Compression detection method for sealing strip with layered structure
Through the layered structure of the seal strip compression detection method, the problem of low awareness of the seal strip compression condition is solved, accurate perception and real-time monitoring of the seal strip compression condition are achieved, and the seal stability and reliability of the sealing object are improved.
Patent Information
- Application Number
- CN202411375585.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-30
- Publication Date
- 2025-02-28
AI Technical Summary
In the prior art, the perception accuracy of the sealing strip is low, resulting in delay in adjusting the compression deviation and affecting the sealing stability of the sealing object.
The seal strip compression detection method with a layered structure is adopted. By obtaining the contact area between multiple sealing layers of the target seal strip, when the predetermined area limit is reached, an electrical signal detection command is issued, detection is performed, signal characteristic information is extracted, and input it into the compression degree prediction model to obtain the compression degree prediction value, and perform compression performance evaluation.
It improves the accuracy and response speed of seal strip compression conditions, reduces the dependence on manual measurement, and ensures real-time online monitoring of sealing performance.
Smart Images
Figure CN119533885A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of sealing strip detection, and in particular to a sealing strip compression detection method with a layered structure. Background Art
[0002] Currently, the perception accuracy of the sealing strip compression condition is not high, which will cause a certain delay in adjusting the sealing strip compression deviation, thereby affecting the sealing stability of the sealed object.
[0003] Specifically, in industrial production, if the tightness of the sealing strips in industrial equipment is not accurately perceived, then during the production and operation of the industrial equipment, the loosening or improper tightness of the sealing strips cannot be detected in time, which may lead to leakage or poor sealing during the production process.
[0004] Such defects not only affect the normal operation of industrial equipment, but may also bring safety hazards and reduced production efficiency. Therefore, improving the accuracy and response speed of the sealing strip compression condition sensing system to ensure the stability and reliability of the sealing object is an urgent problem to be solved. Summary of the invention
[0005] The present application provides a sealing strip compression detection method with a layered structure, which is used to solve the technical problem in the prior art that the sealing strip compression status is perceived with low accuracy, resulting in delayed adjustment of the sealing strip compression deviation and affecting the sealing stability of the sealed object.
[0006] In view of the above problems, the present application provides a sealing strip compression detection method for a layered structure, the method comprising: obtaining a target sealing layer set of a target sealing strip, the target sealing layer set comprising N sealing layers, N≥2; obtaining a target contact area between a first sealing layer and a second sealing layer among the N sealing layers; when the target contact area reaches a predetermined area limit, issuing an electrical signal detection instruction, and detecting the target sealing strip based on the electrical signal detection instruction to obtain a target detection signal timing; reading a predetermined signal feature, and performing feature extraction on the target detection signal timing based on the predetermined signal feature to obtain target signal feature information; using the target signal feature information as input information of a compression prediction model to obtain an output result, wherein the output result comprises a target compression prediction value of the target sealing strip; and performing compression performance evaluation on the target sealing strip based on the target compression prediction value.
[0007] In another aspect of the present application, a sealing strip compression detection system with a layered structure is provided, the system comprising: a sealing information acquisition unit, used to acquire a target sealing layer set of a target sealing strip, the target sealing layer set comprising N sealing layers, N≥2; a contact area acquisition unit, used to acquire a target contact area between a first sealing layer and a second sealing layer in the N sealing layers; a detection instruction execution unit, used to issue an electric signal detection instruction when the target contact area reaches a predetermined area limit, and detect the target sealing strip based on the electric signal detection instruction to obtain a target detection signal timing; a signal feature extraction unit, used to read a predetermined signal feature, and perform feature extraction on the target detection signal timing based on the predetermined signal feature to obtain target signal feature information; an output result acquisition unit, used to use the target signal feature information as input information of a compression prediction model to obtain an output result, wherein the output result includes a target compression prediction value of the target sealing strip; and a compression performance evaluation unit, used to perform compression performance evaluation on the target sealing strip based on the target compression prediction value.
