentration.
[0016] Optionally, the breadth evaluation refers to evaluating the sample extreme value area data marked under the same working conditions;
[0017] The breadth evaluation method includes the product of the shape factor and the gradient mean.
[0018] Optionally, the gradient mean calculation method includes the equivalent circle gradient mean of the extreme area or the mean of the height value and the boundary value.
[0019] Optionally, the shape factor calculation method includes 4*π*extreme region area / (extreme region perimeter*extreme region perimeter).
[0020] Optionally, the machine learning includes at least one of deep learning, reinforcement learning, and swarm intelligence algorithms.
[0021] Optionally, a method for comprehensively evaluating the performance is height value-shape factor*gradient mean.
[0022] Optionally, the three-dimensional reconstruction refers to reconstructing the surface images obtained from different angles of the numerical simulation image to accurately obtain the sample extreme value area data.
[0023] In the embodiment of the present invention, the performance of the extreme value region is comprehensively evaluated by considering the height and width of the extreme value region, thereby avoiding the one-sided influence caused by only the extreme value; the influence of the shape in the width is considered by setting the shape factor, and the closer to a circle, the safer it is; the gradient mean is set to achieve accurate measurement of the gradient change in the extreme value region; the performance under different conditions is predicted by a machine learning algorithm; this method can estimate and predict the extreme value region in two-dimensional images and three-dimensional models; and solves the problem that the prior art cannot accurately evaluate and predict numerical simulation data. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.
[0025] Figure 1 The steps of the non-extreme value evaluation prediction method provided by the embodiment of the present invention are shown;
[0026] Figure 2 A training and prediction diagram showing a non-extreme value evaluation prediction method provided by an embodiment of the present invention;
[0027] Figure 3 Represents the equivalent circular gradient calculation of the non-extreme value evaluation prediction method provided by an embodiment of the present invention; DETAILED DESCRIPTION
[0028] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0029] It should be understood that the references to "one embodiment" or "an embodiment" throughout the specification mean that the specific features, structures, or characteristics associated with the embodiment are included in at least one embodiment of the present invention. Therefore, the references to "in one embodiment" or "in an embodiment" appearing throughout the specification do not necessarily refer to the same embodiment. In addition, these specific features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
[0030] See also Figures 1 to 3 The embodiment of the present invention provides a non-extreme value evaluation prediction method, which is characterized by being used to truly evaluate and predict product performance, and includes the following steps:
[0031] S1: Establish a model based on product performance requirements and conduct at least one numerical simulation, and collect sample data of simulation results;
[0032] In the embodiment of the present invention, the numerical simulation includes multiple scales, from microscopic, mesoscopic to macroscopic, and the present invention is preferably macroscopic finite element simulation;
[0033] The finite element simulation (FEM) specifically includes simulations in different directions such as structure, fluid, dynamics, etc.; it also includes simulations by different methods such as finite volume.
[0034] The product performance requirements include at least one of stress, displacement, temperature, pressure, speed, and concentration. The multiple situations are generally coupled. For example, stress and temperature coupling can obtain thermal stress, which is the mechanical change of the product due to the effect of heat;
[0035] The commonly used finite element software includes but is not limited to ANSYS, HyperWorks, etc.;
[0036] The model building preferably builds a finite element / finite volume model. Building a finite element model requires grid division, loading conditions or working conditions, and debugging operations, and the data obtained is often image data;
[0037] When conducting a finite element simulation and there is a lack of sample data, a sample database can be established by changing different angles and sizes;
[0038] Carrying out multiple finite element simulations requires changing different working conditions; when changing the working conditions to make samples, the corresponding extreme value areas can also be accurately identified.
[0039] S2: 3D reconstruction or 2D marking of the sample extreme value region data, and determining breadth evaluation data, wherein a calculation method of the breadth evaluation includes at least one of a shape factor and a gradient mean;
[0040] In the embodiment of the present invention, the three-dimensional reconstruction uses the corresponding transformation matrix based on the images at different angles to obtain the corresponding three-dimensional model;
[0041] It should be noted that, in the embodiment of the present invention, two-dimensional marking refers to marking the image data using existing marking software;
[0042] Existing labeling software includes labelme, labelimg, etc.
[0043] The breadth evaluation data refers to sample data for marking and identifying the extreme value range required in the calculated result data; the height evaluation data refers to the extreme values calculated in the finite element software, such as the maximum principal stress, the maximum equivalent stress, etc.
[0044] The breadth evaluation calculation method may include a shape factor or a gradient mean or both;
[0045] The shape factor is a factor that considers the influencing factor of the boundary shape within the extreme value region (the desired target region);
[0046] The gradient mean refers to the average value of the performance change in the extreme value area. For example, the marked area is the maximum principal stress range, the color is red, the principal stress range is 30-40MPa, the maximum principal stress is 40MPa, and the gradient changes from 40MPa to 30MPa. The so-called gradient mean refers to the average value of this range.
