Automatic detection method applied to flexible disc assembly
By using multimodal fusion algorithm and neural network model of SIN chaotic sequence on the flexible disk assembly production line, combined with camera and laser displacement sensor, the abnormal points of the flexible disk are automatically detected, solving the inaccuracy of artificial visual detection and Visual fatigue problems improve the reliability and efficiency of detection.
Patent Information
- Application Number
- CN202411070762.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-06
- Publication Date
- 2024-11-22
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
On flexible disk assembly production lines, manual visual inspection cannot guarantee 100% accuracy, and long-term visual inspection leads to visual fatigue and the risk of neglecting bad products.
Using a multimodal fusion algorithm based on SIN chaotic sequence and anomaly point detection neural network model, image and distance data are obtained through on-site cameras and laser displacement sensors, and abnormal points of flexible disks are automatically detected to reduce the visual inspection time of operators. and labor intensity.
Accurate detection of flexible disks is achieved, the reliability and computing efficiency of detection are improved, and the computational complexity and burden on operators is reduced.
Smart Images

Figure CN119007171A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of flexible disk detection, and in particular to an automatic detection method applied to flexible disk assembly. Background Art
[0002] As OEMs continue to tighten quality control, we must continue to explore and improve corresponding quality inspection means and methods during on-site production.
[0003] On the flexible disk assembly production line, the assembled products need to be inspected to see whether their gaskets are well attached and whether the positive and negative directions of the gear ring welding are correct. The inspection method used is manual visual inspection.
[0004] This method requires the operator to visually observe the workpiece and determine by visual inspection whether the gear ring welding direction is correct and the gasket is well attached. The disadvantage is that it cannot guarantee 100% implementation of visual inspection, and visual fatigue during long-term visual inspection may lead to neglect of defects and outflow. Summary of the invention
[0005] In view of the above problems, the present invention provides an automatic detection method for flexible disk assembly, which not only reduces the operator's visual inspection time and reduces labor intensity, but also does not require the operator to visually judge the results. The sensor automatically measures and outputs the results, thereby improving the reliability of detection.
[0006] In order to achieve the above-mentioned purpose and other related purposes, the technical solution provided by the present invention is as follows:
[0007] An automatic detection method for flexible disk assembly, the method comprising:
[0008] M1. Place the flexible disk assembly to be tested on the detection platform, obtain the image data information of the flexible disk assembly in real time based on the on-board camera, and obtain the distance data information of the flexible disk assembly in real time based on the on-board laser displacement sensor;
[0009] M2. Based on the distance data information of the flexible disk assembly and the image data information of the flexible disk assembly, the distance and image of the flexible disk assembly are fused using a multimodal fusion algorithm based on a SIN chaotic sequence to obtain the fusion data information of the distance and image of the flexible disk assembly;
[0010] M3. Based on the fusion data information of the distance and image of the flexible disk assembly, a neural network model for abnormal point detection of the flexible disk assembly is constructed, and the abnormal hyperparameters of the flexible disk assembly are inferred to obtain abnormal hyperparameter data information of the flexible disk assembly;
[0011] M4. Based on the abnormal hyperparameter data information of the flexible disk assembly, establish an abnormal value function F of the flexible disk assembly, calculate the abnormal value of the flexible disk assembly, and output the abnormal value data information of the flexible disk assembly.
[0012] Furthermore, the method further comprises:
[0013] M5. Based on the abnormal value data information of the flexible disk assembly, a preset threshold is set. If the abnormal value of the flexible disk assembly is greater than the preset threshold, the flexible disk assembly is unqualified and is discarded. If the abnormal value of the flexible disk assembly is less than the preset threshold, the flexible disk assembly is qualified and meets the production standard.
[0014] Furthermore, in step M2, the step of fusing the distance and image of the flexible disk assembly using a multi-modal fusion algorithm based on a SIN chaotic sequence includes:
[0015] M21. Based on the distance data information of the flexible disk assembly and the image data information of the flexible disk assembly, a SIN chaotic mapping function G of the image and distance of the flexible disk assembly is established.