[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:
[0009] The method provided in the embodiment of the present application is to obtain a target sealing layer set of a target sealing strip, wherein the target sealing layer set includes N sealing layers, N≥2; obtain a target contact area between the first sealing layer and the second sealing layer in the N sealing layers; when the target contact area reaches a predetermined area limit, issue an electric signal detection instruction, and detect the target sealing strip based on the electric signal detection instruction to obtain a target detection signal timing; read a predetermined signal feature, and extract the target detection signal timing based on the predetermined signal feature to obtain target signal feature information; use the target signal feature information as input information of a compression prediction model to obtain an output result, wherein the output result includes a target compression prediction value of the target sealing strip; and evaluate the compression performance of the target sealing strip based on the target compression prediction value. The method achieves the technical effect of reducing the dependence of the compression condition detection and analysis of the sealing strip on manual measurement, improving the accuracy of the perception of the compression condition of the sealing strip, and ensuring the real-time online sealing performance of the sealing strip. BRIEF DESCRIPTION OF THE DRAWINGS
[0010] Figure 1 A schematic diagram of a process flow of a sealing strip compression detection method for a layered structure provided in the present application;
[0011] Figure 2 A schematic diagram of a process for generating a target polynomial in a sealing strip compression detection method of a layered structure provided in the present application;
[0012] Figure 3A structural schematic diagram of a sealing strip compression detection system with a layered structure provided in the present application.
[0013] Explanation of the reference numerals: sealing information obtaining unit 11 , contact area obtaining unit 12 , detection instruction execution unit 13 , signal feature extraction unit 14 , output result obtaining unit 15 , pressing performance evaluation unit 16 . DETAILED DESCRIPTION
[0014] The present application provides a sealing strip compression detection method with a layered structure, which is used to solve the technical problem in the prior art that the sealing strip compression status perception accuracy is low, resulting in delayed adjustment of the sealing strip compression deviation, and affecting the sealing stability of the sealed object. The method achieves the technical effect of reducing the dependence of the sealing strip compression status detection and analysis on manual measurement, improving the sealing strip compression status perception accuracy, and ensuring the real-time online sealing performance of the sealing strip.
[0015] Below, the technical solutions in the present invention will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments of the present invention. It should be understood that the present invention is not limited to the example embodiments described herein. 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. It should also be noted that, for the convenience of description, only the parts related to the present invention are shown in the accompanying drawings, rather than all of them.
[0016] Embodiment 1
[0017] like Figure 1 As shown, the present application provides a sealing strip compression detection method of a layered structure, comprising:
[0018] A100: Obtain a target sealing layer set of a target sealing strip, wherein the target sealing layer set includes N sealing layers, where N≥2.
[0019] A200: Obtain a target contact area between the first sealing layer and the second sealing layer in the N sealing layers.
[0020] Specifically, it should be understood that a sealing strip is a material used to prevent liquid, gas or dust from entering or leaking out of a closed space, and is usually used as an isolation layer between two surfaces. A sealing layer refers to a material layer used when constructing a sealing strip, which is a part of the sealing strip. The sealing layer can be a single material layer or a combination of multiple material layers.
[0021] In this embodiment, the target sealing strip is an integral product composed of N sealing layers, and the N sealing layers constitute the target sealing layer set.
[0022] The first sealing layer is uniformly coated with dye, the first sealing layer and the second sealing layer are tightly attached, and after the two sealing layers are separated, the coating area of the dye on the surface of the second sealing layer is measured as the target contact area between the first sealing layer and the second sealing layer, and the first sealing layer and the second sealing layer are two random adjacent sealing layers in the target sealing strip.
[0023] A300: When the target contact area reaches a predetermined area limit, an electrical signal detection instruction is issued, and the target sealing strip is detected based on the electrical signal detection instruction to obtain a target detection signal timing.
[0024] In one embodiment, the method steps provided by the present application further include:
[0025] A310: Perform polynomial fitting on the target scatter plot generated based on the target detection signal timing to obtain a target polynomial.
[0026] A320: Obtain a first prediction signal at a first time according to the target polynomial.
[0027] A330: Determine whether the first prediction signal meets the predetermined signal limit and obtain a first judgment result.
[0028] A340: Based on the first judgment result, if the first prediction signal does not meet the predetermined signal limit, a sealing failure warning instruction is issued.