[0047] S3: comprehensively evaluating the performance based on the height and breadth of the extreme value region;
[0048] In the embodiment of the present invention, it is inaccurate to evaluate the performance of the entire model only by height, and the comprehensive evaluation by height and breadth is jointly determined by the combination of the two.
[0049] S4: Train the sample data with a machine learning algorithm and predict the product performance under different conditions in real time.
[0050] Based on the samples of the above comprehensive evaluation, a machine learning algorithm is used to train the model;
[0051] The machine learning algorithms include existing deep learning, reinforcement learning, swarm intelligence algorithms, etc., and also include Yolo, R-CNN, etc.;
[0052] In the embodiment of the present invention, the performance of the extreme value region is comprehensively evaluated by considering the height and width of the extreme value region, thereby avoiding the one-sided influence caused by only the extreme value; the influence of the shape in the width is considered by setting the shape factor, and the closer to a circle, the safer it is; the gradient mean is set to achieve accurate measurement of the gradient change in the extreme value region; the performance under different conditions is predicted by a machine learning algorithm; this method can estimate and predict the extreme value region in two-dimensional images and three-dimensional models; and solves the problem that the prior art cannot accurately evaluate and predict numerical simulation data.
[0053] Optionally, the at least one numerical simulation includes the same working conditions or different working conditions; the acquisition method includes different angles under the same working conditions or different angles under the same color band conditions under different working conditions.
[0054] In the embodiment of the present invention, the color band refers to a color bar in the finite element simulation result image.
[0055] Optionally, the sample data includes at least one of an image, text or video; and the marking includes obtaining an area within a specific color range boundary of the image.
[0056] It should be noted that the existing labeling software is preferably segmentation software, such as labelme.
[0057] Optionally, the height is the maximum product performance requirement obtained through the numerical simulation, and the requirement includes at least one of stress, displacement, temperature, pressure, speed, and concentration.
[0058] Preferably, the breadth evaluation refers to evaluating the sample extreme value area data marked under the same working conditions;
[0059] The breadth evaluation method includes the product of the shape factor and the gradient mean.
[0060] Preferably, the gradient mean calculation method includes the equivalent circle gradient mean of the extreme value region or the uniform gradient mean of the extreme value region.
[0061] It should be noted that the calculation method of the equivalent circular gradient mean can be used for uniform or non-uniform gradients; the non-uniform gradient is based on the gradient assumption that the gradient changes non-uniformly, and the uniform gradient is based on the gradient assumption that the gradient is uniform.
[0062] The calculation method of the equivalent circle gradient mean is the difference between the height value and the regional mean. The regional mean includes the sum of the area ratios of the inner mean of the inscribed circle in the extreme value area and the outer mean of the inscribed circle. The inner mean of the inscribed circle is the gradient average of the inscribed circle area ratio, and the outer mean of the inscribed circle is equivalent to the boundary value.
[0063] Assuming that the gradient is non-uniform, the specific solution method for the equivalent circular gradient is:
[0064] Assume the extreme value area (area S, perimeter L), the radius of the inscribed circle is r, and the amplitude is: Fz = Max-Min,
[0065] The mean value of the inscribed circle non-uniform stress is Td, the ratio of the inscribed circle to the extreme area is B = π*r*r / S, and the mean value of the gradient is Ave = Max-(B*Td+(1-B)*Min);
[0066] The solution of Td can be obtained by equivalent or integral.
[0067] Under the condition of uniform gradient, the solution method also refers to the above formula, the difference is that the value of the mean stress Td of the inscribed circle is the average of the sum of the height value and the inscribed circle boundary value.
[0068] It should be noted that the uniform gradient mean is assumed to be under the condition of uniform gradient in the extreme value region, and can also be used to solve non-uniform gradient.
[0069] The uniform gradient mean value is calculated by the difference between the height value and the regional mean value, and the regional uniform mean value is the average of the sum of the height value and the boundary value.
[0070] The calculation method is: assuming that the height value is Max, the boundary value is Min, and the gradient mean is Ave=Max-(Max+Min) / 2;
[0071] Preferably, the shape factor (Xz) calculation method includes 4*π*extreme region area S / (extreme region perimeter L*extreme region perimeter L).
[0072] Optionally, the machine learning includes at least one of deep learning, reinforcement learning, and swarm intelligence algorithms.
[0073] Preferably, the machine learning is deep learning.
[0074] Optionally, a method for comprehensively evaluating the performance is height value-shape factor*gradient mean.