[0016] Among them, x is the distance data information of the flexible disk assembly, y is the image data information of the flexible disk assembly, f is the relationship function between the image and distance of the flexible disk assembly, g is the SIN function of the image and distance of the flexible disk assembly, α 1 , α 2 and α 3 is the image and distance SIN chaos factor of the flexible disk assembly;
[0017] M22. Based on the SIN chaotic mapping function G of the image and distance of the flexible disk assembly, the SIN chaotic sequence of the image and distance of the flexible disk assembly is characterized to obtain data information of the SIN chaotic sequence of the image and distance of the flexible disk assembly;
[0018] M23 establishes a multimodal fusion function H of the flexible disk assembly based on the data information of the SIN chaotic sequence of the image and distance of the flexible disk assembly.
[0019]
[0020] Among them, z is the data information of the SIN chaotic sequence of the image and distance of the flexible disk assembly, β 1 , β 2 and β 3 is the multimodal fusion factor of the flexible disk assembly, and the distance and image of the flexible disk assembly are fused to obtain fused data information of the distance and image of the flexible disk assembly.
[0021] Furthermore, the relationship function f between the image and the distance of the flexible disk assembly is,
[0022]
[0023] The SIN function g of the image and distance of the flexible disk assembly is,
[0024]
[0025] Wherein, x is the distance data information of the flexible disk assembly, and y is the image data information of the flexible disk assembly.
[0026] Furthermore, the image and distance SIN chaos factor α of the flexible disk assembly 1 , α 2 and α 3 The constraints are:
[0027]
[0028] The multimodal fusion factor β of the flexible disk assembly 1 , β 2 and β 3 The constraint function h is,
[0029]
[0030] Among them, the value range of the constraint function h is (1,2).
[0031] Furthermore, in step M3, the construction of a neural network model for detecting abnormal points of the flexible disk assembly includes:
[0032] M31. Based on the distance and image fusion data information of the flexible disk assembly, a feature extraction function W of the flexible disk assembly is constructed,
[0033]
[0034] Among them, r is the distance of the flexible disk assembly and the fusion data information of the image, λ 1 , 2 and λ 3 is the feature extraction factor of the flexible disk assembly, and the fusion data of the flexible disk assembly is subjected to feature extraction to obtain the fusion feature matrix data information of the distance and image of the flexible disk assembly;
[0035] M32. Inputting the fusion feature data information of the distance and image of the flexible disk assembly into the neural network model for abnormal point detection of the flexible disk assembly for training and learning, and determining the abnormal hyperparameter function S of the flexible disk neural network model,
[0036]
[0037] Where a is the distance of the flexible disk assembly and the fusion feature data information of the image, η 1 and η 2 As the predictor of the abnormal hyperparameters of the flexible disk, the trained neural network model for abnormal point detection of the flexible disk assembly is obtained;
[0038] M33. Based on the trained neural network model for abnormal point detection of the flexible disk assembly, the distance of the flexible disk assembly and the fusion feature data information of the image are input, the abnormal hyperparameters of the flexible disk assembly are inferred, and the abnormal hyperparameter data information of the flexible disk assembly is obtained.
[0039] Furthermore, the feature extraction factor λ of the flexible disk assembly 1 for,
[0040]
[0041] The feature extraction factor λ of the flexible disk assembly 2 for,
[0042]
[0043] The feature extraction factor λ of the flexible disk assembly 3 for,
[0044]
[0045] Among them, r is the distance of the flexible disk assembly and the fusion data information of the image.
[0046] Furthermore, the feature extraction factor λ of the flexible disk assembly 1 , 2 and λ 3 The constraint function v is,
[0047]
[0048] The value range of v is (2,3).
[0049] Furthermore, the abnormal value function F of the flexible disk assembly is,
[0050]
[0051] Among them, c is the abnormal hyperparameter data information of the flexible disk assembly, ω 1 ,ω 2 and ω 3 is the outlier prediction factor of the flexible disk assembly.