[0029] A350: Performing a leakage warning notification of the target sealing strip according to the sealing fault warning instruction.
[0030] In one embodiment, Figure 2 As shown, the method steps provided by the present application also include:
[0031] A311: Generate a first fitting curve according to the first scatter point group in the target scatter point graph.
[0032] A312: Obtain the number of scattered points on the first fitting curve that meet the predetermined constraints, and record it as the first scattered point number value.
[0033] A313: Generate a second fitting curve according to the second scatter point group in the target scatter point graph.
[0034] A314: Obtain the number of scattered points on the second fitting curve that meet the predetermined constraint, and record it as a second scattered point number value.
[0035] A315: When the second scatter point quantity value is greater than the first scatter point quantity value, obtain a second regression polynomial of the second fitting curve.
[0036] A316: Record the second regression polynomial as the target polynomial.
[0037] In one embodiment, the method steps provided by the present application further include:
[0038] A321: Obtain a first signal deviation between the first prediction signal and the predetermined signal limit, and perform life prediction for the target sealing strip based on the first signal deviation.
[0039] Specifically, in this embodiment, the predetermined area limit is the minimum contact area constraint for the first sealing layer and the second sealing layer to achieve a sealing effect. When the target contact area reaches (is less than) the predetermined area limit, it indicates that the sealing effect of the first sealing layer and the second sealing layer may be invalid (the sealing performance of the target sealing strip may be invalid). Based on the premise that there is a risk of invalid sealing, this embodiment performs a sealing compression performance evaluation and analysis of the target sealing strip to determine whether the target sealing strip actually has a sealing failure.
[0040] Specifically, when the target contact area reaches a predetermined area limit, an electrical signal detection instruction is issued, and the electrical signal detection instruction includes conductivity detection instructions of multiple detection cycles.
[0041] The target sealing strip is detected based on the conductivity detection instructions of multiple detection cycles to obtain a target detection signal timing sequence, wherein the target detection signal timing sequence includes a plurality of sealing strip conductivity change timing sequences of multiple detection cycles.
[0042] The conductivity change sequence of each sealing strip is that within a detection cycle, an electric field or current is applied to the target sealing strip based on the conductivity detection instruction, and the response of the target sealing strip to the electric field or current changes over time are recorded.
[0043] Using time as the X-axis and conductivity as the Y-axis, a two-dimensional plane is constructed, and multiple detection cycle data are aligned to the origin of the two-dimensional plane, and then multiple cycle scatter plots are generated based on the multiple sealing strip conductivity change time series of multiple detection cycles of the target detection signal time series. After configuring multiple cycle scatter plots with non-repeated multiple color identifiers, the multiple cycle scatter plots are fitted to the same layer to complete the drawing of the target scatter plot.
[0044] The first scatter point group in the target scatter point diagram is a scatter point group formed by scatter points of a first sealing strip conductivity change time sequence in a first detection period among a plurality of sealing strip conductivity change time sequences in a plurality of detection periods of the target detection signal time sequence.
[0045] The specific method of generating the first fitting curve according to the first scattered point group is as follows:
[0046] The scattered points are sorted according to the X coordinates, and an interpolation function (linear, quadratic or cubic) is constructed for each adjacent scattered point to describe the curve between the two scattered points, and the interpolation functions of multiple groups of adjacent scattered points of the first scattered point group are obtained, and then the interpolation functions of the multiple groups of adjacent scattered points are used to fit the entire curve to generate the first fitting curve.
[0047] The predetermined constraint is that the scatter points are located on the first fitting curve and are native scatter points in the target scatter plot. The number of scatter points on the first fitting curve that meet the predetermined constraint is obtained and recorded as a first scatter point number value, which records the number of scatter points in the target scatter plot that are located on the first fitting curve.
[0048] The same method of fitting to generate the first fitting curve is adopted to generate a second fitting curve according to the second scatter point group in the target scatter point graph, and the number of scatter points on the second fitting curve that meet the predetermined constraints is obtained and recorded as the second scatter point number value.
[0049] When the second scatter point quantity value is greater than the first scatter point quantity value, the form of the interpolation function is determined as the form of the second regression polynomial according to the spline interpolation method (linear, quadratic or cubic) of the second fitting curve.