[0075] It should be noted that the comprehensive evaluation P = Max-Xz*Ave;
[0076] Optionally, the three-dimensional reconstruction refers to reconstructing the surface images obtained from different angles of the numerical simulation image to accurately obtain the sample extreme value area data.
[0077] Example 1 Comprehensive Evaluation
[0078] Assuming that the extreme value area is an irregular image with non-uniform gradient, the maximum principal stress is 320MPa, the boundary of the extreme value area is 280Mpa, the ratio of the inscribed circle to the overall area is 4 / 5, and the non-uniform gradient of the inscribed circle is: 15, then the gradient mean is: Ave = 320-(0.8*15+0.2*280) = 12MPa, the shape factor is 0.92, then the breadth evaluation is G = 11.04, and the comprehensive evaluation is: P = 320-11.04;
[0079] Example 2 Simulation prediction
[0080] The recognition algorithm yolov10 in deep learning is used to identify and predict the thermal stress influence law of the above model. Through labelme marking, the maximum heat affected area is used as the recognition area for training. In the model training, the present invention identifies and calculates the number of image pixel values in the thermal stress affected area under different temperature conditions. After training, this paper monitors and verifies the effectiveness of the model at 200 degrees Celsius. The recognition detection finds that the pixel value size under 200℃ is 10448, the data prediction value is 10530, and the accuracy rate is 99.2%.
[0081] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.
[0082] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present invention.
[0083] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which are within the protection of the present invention.
Claims
1. Non-extreme value evaluation prediction method, characterized in that: Used to truly evaluate and predict product performance, including the following steps: S1: Establish a model according to product performance requirements and perform at least one numerical simulation, and collect simulation results as sample data; S2: 3D reconstructing or 2D marking the sample extreme value region data, and determining breadth evaluation data, wherein the breadth evaluation includes at least one of a shape factor and a gradient mean; S3: comprehensively evaluating the performance based on the height and breadth of the extreme value region; S4: Train the sample data with a machine learning algorithm and predict the performance of the product under different conditions in real time.
2. The non-extreme value evaluation prediction method according to claim 1, characterized in that: The at least one numerical simulation includes the same working condition or different working conditions; the acquisition method includes the same working condition or different working conditions; The acquisition method needs to acquire sample data at different angles under the same color band condition.
3. The non-extreme value evaluation prediction method according to claim 1, characterized in that: The sample data includes at least one of an image, text or video; the marking includes obtaining an area within a specific color range boundary of the image.
4. The non-extreme value evaluation prediction method according to claim 1, characterized in that: The height is the maximum performance requirement of the product obtained through the numerical simulation, and the requirement includes at least one of stress, displacement, temperature, pressure, speed, and concentration.
5. The non-extreme value evaluation prediction method according to claim 1, characterized in that: The breadth evaluation refers to the evaluation of the marked sample extreme value area data under the same working conditions; The breadth evaluation method includes the product of the shape factor and the gradient mean.
6. The non-extreme value evaluation prediction method according to claim 5, characterized in that: The gradient mean calculation method includes the equivalent circular gradient of the extreme value region or the uniform gradient of the extreme value region; The calculation method of the equivalent circle gradient mean is the difference between the height value and the regional mean, the regional mean includes the sum of the area ratios of the inner mean of the inscribed circle and the outer mean of the inscribed circle in the extreme value region, the inner mean of the inscribed circle is the gradient average calculated by the ratio of the inscribed circle area, and the outer mean of the inscribed circle is equivalent to the boundary value; The uniform gradient mean value is calculated by the difference between the height value and the regional mean value, and the regional uniform mean value is the average of the sum of the height value and the boundary value.
7. The non-extreme value evaluation prediction method according to claim 5, characterized in that: The shape factor calculation method includes 4*π*extreme region area / (extreme region perimeter*extreme region perimeter).
8. The non-extreme value evaluation prediction method according to claim 1, characterized in that: The machine learning includes at least one of deep learning, reinforcement learning, and swarm intelligence algorithms.
9. The non-extreme value evaluation prediction method according to claim 1, characterized in that: The method for comprehensively evaluating the performance is height value-shape factor*gradient mean.
10. The non-extreme value evaluation prediction method according to claim 1, characterized in that: The three-dimensional reconstruction refers to the reconstruction of the surface images obtained from different angles of the numerical simulation images to accurately obtain the sample extreme value area data.
Citation Information
Patent Citations
Notched member multi-axis fatigue life prediction method coupled with stress gradient correction
CN108627328A
Paper tube damage defect detection method based on shape features
CN114494225A
System for practicing of artificial intelligence vision model learning using quality evaluation and method thereof
KR102673850B1