[0052] Furthermore, the abnormal value prediction factor ω of the flexible disk assembly1 ,ω 2 and ω 3 The constraints are:
[0053]
[0054] The present invention has the following positive effects:
[0055] 1. The present invention fuses the distance and image of the flexible disk assembly by adopting a multimodal fusion algorithm based on the SIN chaotic sequence, and combines it with a neural network model for abnormal point detection of the flexible disk assembly to infer the abnormal hyperparameters of the flexible disk assembly. It can not only accurately detect each abnormal point of the flexible disk assembly, but also reduce the operator's visual inspection time and labor intensity. There is no need for the operator to visually judge the result. The sensor automatically measures and outputs the result, which improves the reliability of detection.
[0056] 2. The present invention establishes an abnormal value function F of the flexible disk assembly to infer the abnormal value of the flexible disk assembly, which can not only make an overall judgment on the abnormal value of the flexible disk assembly, but also reduce the calculation complexity and improve the overall calculation efficiency and scalability of the system. BRIEF DESCRIPTION OF THE DRAWINGS
[0057] Figure 1 It is a schematic diagram of the method flow of the present invention;
[0058] Figure 2 It is a flow chart of the multi-modal fusion algorithm based on SIN chaotic sequence of the present invention;
[0059] Figure 3 A schematic diagram of the process of constructing a neural network model for abnormal point detection of a flexible disk assembly according to the present invention;
[0060] Figure 4 It is a structural schematic diagram of the present invention.
[0061] Explanation of the numbers in the figure: 1-detection platform, 2-flexible disk assembly to be tested, 3-camera, 4-laser radar. DETAILED DESCRIPTION
[0062] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.
[0063] Example 1: Figure 1 or Figure 4As shown, an automatic detection method for flexible disk assembly, the method comprising:
[0064] M1. The flexible disk assembly to be tested is placed on the detection platform 1, and the image data information of the flexible disk assembly is obtained in real time based on the onboard camera 3, and the distance data information of the flexible disk assembly is obtained in real time based on the onboard laser displacement sensor 4;
[0065] M2. Based on the distance data information of the flexible disk assembly and the image data information of the flexible disk assembly, the distance and image of the flexible disk assembly are fused using a multimodal fusion algorithm based on a SIN chaotic sequence to obtain the fusion data information of the distance and image of the flexible disk assembly;
[0066] M3. Based on the fusion data information of the distance and image of the flexible disk assembly, a neural network model for abnormal point detection of the flexible disk assembly is constructed, and the abnormal hyperparameters of the flexible disk assembly are inferred to obtain abnormal hyperparameter data information of the flexible disk assembly;
[0067] M4. Based on the abnormal hyperparameter data information of the flexible disk assembly, establish an abnormal value function F of the flexible disk assembly, calculate the abnormal value of the flexible disk assembly, and output the abnormal value data information of the flexible disk assembly.
[0068] In this embodiment, the method further includes:
[0069] M5. Based on the abnormal value data information of the flexible disk assembly, a preset threshold is set. If the abnormal value of the flexible disk assembly is greater than the preset threshold, the flexible disk assembly is unqualified and is discarded. If the abnormal value of the flexible disk assembly is less than the preset threshold, the flexible disk assembly is qualified and meets the production standard.
[0070] In this embodiment, if Figure 2 As shown, in step M2, the method of fusing the distance and image of the flexible disk assembly using a multi-modal fusion algorithm based on a SIN chaotic sequence includes:
[0071] M21. Based on the distance data information of the flexible disk assembly and the image data information of the flexible disk assembly, a SIN chaotic mapping function G of the image and distance of the flexible disk assembly is established.
[0072] Among them, x is the distance data information of the flexible disk assembly, y is the image data information of the flexible disk assembly, f is the relationship function between the image and distance of the flexible disk assembly, g is the SIN function of the image and distance of the flexible disk assembly, α 1 , α 2 and α 3 is the image and distance SIN chaos factor of the flexible disk assembly;
[0073] M22. Based on the SIN chaotic mapping function G of the image and distance of the flexible disk assembly, the SIN chaotic sequence of the image and distance of the flexible disk assembly is characterized to obtain data information of the SIN chaotic sequence of the image and distance of the flexible disk assembly;
[0074] M23 establishes a multimodal fusion function H of the flexible disk assembly based on the data information of the SIN chaotic sequence of the image and distance of the flexible disk assembly.