[0050] Then, the scattered point coordinates of the second scattered point group are input into the second regression polynomial to solve the parameter index, so as to obtain the second regression polynomial of the second fitting curve, and the second regression polynomial is recorded as the target polynomial.
[0051] The first time is the time when the target sealing strip is first sent an electric signal to detect the conductivity in the detection cycle. The first time is input as input data into the target polynomial to calculate the first prediction signal of the first time. The first prediction signal is the return signal prediction value after the current target sealing strip receives the electric signal of the fixed value electric signal detection instruction. Compared with the measured information, it excludes the interference of physical environmental factors. The first prediction signal is the value returned by the target sealing strip to characterize its own conductivity, including but not limited to voltage value and resistance value.
[0052] The predetermined signal limit value is consistent with the first prediction signal unit, and the predetermined signal limit value is the minimum conductivity value of the target sealing strip without sealing failure. It is judged whether the first prediction signal meets the predetermined signal limit value to obtain a first judgment result. If the first judgment result is that the first prediction signal does not meet the predetermined signal limit value, a sealing failure warning instruction is issued, and a leakage warning notification of the target sealing strip is performed according to the sealing failure warning instruction.
[0053] This embodiment is based on conductivity testing and solving conductivity relationship equations, and achieves the technical effect of eliminating interference from external physical factors, analyzing the conductivity of the sealing strip, and quickly and accurately determining whether the sealing strip has a sealing fault.
[0054] If the first judgment result is that the first prediction signal meets the predetermined signal limit, the first prediction signal is used as the numerator and the predetermined signal limit is used as the denominator to calculate the deviation percentage, and 1 minus the deviation percentage is used as the first signal deviation.
[0055] The design life of the target sealing strip is obtained interactively, and the first signal deviation is multiplied by the design life as a prediction result of the life prediction of the target sealing strip based on the first signal deviation. A replacement prompt instruction for the target sealing strip is generated according to the prediction result. When the future use time of the target sealing strip reaches the prediction result, the replacement prompt instruction is sent to replace the target sealing strip, thereby achieving the technical effect of ensuring that the sealing strip is replaced in time before losing its sealing performance.
[0056] A400: Read the predetermined signal characteristics, and extract the characteristics of the target detection signal timing based on the predetermined signal characteristics to obtain target signal characteristic information.
[0057] In one embodiment, the predetermined signal features include predetermined time domain signal features, predetermined frequency domain signal features, and predetermined fusion domain signal features. The method step A400 provided in the present application also includes:
[0058] A410: Based on the predetermined time domain signal characteristics, feature extraction is performed on the target detection signal timing to obtain a target time domain feature set.
[0059] A420: Based on the predetermined frequency domain signal characteristics, feature extraction is performed on the target detection signal time series after Fourier transformation to obtain a target frequency domain feature set.
[0060] A430: Based on the predetermined fusion domain signal features, feature extraction is performed on the target detection signal time series after modal decomposition processing to obtain a target fusion domain feature set.
[0061] A440: The target time domain feature set, the target frequency domain feature set and the target fusion domain feature set together constitute the target signal feature information.
[0062] In one embodiment, the method steps provided by the present application further include:
[0063] A431: Obtain a target modal decomposition result, wherein the target modal decomposition result includes multiple modal components.
[0064] A432: Introduce a modal component weight evaluation function to perform weight evaluation on the first modal component among the multiple modal components to obtain a first weight.
[0065] A433: Generate a descending list of the multiple modal components in combination with the first weight, and extract the target modal component in the descending list.
[0066] A434: Using a target modal component feature set obtained by performing feature extraction on the target modal component based on the predetermined fusion domain signal feature as the target fusion domain feature set.
[0067] In one embodiment, the expression of the modal component weight evaluation function is as follows:
[0068]
[0069] Wherein, w(x) represents the weight of the modal component x in the multiple modal components, amount(x) represents the characteristic information amount of the modal component x, amount(X) represents the total characteristic information amount of the target detection signal time series, cor(x) represents the correlation coefficient value between the modal component x and the target detection signal time series, α and β are the first coefficient and the second coefficient respectively, and
[0070] α+β=1.
[0071] Specifically, in this embodiment, predetermined signal features are read, and the predetermined signal features include predetermined time domain signal features, predetermined frequency domain signal features, and predetermined fusion domain signal features.
[0072] It should be understood that time domain signal features refer to features that analyze and describe the characteristics and changes of a signal on the time axis, including the signal's amplitude, waveform, period, pulse width, rise time, fall time, etc. Frequency domain signal features refer to features that analyze and describe the characteristics of a signal in the frequency domain, including the signal's frequency components, spectrum distribution, and frequency characteristics. Fusion domain signal features refer to features that combine the characteristics of the time domain and frequency domain for analysis.
[0073] The target detection signal timing sequence includes multiple sealing strip conductivity change timing sequences of multiple detection cycles. In this embodiment, multiple detection cycles are connected to indirectly connect multiple sealing strip conductivity change timing sequences to obtain a full-cycle conductivity change timing sequence.
[0074] This embodiment extracts features of the full-cycle conductivity change time series based on the predetermined time domain signal features to obtain a target time domain feature set, and extracts features of the full-cycle conductivity change time series converted from the time domain to the frequency domain via Fourier transform based on the predetermined frequency domain signal features to obtain a target frequency domain feature set.
[0075] The specific method of performing modal decomposition analysis on the full-cycle conductivity change time series and obtaining the target fusion domain feature set is as follows:
[0076] A modal component weight evaluation function is pre-constructed, and the expression of the modal component weight evaluation function is as follows:
[0077]
[0078] Among them, w(x) represents the weight of the modal component x among the multiple modal components, amount(x) represents the characteristic information amount of the modal component x, amount(X) represents the total characteristic information amount of the target detection signal time series, cor(x) represents the correlation coefficient value between the modal component x and the target detection signal time series, α and β are the first coefficient and the second coefficient respectively, and α+β=1, the first coefficient and the second coefficient are essentially weights, and this embodiment does not impose numerical limitations on the specific weight values of the first coefficient and the second coefficient.
[0079] Modal decomposition analysis is performed on all conductivity signals in the full-cycle conductivity change time series to obtain a target modal decomposition result, wherein the target modal decomposition result includes modal components of multiple frequencies, each modal component representing an oscillation or change within a specific frequency range in the signal of the full-cycle conductivity change time series.
[0080] The characteristic information of the first modal component among the multiple modal components includes frequency, amplitude and phase, and the total characteristic information of the target detection signal time series is the combination of the characteristic information of the multiple modal components, specifically including the frequency, amplitude and phase of each modal component.
[0081] The method for calculating the correlation coefficient value between the first modal component and the target detection signal time series is as follows: performing mean calculation on the frequencies of multiple modal components of the total characteristic information amount to obtain the frequency mean, performing mean calculation on the amplitudes of multiple modal components of the total characteristic information amount to obtain the amplitude mean, and performing mean calculation on the phases of multiple modal components of the total characteristic information amount to obtain the phase mean.
[0082] The Euclidean distances between the frequency, amplitude and phase of the first modal component and the frequency mean, amplitude mean and phase mean are calculated as the correlation coefficient values between the first modal component and the target detection signal time series.
[0083] The characteristic information amount of the first modal component is unit normalized, and after the total characteristic information amount of the target detection signal time series is unit normalized, the normalization result and the correlation coefficient value between the first modal component and the target detection signal time series, and the weight pre-assigned values of the first coefficient and the second coefficient are substituted into the modal component weight evaluation function for weight evaluation to obtain the first weight.
[0084] The same method is adopted to obtain multiple weights of the multiple modal components, and then the multiple weights are serialized to generate a descending list of the multiple modal components, and the modal component corresponding to the maximum weight value is extracted from the descending list as the target modal component.
[0085] The predetermined fusion domain signal features include the predetermined time domain signal features and the predetermined frequency domain signal features. Based on the predetermined time domain signal features, the conductivity change time series of the target modal component is feature extracted to obtain a modal component time domain feature set. Based on the predetermined frequency domain signal features, the conductivity change time series of the target modal component converted from the time domain to the frequency domain by Fourier transform is feature extracted to obtain a modal component frequency domain feature set. The modal component time domain feature set and the modal component frequency domain feature set constitute a target modal component feature set, and the target modal component feature set is used as the target fusion domain feature set.