[0075]
[0076] Among them, z is the data information of the SIN chaotic sequence of the image and distance of the flexible disk assembly, β 1 , β 2 and β 3 is the multimodal fusion factor of the flexible disk assembly, and the distance and image of the flexible disk assembly are fused to obtain fused data information of the distance and image of the flexible disk assembly.
[0077] In this embodiment, the relationship function f between the image and the distance of the flexible disk assembly is,
[0078] The SIN function g of the image and distance of the flexible disk assembly is,
[0079]
[0080] Wherein, x is the distance data information of the flexible disk assembly, and y is the image data information of the flexible disk assembly.
[0081] In this embodiment, the image and distance SIN chaos factor α of the flexible disk assembly 1 , α 2 and α 3 The constraints are:
[0082]
[0083] The multimodal fusion factor β of the flexible disk assembly 1 , β 2 and β 3 The constraint function h is,
[0084]
[0085] Among them, the value range of the constraint function h is (1,2).
[0086] In this embodiment, multimodal fusion refers to the fusion of multiple different types of data modalities into a unified system to improve the overall performance and scalability of the system. The multimodal fusion factors will also be adaptively adjusted to ensure the integrity and fit of the distance and image fusion of the flexible disk assembly, thereby improving the overall stability of the system.
[0087] Embodiment 2: Based on the method for automatic detection of flexible disk assembly in Embodiment 1, the present invention is further illustrated and described below.
[0088] like Figure 1 or Figure 4 As shown, an automatic detection method for flexible disk assembly, the method comprising:
[0089] M1. The flexible disk assembly to be tested is placed on the detection platform 1, and the image data information of the flexible disk assembly is obtained in real time based on the onboard camera 3, and the distance data information of the flexible disk assembly is obtained in real time based on the onboard laser displacement sensor 4;
[0090] M2. Based on the distance data information of the flexible disk assembly and the image data information of the flexible disk assembly, the distance and image of the flexible disk assembly are fused using a multimodal fusion algorithm based on a SIN chaotic sequence to obtain the fusion data information of the distance and image of the flexible disk assembly;
[0091] M3. Based on the fusion data information of the distance and image of the flexible disk assembly, a neural network model for abnormal point detection of the flexible disk assembly is constructed, and the abnormal hyperparameters of the flexible disk assembly are inferred to obtain abnormal hyperparameter data information of the flexible disk assembly;
[0092] M4. Based on the abnormal hyperparameter data information of the flexible disk assembly, establish an abnormal value function F of the flexible disk assembly, calculate the abnormal value of the flexible disk assembly, and output the abnormal value data information of the flexible disk assembly.
[0093] In this embodiment, if Figure 3 As shown, in step M3, the construction of a neural network model for detecting abnormal points of the flexible disk assembly includes:
[0094] M31. Based on the distance and image fusion data information of the flexible disk assembly, a feature extraction function W of the flexible disk assembly is constructed,
[0095]
[0096] Among them, r is the distance of the flexible disk assembly and the fusion data information of the image, λ 1 , 2 and λ 3is the feature extraction factor of the flexible disk assembly, and the fusion data of the flexible disk assembly is subjected to feature extraction to obtain the fusion feature matrix data information of the distance and image of the flexible disk assembly;
[0097] M32. Inputting the fusion feature data information of the distance and image of the flexible disk assembly into the neural network model for abnormal point detection of the flexible disk assembly for training and learning, and determining the abnormal hyperparameter function S of the flexible disk neural network model,
[0098]
[0099] Where a is the distance of the flexible disk assembly and the fusion feature data information of the image, η 1 and η 2 As the predictor of the abnormal hyperparameters of the flexible disk, the trained neural network model for abnormal point detection of the flexible disk assembly is obtained;
[0100] M33. Based on the trained neural network model for abnormal point detection of the flexible disk assembly, the distance of the flexible disk assembly and the fusion feature data information of the image are input, the abnormal hyperparameters of the flexible disk assembly are inferred, and the abnormal hyperparameter data information of the flexible disk assembly is obtained.
[0101] In this embodiment, the feature extraction factor λ of the flexible disk assembly 1 for,
[0102]
[0103] The feature extraction factor λ of the flexible disk assembly 2 for,
[0104]
[0105] The feature extraction factor λ of the flexible disk assembly 3 for,
[0106]
[0107] Among them, r is the distance of the flexible disk assembly and the fusion data information of the image.
[0108] In this embodiment, the feature extraction factor λ of the flexible disk assembly 1 , 2 and λ 3 The constraint function v is,
[0109]
[0110] The value range of v is (2,3).
[0111] In this embodiment, the abnormal value function F of the flexible disk assembly is,
[0112]
[0113] Among them, c is the abnormal hyperparameter data information of the flexible disk assembly, ω 1 ,ω 2 and ω 3 is the outlier prediction factor of the flexible disk assembly.
[0114] In this embodiment, the abnormal value prediction factor ω of the flexible disk assembly 1 ,ω 2 and ω 3 The constraints are:
[0115]
[0116] In this embodiment, the present invention provides an automatic detection system for flexible disk assembly, including a computer device, which is programmed or configured to execute any one of the steps of the automatic detection method for flexible disk assembly.
[0117] In this embodiment, the present invention provides a computer-readable storage medium storing a computer program programmed or configured to execute any one of the automatic detection methods for flexible disk assembly.
[0118] Any reference to memory, storage, database or other media used in the embodiments provided in the present application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM) and memory bus dynamic RAM (RDRAM).
[0119] In summary, the present invention not only reduces the operator's visual inspection time and reduces labor intensity, but also eliminates the need for the operator to visually judge the results. The sensor automatically measures and outputs the results, thereby improving the reliability of the detection.
[0120] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.
Claims
1. An automatic detection method for flexible disk assembly, characterized in that: The method comprises: M1. Place the flexible disk assembly to be tested on the detection platform, obtain the image data information of the flexible disk assembly in real time based on the on-board camera, and obtain the distance data information of the flexible disk assembly in real time based on the on-board laser displacement sensor; M2. Based on the distance data information of the flexible disk assembly and the image data information of the flexible disk assembly, the distance and image of the flexible disk assembly are fused using a multimodal fusion algorithm based on a SIN chaotic sequence to obtain the fusion data information of the distance and image of the flexible disk assembly; M3. Based on the fusion data information of the distance and image of the flexible disk assembly, a neural network model for abnormal point detection of the flexible disk assembly is constructed, and the abnormal hyperparameters of the flexible disk assembly are inferred to obtain abnormal hyperparameter data information of the flexible disk assembly; M4. Based on the abnormal hyperparameter data information of the flexible disk assembly, establish an abnormal value function F of the flexible disk assembly, calculate the abnormal value of the flexible disk assembly, and output the abnormal value data information of the flexible disk assembly.
2. The automatic detection method for flexible disk assembly according to claim 1, characterized in that: The method further comprises: M5. Based on the abnormal value data information of the flexible disk assembly, a preset threshold is set. If the abnormal value of the flexible disk assembly is greater than the preset threshold, the flexible disk assembly is unqualified and is discarded. If the abnormal value of the flexible disk assembly is less than the preset threshold, the flexible disk assembly is qualified and meets the production standard.