[0086] The target time domain feature set, the target frequency domain feature set and the target fusion domain feature set together constitute the target signal feature information.
[0087] This embodiment is based on feature extraction to achieve the technical effect of obtaining target signal feature information for predicting the sealing strip compression degree, and indirectly achieves the technical effect of improving the prediction accuracy of the sealing strip compression degree.
[0088] A500: Using the target signal characteristic information as input information of a compression prediction model to obtain an output result, wherein the output result includes a target compression prediction value of the target sealing strip.
[0089] Specifically, in this embodiment, multiple sample sealing strips of the same model as the target sealing strip and multiple sample compression degrees are obtained interactively, and multiple sample signal characteristic information of the multiple sample sealing strips is obtained by the same method as that of obtaining the target signal characteristic information.
[0090] Based on unspecific data feature analysis models such as back propagation neural networks or recurrent neural networks, a compression prediction model is constructed, wherein the input data of the compression prediction model is the characteristic information of the sealing strip signal, and the output is the predicted value of the compression of the sealing strip.
[0091] A conventional training method of an unspecified data feature analysis model such as a back propagation neural network or a recurrent neural network is used to train a compression prediction model using the multiple sample signal feature information and multiple sample compressions until the output accuracy of the compression prediction model meets expectations (≥98%).
[0092] The target signal characteristic information is used as input information of a compression degree prediction model to obtain an output result, where the output result is a target compression degree prediction value of the target sealing strip.
[0093] This embodiment predicts the compression degree of the target sealing strip based on the pre-built data analysis model, thereby achieving the technical effect of improving the approximation between the compression degree and the actual compression degree condition of the target sealing strip.
[0094] A600: Evaluate the compression performance of the target sealing strip based on the target compression degree prediction value.
[0095] The target compression degree parameter and the compression degree deviation threshold of the target sealing strip are interactively obtained, and the deviation is calculated according to the target compression degree prediction value and the target compression degree parameter to obtain the compression degree deviation coefficient.
[0096] If the compression deviation coefficient falls within the compression deviation threshold, it indicates that the compression condition of the current target sealing strip is normal. Conversely, if the compression deviation coefficient falls within the compression deviation threshold, it indicates that the compression condition of the current target sealing strip is abnormal, and the sealing strip needs to be replaced or the relative position of the sealing strip and the sealing object needs to be adjusted.
[0097] This embodiment achieves the technical effect of reducing the dependence of the sealing strip compression condition detection and analysis on manual measurement, improving the accuracy of the sealing strip compression condition perception, and ensuring the real-time online sealing performance of the sealing strip.
[0098] Embodiment 2
[0099] Based on the same inventive concept as the method for detecting the compression of a sealing strip with a layered structure in the above-mentioned embodiment, Figure 3 As shown, the present application provides a sealing strip compression detection system with a layered structure, wherein the system comprises:
[0100] The sealing information obtaining unit 11 is used to obtain a target sealing layer set of a target sealing strip, where the target sealing layer set includes N sealing layers, where N≥2.
[0101] The contact area obtaining unit 12 is used to obtain a target contact area between the first sealing layer and the second sealing layer in the N sealing layers.
[0102] The detection instruction execution unit 13 is used to issue an electrical signal detection instruction when the target contact area reaches a predetermined area limit, and detect the target sealing strip based on the electrical signal detection instruction to obtain a target detection signal timing.
[0103] The signal feature extraction unit 14 is used to read the predetermined signal feature, and extract the feature of the target detection signal timing based on the predetermined signal feature to obtain target signal feature information.
[0104] The output result obtaining unit 15 is used to use the target signal characteristic information as input information of a compression degree prediction model to obtain an output result, wherein the output result includes a target compression degree prediction value of the target sealing strip.
[0105] The compression performance evaluation unit 16 is used to evaluate the compression performance of the target sealing strip based on the target compression degree prediction value.
[0106] In one embodiment, the signal feature extraction unit 14 further includes:
[0107] The predetermined signal features include predetermined time domain signal features, predetermined frequency domain signal features and predetermined fusion domain signal features, and the method includes: based on the predetermined time domain signal features, performing feature extraction on the target detection signal time series to obtain a target time domain feature set; based on the predetermined frequency domain signal features, performing feature extraction on the target detection signal time series after Fourier transformation to obtain a target frequency domain feature set; based on the predetermined fusion domain signal features, performing feature extraction on the target detection signal time series after modal decomposition processing to obtain a target fusion domain feature set; the target time domain feature set, the target frequency domain feature set and the target fusion domain feature set together constitute the target signal feature information.
[0108] In one embodiment, the signal feature extraction unit 14 further includes:
[0109] Obtain a target modal decomposition result, wherein the target modal decomposition result includes multiple modal components; introduce a modal component weight evaluation function to perform weight evaluation on a first modal component among the multiple modal components to obtain a first weight; generate a descending list of the multiple modal components in combination with the first weight, and extract the target modal component in the descending list; and use a target modal component feature set obtained by performing feature extraction on the target modal component based on the predetermined fusion domain signal feature as the target fusion domain feature set.
[0110] In one embodiment, the signal feature extraction unit 14 further includes:
[0111] The expression of the modal component weight evaluation function is as follows:
[0112]
[0113] Among them, W(x) represents the weight of the modal component x among the multiple modal components, amount(x) represents the characteristic information amount of the modal component x, amount(X) represents the total characteristic information amount of the target detection signal time series, cor(x) represents the correlation coefficient value between the modal component x and the target detection signal time series, α and β are the first coefficient and the second coefficient respectively, and α+β=1.
[0114] In one embodiment, the detection instruction execution unit 13 further includes:
[0115] Performing polynomial fitting on a target scatter plot generated based on the target detection signal timing to obtain a target polynomial; obtaining a first prediction signal at a first time based on the target polynomial; determining whether the first prediction signal meets a predetermined signal limit to obtain a first determination result; based on the first determination result, issuing a sealing fault warning instruction if the first prediction signal does not meet the predetermined signal limit; and performing a leakage warning notification of the target sealing strip based on the sealing fault warning instruction.
[0116] In one embodiment, the detection instruction execution unit 13 further includes:
[0117] A first fitting curve is generated according to a first scatter point group in the target scatter point graph; the number of scatter points on the first fitting curve that meet a predetermined constraint is obtained, recorded as a first scatter point quantity value; a second fitting curve is generated according to a second scatter point group in the target scatter point graph; the number of scatter points on the second fitting curve that meet the predetermined constraint is obtained, recorded as a second scatter point quantity value; when the second scatter point quantity value is greater than the first scatter point quantity value, a second regression polynomial of the second fitting curve is obtained; and the second regression polynomial is recorded as the target polynomial.
[0118] In one embodiment, the detection instruction execution unit 13 further includes:
[0119] A first signal deviation between the first prediction signal and the predetermined signal limit is obtained, and a lifespan prediction is performed on the target sealing strip based on the first signal deviation.
[0120] Any of the methods or steps described above may be stored as computer instructions or programs in various types of computer memories, and the computer instructions or programs may be recognized by various types of computer processors to implement any of the methods or steps described above.
[0121] Based on the above specific embodiments of the present invention, any improvements and modifications made to the present invention by those skilled in the art without departing from the principles of the present invention shall fall within the patent protection scope of the present invention.
Claims
1. A method for detecting compression of a sealing strip of a layered structure, characterized in that: include: Acquire a target sealing layer set of a target sealing strip, wherein the target sealing layer set includes N sealing layers, where N≥2; Obtaining a target contact area between a first sealing layer and a second sealing layer in the N sealing layers; When the target contact area reaches a predetermined area limit, an electrical signal detection instruction is issued, and the target sealing strip is detected based on the electrical signal detection instruction to obtain a target detection signal timing sequence; Reading a predetermined signal feature, and extracting a feature of the target detection signal timing based on the predetermined signal feature to obtain target signal feature information; Using the target signal characteristic information as input information of a compression prediction model to obtain an output result, wherein the output result includes a target compression prediction value of the target sealing strip; The compression performance of the target sealing strip is evaluated based on the target compression degree prediction value.
2. The method according to claim 1, characterized in that The predetermined signal features include predetermined time domain signal features, predetermined frequency domain signal features and predetermined fusion domain signal features, and the method includes: Based on the predetermined time domain signal feature, feature extraction is performed on the target detection signal time sequence to obtain a target time domain feature set; Based on the predetermined frequency domain signal features, feature extraction is performed on the target detection signal time series after Fourier transformation to obtain a target frequency domain feature set; Based on the predetermined fusion domain signal features, feature extraction is performed on the target detection signal time series after modal decomposition processing to obtain a target fusion domain feature set; The target time domain feature set, the target frequency domain feature set and the target fusion domain feature set together constitute the target signal feature information.
3. The method according to claim 2, characterized in that Methods include: Obtaining a target modal decomposition result, wherein the target modal decomposition result includes multiple modal components; Introducing a modal component weight evaluation function to perform weight evaluation on a first modal component among the multiple modal components to obtain a first weight; generating a descending list of the plurality of modal components in combination with the first weights, and extracting a target modal component from the descending list; A target modal component feature set obtained by performing feature extraction on the target modal component based on the predetermined fusion domain signal feature is used as the target fusion domain feature set.
4. The method according to claim 3, characterized in that The expression of the modal component weight evaluation function is as follows: Among them, w(x) represents the weight of the modal component x among the multiple modal components, amount(x) represents the characteristic information amount of the modal component x, amount(X) represents the total characteristic information amount of the target detection signal time series, cor(x) represents the correlation coefficient value between the modal component x and the target detection signal time series, α and β are the first coefficient and the second coefficient respectively, and α+β=1.
5. The method according to claim 1, characterized in that The method also includes: Performing polynomial fitting on a target scatter plot generated based on the target detection signal timing to obtain a target polynomial; Obtaining a first prediction signal at a first time according to the target polynomial; Determining whether the first prediction signal meets a predetermined signal limit, and obtaining a first determination result; According to the first judgment result, if the first prediction signal does not meet the predetermined signal limit, issuing a sealing failure warning instruction; An air leakage warning notification of the target sealing strip is performed according to the sealing failure warning instruction.
6. The method according to claim 5, characterized in that include: Generate a first fitting curve according to a first scatter point group in the target scatter point graph; Obtaining the number of scattered points on the first fitting curve that meet the predetermined constraints, recorded as a first scattered point number value; generating a second fitting curve according to a second scatter point group in the target scatter point graph; Obtaining the number of scattered points on the second fitting curve that meet the predetermined constraint, recorded as a second scattered point number value; When the second scatter point quantity value is greater than the first scatter point quantity value, obtaining a second regression polynomial of the second fitting curve; The second regression polynomial is recorded as the target polynomial.
7. The method according to claim 5, characterized in that Methods include: A first signal deviation between the first prediction signal and the predetermined signal limit is obtained, and a lifespan prediction is performed on the target sealing strip based on the first signal deviation.
8. A multi-source data planning system based on the city brain, characterized in that: The steps for implementing the method according to any one of claims 1 to 7 include: A sealing information obtaining unit, used to obtain a target sealing layer set of a target sealing strip, wherein the target sealing layer set includes N sealing layers, where N≥2; A contact area obtaining unit, used for obtaining a target contact area between a first sealing layer and a second sealing layer in the N sealing layers; a detection instruction execution unit, configured to issue an electrical signal detection instruction when the target contact area reaches a predetermined area limit, and detect the target sealing strip based on the electrical signal detection instruction to obtain a target detection signal timing sequence; A signal feature extraction unit, used to read a predetermined signal feature, and extract a feature of the target detection signal timing based on the predetermined signal feature to obtain target signal feature information; an output result obtaining unit, configured to use the target signal characteristic information as input information of a compression prediction model to obtain an output result, wherein the output result includes a target compression prediction value of the target sealing strip; A compression performance evaluation unit is used to evaluate the compression performance of the target sealing strip based on the target compression degree prediction value.