3. The automatic detection method for flexible disk assembly according to claim 1, characterized in that: In step M2, the method of fusing the distance and image of the flexible disk assembly using a multi-modal fusion algorithm based on a SIN chaotic sequence includes: M21. Based on the distance data information of the flexible disk assembly and the image data information of the flexible disk assembly, a SIN chaotic mapping function G of the image and distance of the flexible disk assembly is established. Among them, x is the distance data information of the flexible disk assembly, y is the image data information of the flexible disk assembly, f is the relationship function between the image and distance of the flexible disk assembly, g is the SIN function of the image and distance of the flexible disk assembly, α1, α2 and α3 are the SIN chaotic factors of the image and distance of the flexible disk assembly; M22. Based on the SIN chaotic mapping function G of the image and distance of the flexible disk assembly, the SIN chaotic sequence of the image and distance of the flexible disk assembly is characterized to obtain data information of the SIN chaotic sequence of the image and distance of the flexible disk assembly; M23. Based on the data information of the SIN chaotic sequence of the image and distance of the flexible disk assembly, a multimodal fusion function H of the flexible disk assembly is established. Among them, z is the data information of the SIN chaotic sequence of the image and distance of the flexible disk assembly, β1, β2 and β3 are the multimodal fusion factors of the flexible disk assembly, and the distance and image of the flexible disk assembly are fused to obtain the fused data information of the distance and image of the flexible disk assembly.
4. The automatic detection method for flexible disk assembly according to claim 3 is characterized in that: The relationship function f between the image and distance of the flexible disk assembly is, The SIN function g of the image and distance of the flexible disk assembly is, Wherein, x is the distance data information of the flexible disk assembly, and y is the image data information of the flexible disk assembly.
5. The automatic detection method for flexible disk assembly according to claim 3 is characterized in that: The constraints of the image and distance SIN chaotic factors α1, α2 and α3 of the flexible disk assembly are: The constraint function h of the multimodal fusion factors β1, β2 and β3 of the flexible disk assembly is, Among them, the value range of the constraint function h is (1,2).
6. The automatic detection method for flexible disk assembly according to claim 1, characterized in that: In step M3, the construction of a neural network model for detecting abnormal points of the flexible disk assembly includes: M31. Based on the distance and image fusion data information of the flexible disk assembly, a feature extraction function W of the flexible disk assembly is constructed, Among them, r is the fusion data information of the distance and image of the flexible disk assembly, λ1, λ2 and λ3 are the feature extraction factors of the flexible disk assembly, and the fusion data of the flexible disk assembly is subjected to feature extraction to obtain the fusion feature matrix data information of the distance and image of the flexible disk assembly; M32. Inputting the fusion feature data information of the distance and image of the flexible disk assembly into the neural network model for abnormal point detection of the flexible disk assembly for training and learning, and determining the abnormal hyperparameter function S of the flexible disk neural network model, Among them, a is the fusion feature data information of the distance and image of the flexible disk assembly, η1 and η2 are the prediction factors of the abnormal hyperparameters of the flexible disk, and the trained neural network model for abnormal point detection of the flexible disk assembly is obtained; M33. Based on the trained neural network model for abnormal point detection of the flexible disk assembly, the distance of the flexible disk assembly and the fusion feature data information of the image are input, the abnormal hyperparameters of the flexible disk assembly are inferred, and the abnormal hyperparameter data information of the flexible disk assembly is obtained.
7. The automatic detection method for flexible disk assembly according to claim 6, characterized in that: The feature extraction factor λ1 of the flexible disk assembly is, The feature extraction factor λ2 of the flexible disk assembly is, The feature extraction factor λ3 of the flexible disk assembly is, Among them, r is the distance of the flexible disk assembly and the fusion data information of the image.
8. The automatic detection method for flexible disk assembly according to claim 7, characterized in that: The constraint function v of the feature extraction factors λ1, λ2 and λ3 of the flexible disk assembly is, The value range of v is (2,3).
9. The automatic detection method for flexible disk assembly according to claim 1, characterized in that: The abnormal value function F of the flexible disk assembly is, Among them, c is the abnormal hyperparameter data information of the flexible disk assembly, and ω1, ω2 and ω3 are the abnormal value estimation factors of the flexible disk assembly.
10. The automatic detection method for flexible disk assembly according to claim 9, characterized in that: The constraints of the abnormal value prediction factors ω1, ω2 and ω3 of the flexible disk assembly are:
