Ring spinning yarn breakage detection method and system based on signal processing

Abstract

The present application relates to the technical field of broken yarn detection, and particularly relates to a ring spinning broken yarn detection method and system based on signal processing. The method comprises the following steps: collecting video data of a ring spinning operation process to obtain ring spinning operation process video data; extracting audio signals from the ring spinning operation process video data and performing parallel baseline drift removal processing to obtain audio signal baseline optimized data; performing ring spinning broken yarn audio signal condition analysis according to the audio signal baseline optimized data to obtain broken yarn audio signal condition data; performing broken yarn factor analysis according to the broken yarn audio signal condition data and performing spinning behavior control optimization to obtain spinning behavior control optimization data; performing control time period intervention on the broken yarn factor data and performing data encoding to obtain signal condition time period intervention encoding data and send the signal condition time period intervention encoding data to a cloud platform to perform ring spinning broken yarn detection. The present application optimizes the broken yarn detection technology to make the detection technology more accurate.

Description

Technical Field

[0001] This invention relates to the field of yarn breakage detection technology, and in particular to a method and system for detecting yarn breakage in ring spinning based on signal processing. Background Technology

[0002] Due to various factors such as fiber quality, machine wear, and environmental conditions, yarn breakage frequently occurs, affecting production efficiency and product quality. Therefore, utilizing signal processing technology to analyze and process signals during the spinning process to detect yarn breakage and take timely corrective measures is crucial for improving textile production efficiency and reducing production costs. However, traditional signal processing-based methods for detecting yarn breakage in ring spinning suffer from issues such as signal baseline drift affecting the signal and inaccurate yarn breakage signal analysis. Summary of the Invention

[0003] Therefore, it is necessary to provide a signal processing-based method for detecting yarn breakage in ring spinning to solve at least one of the aforementioned technical problems.

[0004] To achieve the above objectives, a ring spinning yarn breakage detection method based on signal processing is provided, the method comprising the following steps:

[0005] Step S1: Collect video data of the ring spinning process using electronic monitoring equipment to obtain video data of the ring spinning process; extract audio signals from the video data of the ring spinning process to obtain audio signals of the ring spinning process; perform parallel baseline drift removal processing on the audio signals of the ring spinning process to obtain baseline optimized data of the audio signals.

[0006] Step S2: Match the spinning stretching audio signal based on the baseline optimization data of the audio signal to obtain the spinning stretching audio signal matching data; perform spinning stretching audio signal change analysis on the matching data to obtain the spinning stretching audio signal change data; perform energy criticality calculation on the spinning stretching audio signal change data to obtain the audio signal energy critical data; perform ring spinning yarn breakage audio signal condition analysis based on the audio signal energy critical data to obtain the yarn breakage audio signal condition data.

[0007] Step S3: Perform sound frequency signal energy mapping processing under different sound frequency signal conditions based on the yarn breakage sound frequency signal condition data to obtain yarn breakage sound frequency signal energy mapping data; perform yarn breakage factor analysis based on the yarn breakage sound frequency signal energy mapping data to obtain yarn breakage factor data; and optimize spinning behavior control based on the yarn breakage factor data to obtain spinning behavior control optimization data.

[0008] Step S4: Based on the spinning behavior control optimization data, control the yarn breakage factor data to obtain signal condition time period control intervention data; encode the signal condition time period control intervention data to obtain signal condition time period intervention coded data; send the signal condition time period intervention coded data to the cloud platform to perform ring spinning yarn breakage detection.

[0009] This invention collects video data of the ring spinning process using electronic monitoring equipment and extracts audio signals, facilitating real-time monitoring of the spinning machine's operating status. This monitoring helps operators promptly identify anomalies and take corrective measures, thereby improving production efficiency and product quality. Parallel baseline drift removal effectively optimizes the audio signal baseline, reducing noise and interference and improving signal clarity and accuracy. This aids in subsequent data analysis and processing. By matching and analyzing the audio signals, the changing patterns of the audio signals during spinning stretching can be understood, providing a deeper understanding of the spinning machine's operating status and performance. This allows for timely adjustment of the spinning machine's parameters to achieve better production results. By calculating the energy threshold of the audio signals and analyzing yarn breakage conditions, potential yarn breakage during spinning can be predicted, allowing for corresponding preventative measures, such as adjusting spinning machine parameters or replacing consumables, to avoid production interruptions and losses. Analyzing energy mapping data under different yarn breakage audio signal conditions provides a deeper understanding of various factors affecting yarn breakage, such as fiber quality, spinning machine parameters, and environmental conditions. This helps identify and understand the root causes of yarn breakage, providing a basis for taking appropriate preventative measures. Based on yarn breakage factor data, by optimizing spinning behavior control, the operating mode, parameter settings, or operation methods of the spinning machine can be adjusted to reduce yarn breakage. This optimization can improve the stability and controllability of the spinning process, thereby increasing production efficiency and product quality. Based on the optimized spinning behavior control data, control intervention strategies for signal condition periods can be determined, that is, controlling the spinning machine during specific time periods to prevent yarn breakage events. Encoding the control intervention data and sending it to a cloud platform enables remote monitoring and control, thereby allowing for timely adjustments to the spinning machine's operating status and avoiding potential production losses.

[0010] Preferably, step S1 includes the following steps:

[0011] Step S11: Collect video data of the ring spinning process using electronic monitoring equipment to obtain video data of the ring spinning process;

[0012] Step S12: Extract audio signals from the video data of the ring spinning process to obtain the ring spinning operation audio signal;

[0013] Step S13: Perform signal filtering on the spinning operation audio signal to obtain the spinning operation audio signal filtering data;

[0014] Step S14: Perform parallel baseline drift removal processing on the ring spinning operation audio signal based on the filtered data of the spinning operation audio signal to obtain the optimized baseline data of the audio signal.

[0015] This invention uses electronic monitoring equipment to collect video data of the ring spinning process, capturing the real-time operation and status of the spinning machine. This helps to promptly detect abnormalities, such as machine malfunctions or fiber entanglement, allowing for timely measures to prevent production interruptions or quality issues. Audio signal extraction from the video data reveals sound signals related to the spinning machine's operation. These sound signals may contain information about the machine's operating status, load, and potential problems, facilitating real-time monitoring and analysis. Filtering the extracted audio signals removes noise and interference, improving signal quality and clarity. This aids in subsequent data analysis and processing, ensuring accurate information is obtained from the sound signals. Parallel baseline drift removal processing of the filtered audio signals further optimizes signal quality and reduces the impact of baseline variations on data analysis. This helps ensure that the data extracted from the audio signals is more stable and reliable.

[0016] Preferably, step S14 includes the following steps:

[0017] Step S141: Perform time-series annotation on the ring spinning operation audio signal to obtain audio signal time-series annotation data; perform signal segmentation on the audio signal time-series annotation data based on the filtered data of the spinning operation audio signal to obtain audio signal segmentation annotation data;

[0018] Step S142: Perform time-domain feature transformation on the segmented labeled data of the audio signal to obtain time-domain feature data of the audio signal; perform sparse structure analysis of the signal frequency of different signal segments on the time-domain feature data of the audio signal to obtain sparse structure data of the signal frequency.

[0019] Step S143: Based on the sparse structure data of the signal frequency, perform baseline estimation for different signal segments on the time-domain feature data of the audio signal to obtain baseline estimation data of the audio signal; based on the baseline estimation data of the audio signal, perform signal baseline matrix transformation on the time-domain feature data of the audio signal to obtain the signal baseline estimation matrix;

[0020] Step S144: Perform low-rank approximation on the signal baseline estimation matrix to obtain low-rank approximation data of the signal baseline; perform parallel interpolation on the time-domain feature data of the audio signal based on the low-rank approximation data of the signal baseline to obtain low-rank parallel interpolation data of the signal baseline.

[0021] Step S145: Perform parallel baseline drift removal processing on the audio signal segmentation labeling data based on the low-rank approximation data of the signal baseline and the parallel interpolation data of the low-rank signal baseline to obtain the optimized data of the audio signal baseline.

[0022] This invention uses time-series annotation and signal segmentation to divide and annotate the audio signals from ring spinning operations, enabling better subsequent analysis. This helps distinguish the operating status and problems of the spinning machine at different stages. Time-domain feature transformation and frequency sparsity analysis of the audio signal segments allow for the acquisition of signal characteristic information from different perspectives, such as frequency distribution and waveform characteristics. This contributes to a more comprehensive understanding of the characteristics and variation patterns of the audio signals. By performing baseline estimation and low-rank approximation on the time-domain characteristic data of the audio signals, the influence of baseline drift in the signals can be effectively reduced, thereby improving the accuracy and stability of the signals. This helps reduce errors and interference during data processing. Parallel interpolation based on the low-rank approximation data can further optimize the signal quality, and parallel baseline drift removal processing is performed on the audio signal segments using the low-rank approximation data. This helps ensure the stability and accuracy of the signals, providing a reliable foundation for subsequent data analysis and applications.

[0023] Preferably, step S2 includes the following steps:

[0024] Step S21: Extract the spinning stretching operation process data from the video data of the ring spinning operation to obtain the spinning stretching operation process data;

[0025] Step S22: Based on the baseline optimization data of the audio signal, perform spinning stretching audio signal matching on the yarn stretching operation data to obtain spinning stretching audio signal matching data;

[0026] Step S23: Perform spinning stretching audio frequency change analysis on the matching data of the spinning stretching audio frequency signal to obtain the spinning stretching audio frequency signal change data;

[0027] Step S24: Perform energy criticality calculation on the change data of the spinning stretching audio signal to obtain the energy critical data of the audio signal;

[0028] Step S25: Based on the spinning stretching audio signal change data and the audio signal energy critical data, perform ring spinning yarn breakage audio signal condition analysis to obtain yarn breakage audio signal condition data.

[0029] This invention extracts spinning and stretching operation data from video data to obtain detailed information about the spinning machine's operating status and process, including the degree of stretching and speed. This helps monitor the spinning machine's operation. By optimizing the audio signal baseline and matching the spinning and stretching operation data with the audio signal, the sound-related parts of the operation can be identified, facilitating further analysis of the correlation between sound and operation. Change analysis of the matching audio signal data during spinning and stretching reveals how the sound changes during the process, potentially reflecting the sound characteristics at different stretching stages, which helps identify anomalies. Energy threshold calculations are performed on the changes in the spinning and stretching audio signal data to determine the sound energy threshold, helping to judge the sound intensity and changes, and further analyze the relationship between sound and yarn breakage. Based on the changes in the spinning and stretching audio signal data and the audio signal energy threshold data, conditional analysis of the yarn breakage audio signal can identify the sound characteristics at the time of yarn breakage, aiding in real-time monitoring and prevention of yarn breakage events.

[0030] Preferably, step S24 includes the following steps:

[0031] Step S241: Perform frequency domain transformation on the spinning stretching audio signal change data to obtain stretching audio signal frequency domain change data; perform frequency discrete structure analysis on the stretching audio signal frequency domain change data to obtain frequency discrete structure data.

[0032] Step S242: Divide the frequency domain variation data of the stretched audio signal into signal frequency structure regions based on the discrete structure data of the signal frequency to obtain signal frequency structure region data; extract the frequency extreme value intervals of different regions from the signal frequency structure region data to obtain the regional frequency extreme value intervals.

[0033] Step S243: Calculate the wavelet coefficients of different regions of the signal frequency structure region data according to the regional frequency extreme value intervals to obtain the regional frequency wavelet coefficients; perform multi-scale frequency energy analysis of different regions of the signal frequency structure region data according to the regional frequency wavelet coefficients to obtain the regional multi-scale frequency energy data.

[0034] Step S244: Extract frequency energy boundary intervals for different regions from the multi-scale frequency energy data of the region to obtain frequency energy boundary intervals; evaluate the frequency energy amplification rate for different regions from the frequency energy boundary intervals to obtain frequency energy amplification rate evaluation data.

[0035] Step S245: Calculate the variance of the frequency energy amplification rate assessment data to obtain the rate amplification variance data; perform frequency energy normal distribution structure analysis on the regional multi-scale frequency energy data based on the rate amplification variance data and the frequency energy boundary interval to obtain frequency energy normal distribution data; perform multi-stage sampling of the normal distribution structure on the regional multi-scale frequency energy data based on the frequency energy normal distribution data to obtain frequency energy multi-stage sampling data.

[0036] Step S246: Apply energy structure complexity constraints to the multi-stage sampling data of frequency energy to obtain energy structure complexity constraint data;

[0037] Step S247: Calculate the fractal dimension of frequency energy in different regions based on the multi-stage sampling data of frequency energy and the energy space complexity constraint data, and obtain the frequency energy fractal dimension data.

[0038] Step S248: Perform energy criticality calculation on the regional multi-scale frequency energy data based on the frequency energy fractal dimension data to obtain the audio signal energy criticality data.

[0039] This invention decomposes audio signals into frequency structure regions through frequency domain transformation and discrete structure analysis, helping to identify different frequency components in the sound and providing a foundation for subsequent analysis. Wavelet coefficient calculation and multi-scale frequency energy analysis within different frequency structure regions provide a deeper understanding of the energy distribution of sound across different frequency ranges, aiding in the detection of frequency energy changes and anomalies. By extracting frequency energy boundary intervals and evaluating their amplification rate and variance, the changing trends and fluctuations of sound energy can be determined, helping to identify abnormal sounds and predict yarn breakage events. Structural analysis and multi-stage sampling of the normal distribution of frequency energy provide a more comprehensive understanding of the sound energy distribution and identify possible anomalies or trend changes. By constraining the complexity of the energy structure and calculating the fractal dimension of frequency energy, the complexity and regularity of the sound signal can be further analyzed, thereby better identifying anomalies. Finally, based on the frequency energy fractal dimension data, the energy threshold of the audio signal is calculated to determine the energy difference between normal and abnormal sounds, thus achieving effective identification and prediction of yarn breakage sounds.

[0040] Preferably, step S25 includes the following steps:

[0041] Step S251: Perform periodicity analysis on the spinning stretching audio signal change data to obtain periodicity data; perform signal abrupt change point analysis on the spinning stretching audio signal change data based on the periodicity data to obtain stretching signal abrupt change point data.

[0042] Step S252: Evaluate the instantaneous intensity of the signal abrupt change at the tensile signal abrupt change point data to obtain the instantaneous intensity data at the signal abrupt change point; calculate the difference in instantaneous intensity abrupt change rate at the instantaneous intensity data at the signal abrupt change point to obtain the difference in instantaneous intensity abrupt change rate data.

[0043] Step S253: Based on the instantaneous intensity increase rate difference data and the critical data of audio signal energy, perform audio signal condition analysis on the yarn breakage in ring spinning to obtain the audio signal condition data for yarn breakage.

[0044] This invention identifies periodic changes in spinning stretching audio signal data, such as periodic oscillations or fluctuations, through periodic analysis. Abrupt point analysis detects sudden changes or events in the signal, which may be caused by machine malfunctions or production process anomalies. These analyses help to promptly identify abnormal changes in the signal and facilitate subsequent analysis and processing. Instantaneous intensity assessment of abrupt change points in the stretching signal quantifies their intensity or importance, thus identifying which abrupt changes pose a higher risk or impact. Calculating the difference in instantaneous intensity amplification rates helps assess the rate of change in abrupt change point intensity, which is crucial for determining the severity and urgency of the abrupt change. Combining the instantaneous intensity amplification rate difference data with audio signal energy critical data allows for conditional analysis of ring spinning yarn breakage audio signals. This means that by comparing signal characteristics and critical values, the probability and severity of yarn breakage events can be determined. This facilitates timely measures to prevent or reduce yarn breakage events, improving production efficiency and product quality.

[0045] Preferably, step S3 includes the following steps:

[0046] Step S31: Based on the yarn breakage audio signal condition data, perform audio signal energy mapping processing on the audio signal energy critical data under different yarn breakage audio signal conditions to obtain yarn breakage audio signal energy mapping data;

[0047] Step S32: Analyze the factors of yarn breakage based on the energy mapping data of the yarn breakage audio signal to obtain yarn breakage factor data, which includes yarn breakage tension factor data and yarn breakage speed factor data;

[0048] Step S33: Optimize spinning behavior control based on the energy mapping data of the yarn breakage audio signal and the yarn breakage factor data to obtain optimized spinning behavior control data.

[0049] This invention processes the critical energy data of the audio signal based on the conditions of the yarn breakage audio signal to obtain yarn breakage audio signal energy mapping data. The main purpose of this step is to map the energy data of the audio signal to different yarn breakage conditions, so as to further analyze and understand the characteristics and changes of the audio signal under different conditions. Through this mapping process, the characteristics and patterns of yarn breakage events can be more clearly identified. Based on the yarn breakage audio signal energy mapping data, yarn breakage factor analysis is performed to obtain yarn breakage factor data, including yarn breakage tension factor data and yarn breakage speed factor data. The purpose of this step is to identify the key factors affecting the occurrence of yarn breakage events, such as spinning tension and spinning speed. By analyzing these factors, the root causes of yarn breakage events can be understood in depth, and corresponding improvement measures can be proposed. Combining the yarn breakage audio signal energy mapping data and the yarn breakage factor data, spinning behavior control optimization is performed. The purpose of this step is to adjust the parameters and control strategies in the spinning process according to the analysis results to minimize the occurrence of yarn breakage events and optimize the stability and efficiency of spinning behavior. Through real-time monitoring and adjustment, changes and anomalies in the production process can be responded to in a timely manner, thereby improving the stability and output of the production line.

[0050] Preferably, step S33 includes the following steps:

[0051] Step S331: Based on the yarn breakage audio signal energy mapping data, perform yarn breakage tension factor data and yarn breakage speed factor data to obtain yarn breakage tension signal energy matching data and yarn breakage speed signal energy matching data;

[0052] Step S332: Perform correlation and coupling analysis on the energy matching data of the yarn breakage speed signal and the energy matching data of the yarn breakage tension signal to obtain the correlation and coupling data of the yarn breakage speed;

[0053] Step S333: Perform deviation adjustment factor analysis on the yarn breakage speed-related tension coupling data to obtain speed-related tension potential factor data;

[0054] Step S334: Use the decision tree model to construct a potential factor association model for the potential factor data of speed-related tension, and obtain the speed-tension factor association model; learn the potential factor linear relationship model based on the yarn breakage tension signal energy matching data and the yarn breakage speed signal energy matching data to obtain the potential factor linear relationship model.

[0055] Step S335: Dynamically adjust the spinning speed according to the linear relationship model of potential factors to obtain dynamic adjustment data of spinning speed; perform tension intensity matching on the dynamic adjustment data of spinning speed to obtain tension intensity matching data;

[0056] Step S336: Optimize spinning behavior control based on dynamic adjustment data of spinning speed and tension intensity matching data to obtain optimized spinning behavior control data.

[0057] In this step, the invention performs energy matching on yarn breakage tension factor data and yarn breakage speed factor data based on the yarn breakage audio signal energy mapping data, obtaining yarn breakage tension signal energy matching data and yarn breakage speed signal energy matching data. This process correlates energy data with yarn breakage events, providing crucial data support for subsequent analysis. By using the yarn breakage speed signal energy matching data to perform correlation coupling analysis on the yarn breakage tension signal energy matching data, yarn breakage speed-related tension coupling data is obtained. This helps identify the correlation between speed and tension in yarn breakage events, further revealing the occurrence mechanism of yarn breakage events. In this step, deviation adjustment factor analysis is performed on the yarn breakage speed-related tension coupling data to obtain speed-related tension potential factor data. This analysis helps identify potential influencing factors between speed and tension, thereby better understanding the root cause of yarn breakage events. A decision tree model is used to construct a potential factor correlation model for the speed-related tension potential factor data, obtaining a speed-tension factor correlation model. This model helps understand the correlation between different factors and provides a foundation for subsequent analysis. Based on the potential factor linear relationship model, spinning speed is dynamically adjusted to obtain spinning speed dynamic adjustment data. This allows for timely adjustments to the spinning speed based on changes in potential factors during the spinning process, minimizing yarn breakage. Tension matching is then performed on the dynamic adjustment data of the spinning speed to obtain tension matching data. This step helps ensure that an appropriate level of spinning tension is maintained while adjusting the spinning speed, further reducing the risk of yarn breakage. Finally, spinning behavior control optimization is performed based on the dynamic adjustment data of the spinning speed and the tension matching data to obtain optimized spinning behavior control data. This step optimizes the parameters and control strategies in the spinning process by comprehensively considering factors such as speed and tension, thereby improving production efficiency, reducing costs, and minimizing yarn breakage.

[0058] Preferably, the deviation adjustment factor analysis of the yarn breakage speed-related tension coupling data includes the following steps:

[0059] Data standardization processing is performed on the yarn breakage speed-related tension coupling data to obtain the associated standard coupling data;

[0060] Based on the associated standard coupling data, the speed-tension coupling data of yarn breakage speed is processed by speed-tension correlation normal distribution to obtain speed-tension correlation normal distribution data;

[0061] The dispersion of the velocity-tension correlated normal distribution data is calculated to obtain the dispersion data.

[0062] Based on the data on the degree of dispersion of the velocity-tension correlation normal distribution, the normal discrete distribution AIC deviation is adjusted to obtain the normal distribution AIC deviation adjusted data;

[0063] Based on the normal distribution AIC deviation adjustment data, deviation adjustment factor analysis was performed on the yarn breakage speed-related tension coupling data to obtain speed-related tension potential factor data.

[0064] This invention standardizes the speed-tension coupling data related to yarn breakage. Standardization ensures the data has a uniform scale, eliminating the influence of different dimensions and making the data easier to compare and analyze. This helps ensure that subsequent analysis accurately reflects the relationships between the variables. The speed-tension coupling data is then processed according to a normal distribution. Converting the data to a normal distribution makes it more consistent with the assumptions of statistical analysis, leading to more accurate subsequent analysis. The dispersion of the speed-tension coupling normal distribution data is calculated. This step aims to assess the dispersion of the data, i.e., the degree of data variation. This helps determine the stability and consistency of the data, providing a basis for further analysis. Based on the dispersion data, the speed-tension coupling normal distribution data undergoes AIC (Akaike Information Criterion) bias adjustment. AIC is a commonly used model selection criterion that penalizes the model to balance goodness of fit and complexity. This step further optimizes the data distribution and improves the model's accuracy. Finally, based on the AIC bias-adjusted data, bias adjustment factor analysis is performed on the speed-tension coupling data to obtain potential factors related to speed-tension. This step can identify and assess potential factors affecting the relationship between yarn breakage speed and tension, thereby providing a better understanding of the mechanisms and influencing factors of yarn breakage events.

[0065] Preferably, the present invention also provides a signal processing-based ring spinning yarn breakage detection system for performing the signal processing-based ring spinning yarn breakage detection method described above. The signal processing-based ring spinning yarn breakage detection system includes:

[0066] The signal baseline optimization module is used to acquire video data of the ring spinning process through electronic monitoring equipment to obtain video data of the ring spinning process; extract audio signals from the video data of the ring spinning process to obtain audio signals of the ring spinning process; and perform parallel baseline drift removal processing on the audio signals of the ring spinning process to obtain baseline optimized data of the audio signals.

[0067] The yarn breakage audio signal condition analysis module is used to perform yarn stretching audio signal matching based on the audio signal baseline optimization data to obtain yarn stretching audio signal matching data; to perform yarn stretching audio signal change analysis on the yarn stretching audio signal matching data to obtain yarn stretching audio signal change data; to perform energy criticality calculation on the yarn stretching audio signal change data to obtain audio signal energy critical data; and to perform ring spinning yarn breakage audio signal condition analysis based on the audio signal energy critical data to obtain yarn breakage audio signal condition data.

[0068] The spinning behavior control optimization module is used to perform sound frequency signal energy mapping processing under different yarn breakage sound frequency signal conditions based on the yarn breakage sound frequency signal condition data, to obtain yarn breakage sound frequency signal energy mapping data; to perform yarn breakage factor analysis based on the yarn breakage sound frequency signal energy mapping data, to obtain yarn breakage factor data; and to optimize spinning behavior control based on the yarn breakage factor data, to obtain spinning behavior control optimization data.

[0069] The signal condition time period intervention coding module is used to intervene in the control time period of yarn breakage factor data based on the spinning behavior control optimization data to obtain signal condition time period control intervention data; the signal condition time period control intervention data is encoded to obtain signal condition time period intervention coded data, and the signal condition time period intervention coded data is sent to the cloud platform to perform ring spinning yarn breakage detection.

[0070] The beneficial effects of this invention are that it provides a signal processing-based method for detecting yarn breakage in ring spinning, which is an optimization of a traditional signal processing-based method for detecting yarn breakage in ring spinning. This method solves the problems of signal baseline drift affecting the signal and inaccurate yarn breakage signal analysis in the traditional method, reduces the impact of signal baseline drift on the signal, and improves the accuracy of yarn breakage signal analysis. Attached Figure Description

[0071] Figure 1 This is a schematic diagram of the steps in a signal processing-based method for detecting yarn breakage in ring spinning.

[0072] Figure 2 for Figure 1 A detailed flowchart illustrating the implementation steps of step S2.

[0073] Figure 3 for Figure 1 A detailed flowchart illustrating the implementation steps of step S3.

[0074] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0075] The technical method of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of this invention.

[0076] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0077] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0078] To achieve the above objectives, please refer to Figures 1 to 3 A method for detecting yarn breakage in ring spinning based on signal processing, the method comprising the following steps:

[0079] Step S1: Collect video data of the ring spinning process using electronic monitoring equipment to obtain video data of the ring spinning process; extract audio signals from the video data of the ring spinning process to obtain audio signals of the ring spinning process; perform parallel baseline drift removal processing on the audio signals of the ring spinning process to obtain baseline optimized data of the audio signals.

[0080] Step S2: Match the spinning stretching audio signal based on the baseline optimization data of the audio signal to obtain the spinning stretching audio signal matching data; perform spinning stretching audio signal change analysis on the matching data to obtain the spinning stretching audio signal change data; perform energy criticality calculation on the spinning stretching audio signal change data to obtain the audio signal energy critical data; perform ring spinning yarn breakage audio signal condition analysis based on the audio signal energy critical data to obtain the yarn breakage audio signal condition data.

[0081] Step S3: Perform sound frequency signal energy mapping processing under different sound frequency signal conditions based on the yarn breakage sound frequency signal condition data to obtain yarn breakage sound frequency signal energy mapping data; perform yarn breakage factor analysis based on the yarn breakage sound frequency signal energy mapping data to obtain yarn breakage factor data; and optimize spinning behavior control based on the yarn breakage factor data to obtain spinning behavior control optimization data.

[0082] Step S4: Based on the spinning behavior control optimization data, the control period intervention of the yarn breakage factor data is applied to obtain the signal condition period control intervention data; the signal condition period control intervention data is encoded to obtain the signal condition period intervention encoded data, and the signal condition period intervention encoded data is sent to the cloud platform to perform ring spinning yarn breakage detection.

[0083] In this embodiment of the invention, reference is made to Figure 1 The above is a schematic flowchart of the ring spinning yarn breakage detection method based on signal processing according to the present invention. In this example, the ring spinning yarn breakage detection method based on signal processing includes the following steps:

[0084] Step S1: Collect video data of the ring spinning process using electronic monitoring equipment to obtain video data of the ring spinning process; extract audio signals from the video data of the ring spinning process to obtain audio signals of the ring spinning process; perform parallel baseline drift removal processing on the audio signals of the ring spinning process to obtain baseline optimized data of the audio signals.

[0085] In this embodiment of the invention, electronic monitoring equipment is used to collect video data of the ring spinning process. This involves installing cameras or other video acquisition devices to capture real-time images of the ring spinning operation. The collected video data is then processed to extract audio signals from the ring spinning process. This can be achieved using audio processing software or algorithms to extract sound signals from the video. Parallel baseline drift removal processing is then performed on the extracted audio signals. The purpose of this step is to remove baseline drift from the signal to accurately capture the fluctuations in the audio signals and ensure the accuracy of subsequent analysis.

[0086] Step S2: Match the spinning stretching audio signal based on the baseline optimization data of the audio signal to obtain the spinning stretching audio signal matching data; perform spinning stretching audio signal change analysis on the matching data to obtain the spinning stretching audio signal change data; perform energy criticality calculation on the spinning stretching audio signal change data to obtain the audio signal energy critical data; perform ring spinning yarn breakage audio signal condition analysis based on the audio signal energy critical data to obtain the yarn breakage audio signal condition data.

[0087] In this embodiment of the invention, spinning stretching audio signal matching is performed based on baseline-optimized audio signal data. This involves comparing the audio signal with a pre-determined pattern or reference signal to identify specific audio signal patterns during the spinning stretching process, and analyzing the matched spinning stretching audio signal data to understand the changes in the audio signal. This involves methods such as spectrum analysis and waveform analysis to obtain the characteristics and trends of the audio signal, and performing energy criticality calculations on the spinning stretching audio signal change data. This step aims to determine the energy critical value of the audio signal, i.e., at what energy level the audio signal exhibits the possibility of yarn breakage, and performing ring spinning yarn breakage audio signal condition analysis based on the audio signal energy criticality data. This involves associating the audio signal with yarn breakage events, identifying the correlation between audio signal characteristics and yarn breakage events, and extracting key audio signal condition data for subsequent analysis or application.

[0088] Step S3: Perform sound frequency signal energy mapping processing under different sound frequency signal conditions based on the yarn breakage sound frequency signal condition data to obtain yarn breakage sound frequency signal energy mapping data; perform yarn breakage factor analysis based on the yarn breakage sound frequency signal energy mapping data to obtain yarn breakage factor data; and optimize spinning behavior control based on the yarn breakage factor data to obtain spinning behavior control optimization data.

[0089] In this embodiment of the invention, based on the yarn breakage audio signal condition data, audio signal energy mapping processing is performed for different yarn breakage audio signal conditions. This may involve associating the audio signal data with specific yarn breakage conditions and mapping it into an energy space for subsequent analysis and processing. Based on the obtained yarn breakage audio signal energy mapping data, yarn breakage factor analysis is performed. This includes analyzing and comparing the audio signal energy distribution under different yarn breakage conditions to determine the main factors causing yarn breakage. Based on the yarn breakage factor data, spinning behavior control optimization is performed. This may include adjusting spinning equipment parameters, changing the production process, or taking other measures to minimize or eliminate yarn breakage events and improve production efficiency and product quality.

[0090] Step S4: Based on the spinning behavior control optimization data, the control period intervention of the yarn breakage factor data is applied to obtain the signal condition period control intervention data; the signal condition period control intervention data is encoded to obtain the signal condition period intervention encoded data, and the signal condition period intervention encoded data is sent to the cloud platform to perform ring spinning yarn breakage detection.

[0091] In this embodiment of the invention, control intervention is implemented on yarn breakage factor data based on spinning behavior control optimization data. This means that, according to a pre-set control strategy, intervention is performed on the spinning process during specific time periods to minimize the occurrence of yarn breakage events. The data generated by the control intervention is encoded. This may involve encoding information such as the timing, measures, and effects of the control intervention into digital form for subsequent analysis and recording, and sending the encoded signal condition intervention data to a cloud platform. This can be achieved by sending the data to a cloud server via a network connection for real-time monitoring, analysis, and feedback, and by performing ring spinning yarn breakage detection through the cloud platform. The cloud platform may perform yarn breakage detection and analysis based on the received data and take further measures as necessary, such as sending alarms or adjusting production parameters.

[0092] This invention collects video data of the ring spinning process using electronic monitoring equipment and extracts audio signals, facilitating real-time monitoring of the spinning machine's operating status. This monitoring helps operators promptly identify anomalies and take corrective measures, thereby improving production efficiency and product quality. Parallel baseline drift removal effectively optimizes the audio signal baseline, reducing noise and interference and improving signal clarity and accuracy. This aids in subsequent data analysis and processing. By matching and analyzing the audio signals, the changing patterns of the audio signals during spinning stretching can be understood, providing a deeper understanding of the spinning machine's operating status and performance. This allows for timely adjustment of the spinning machine's parameters to achieve better production results. By calculating the energy threshold of the audio signals and analyzing yarn breakage conditions, potential yarn breakage during spinning can be predicted, allowing for corresponding preventative measures, such as adjusting spinning machine parameters or replacing consumables, to avoid production interruptions and losses. Analyzing energy mapping data under different yarn breakage audio signal conditions provides a deeper understanding of various factors affecting yarn breakage, such as fiber quality, spinning machine parameters, and environmental conditions. This helps identify and understand the root causes of yarn breakage, providing a basis for taking appropriate preventative measures. Based on yarn breakage factor data, by optimizing spinning behavior control, the operating mode, parameter settings, or operation methods of the spinning machine can be adjusted to reduce yarn breakage. This optimization can improve the stability and controllability of the spinning process, thereby increasing production efficiency and product quality. Based on the optimized spinning behavior control data, control intervention strategies for signal condition periods can be determined, that is, controlling the spinning machine during specific time periods to prevent yarn breakage events. Encoding the control intervention data and sending it to a cloud platform enables remote monitoring and control, thereby allowing for timely adjustments to the spinning machine's operating status and avoiding potential production losses.

[0093] Preferably, step S1 includes the following steps:

[0094] Step S11: Collect video data of the ring spinning process using electronic monitoring equipment to obtain video data of the ring spinning process;

[0095] Step S12: Extract audio signals from the video data of the ring spinning process to obtain the ring spinning operation audio signal;

[0096] Step S13: Perform signal filtering on the spinning operation audio signal to obtain the spinning operation audio signal filtering data;

[0097] Step S14: Perform parallel baseline drift removal processing on the ring spinning operation audio signal based on the filtered data of the spinning operation audio signal to obtain the optimized baseline data of the audio signal.

[0098] In this embodiment of the invention, firstly, video data of the ring spinning process is acquired using electronic monitoring equipment. This includes installing appropriate cameras or other video acquisition devices to capture the ring spinning operation. Subsequently, the video data is transmitted to a computer or data storage device for further processing. Next, the audio signals extracted from the video data are processed. This requires using audio processing software or programming tools to extract and analyze the sound from the video data to obtain the audio signals of the ring spinning process. The audio signals extracted from the video data are then subjected to signal filtering. Signal filtering is a common signal processing technique used to remove noise, adjust frequency or amplitude, etc. This can be achieved through digital signal processing algorithms or filters. Finally, based on the filtered audio signal data of the ring spinning operation, parallel baseline drift removal processing is performed on the ring spinning operation audio signal. This may involve using specific algorithms or techniques to remove baseline drift in the ring spinning operation audio signal, thereby optimizing the sound data to obtain more accurate results.

[0099] This invention uses electronic monitoring equipment to collect video data of the ring spinning process, capturing the real-time operation and status of the spinning machine. This helps to promptly detect abnormalities, such as machine malfunctions or fiber entanglement, allowing for timely measures to prevent production interruptions or quality issues. Audio signal extraction from the video data reveals sound signals related to the spinning machine's operation. These sound signals may contain information about the machine's operating status, load, and potential problems, facilitating real-time monitoring and analysis. Filtering the extracted audio signals removes noise and interference, improving signal quality and clarity. This aids in subsequent data analysis and processing, ensuring accurate information is obtained from the sound signals. Parallel baseline drift removal processing of the filtered audio signals further optimizes signal quality and reduces the impact of baseline variations on data analysis. This helps ensure that the data extracted from the audio signals is more stable and reliable.

[0100] Preferably, step S14 includes the following steps:

[0101] Step S141: Perform time-series annotation on the ring spinning operation audio signal to obtain audio signal time-series annotation data; perform signal segmentation on the audio signal time-series annotation data based on the filtered data of the spinning operation audio signal to obtain audio signal segmentation annotation data;

[0102] Step S142: Perform time-domain feature transformation on the segmented labeled data of the audio signal to obtain time-domain feature data of the audio signal; perform sparse structure analysis of the signal frequency of different signal segments on the time-domain feature data of the audio signal to obtain sparse structure data of the signal frequency.

[0103] Step S143: Based on the sparse structure data of the signal frequency, perform baseline estimation for different signal segments on the time-domain feature data of the audio signal to obtain baseline estimation data of the audio signal; based on the baseline estimation data of the audio signal, perform signal baseline matrix transformation on the time-domain feature data of the audio signal to obtain the signal baseline estimation matrix;

[0104] Step S144: Perform low-rank approximation on the signal baseline estimation matrix to obtain low-rank approximation data of the signal baseline; perform parallel interpolation on the time-domain feature data of the audio signal based on the low-rank approximation data of the signal baseline to obtain low-rank parallel interpolation data of the signal baseline.

[0105] Step S145: Perform parallel baseline drift removal processing on the audio signal segmentation labeling data based on the low-rank approximation data of the signal baseline and the parallel interpolation data of the low-rank signal baseline to obtain the optimized data of the audio signal baseline.

[0106] In this embodiment of the invention, firstly, the audio signal of ring spinning operation is time-series labeled. This involves aligning the audio signal with the time axis of the video data and marking the audio signal characteristics at each time point or time period. Based on the filtered data of the audio signal of the spinning operation, the audio signal time-series labeled data is segmented. This means dividing the audio signal into different segments or sections for further analysis of each segment. Time-domain feature transformation is performed on each segment in the audio signal segmented labeled data, including extracting information such as the amplitude, waveform shape, and time-domain statistical features of the segment, to further analyze the time-domain characteristics of the sound. Sparse structure analysis of the signal frequency of different signal segments is performed on the audio signal time-domain feature data, including analyzing the frequency distribution and frequency variation patterns in each segment to understand the characteristics and structure of the sound in the frequency domain. Based on the sparse structure data of the signal frequency, baseline estimation is performed on the audio signal time-domain feature data for different signal segments. This means estimating the baseline (i.e., trend or basic shape) of each signal segment based on the frequency sparse structure of the audio signal for subsequent processing. Based on the baseline estimation data, a signal baseline matrix transformation is performed on the audio signal's time-domain feature data, involving applying the baseline estimation data to the original feature data to obtain the corresponding signal baseline estimation matrix. A low-rank approximation is then performed on the signal baseline estimation matrix. This means reducing the matrix's rank to decrease data complexity and storage requirements while retaining as much information as possible. Parallel interpolation is then performed on the audio signal's time-domain feature data based on the low-rank baseline approximation data, involving using interpolation algorithms to estimate signal values ​​between missing data points, thereby filling in missing or sparse portions of the data. Finally, parallel baseline drift removal is performed on the segmented labeled audio signal data based on the low-rank baseline approximation data and the parallel low-rank baseline interpolation data. This means using the baseline estimation data and interpolation data to remove baseline drift from the original signal to optimize signal quality and accuracy.

[0107] This invention uses time-series annotation and signal segmentation to segment the audio signals from ring spinning operations. This allows the audio signals to be divided and labeled according to different time periods, facilitating subsequent analysis. This helps distinguish the operating status and problems of the spinning machine at different stages. Time-domain feature transformation and frequency sparsity analysis of the segmented audio signals allow for the acquisition of signal characteristic information from different perspectives, such as frequency distribution and waveform characteristics. This contributes to a more comprehensive understanding of the characteristics and variation patterns of the audio signals. By performing baseline estimation and low-rank approximation on the time-domain characteristic data of the audio signals, the influence of baseline drift in the signals can be effectively reduced, thereby improving the accuracy and stability of the signals. This helps reduce errors and interference during data processing. Parallel interpolation based on the low-rank approximation data can further optimize the signal quality, and parallel baseline drift removal processing is performed on the segmented audio signals using the low-rank approximation data. This helps ensure the stability and accuracy of the signals, providing a reliable foundation for subsequent data analysis and applications.

[0108] Preferably, step S2 includes the following steps:

[0109] Step S21: Extract the spinning stretching operation process data from the video data of the ring spinning operation to obtain the spinning stretching operation process data;

[0110] Step S22: Based on the baseline optimization data of the audio signal, perform spinning stretching audio signal matching on the yarn stretching operation data to obtain spinning stretching audio signal matching data;

[0111] Step S23: Perform spinning stretching audio frequency change analysis on the matching data of the spinning stretching audio frequency signal to obtain the spinning stretching audio frequency signal change data;

[0112] Step S24: Perform energy criticality calculation on the change data of the spinning stretching audio signal to obtain the energy critical data of the audio signal;

[0113] Step S25: Based on the spinning stretching audio signal change data and the audio signal energy critical data, perform ring spinning yarn breakage audio signal condition analysis to obtain yarn breakage audio signal condition data.

[0114] As an example of the present invention, reference is made to... Figure 2 As shown, in this example, step S2 includes:

[0115] Step S21: Extract the spinning stretching operation process data from the video data of the ring spinning operation to obtain the spinning stretching operation process data;

[0116] In this embodiment of the invention, the video data of the ring spinning process is preprocessed, including removing existing noise and smoothing the video sequence to reduce interference. Frame sequences are extracted from the preprocessed video data, usually at an appropriate frame rate, to capture key information of the spinning stretching operation. Image processing techniques, such as edge detection and color segmentation, are used to detect and extract the spinning stretching operation area in the video frame. Each extracted spinning stretching operation area is analyzed to calculate relevant indicators such as the degree of stretching and speed, and recorded as spinning stretching operation process data.

[0117] Step S22: Based on the baseline optimization data of the audio signal, perform spinning stretching audio signal matching on the yarn stretching operation data to obtain spinning stretching audio signal matching data;

[0118] In this embodiment of the invention, the collected audio signal is subjected to baseline optimization processing to eliminate possible background noise, drift and other interference factors. The optimized audio signal is then matched with the spinning stretching operation process data obtained in step S21. Correlation analysis, pattern matching and other methods are usually used. Based on the matching results, the correspondence and matching degree between the audio signal and the spinning stretching operation are extracted to form spinning stretching audio signal matching data.

[0119] Step S23: Perform spinning stretching audio frequency change analysis on the matching data of the spinning stretching audio frequency signal to obtain the spinning stretching audio frequency signal change data;

[0120] In this embodiment of the invention, spectrum analysis is performed on the matching data of the spinning stretching audio signal. Techniques such as Fast Fourier Transform (FFT) are typically used to convert the time-domain signal into a frequency-domain signal. Feature parameters, such as frequency changes and spectral morphology, are extracted from the spectrum to capture the changing trend of the audio signal over time. The changing patterns of the spectral features are analyzed to identify the main changing patterns of the audio signal during the spinning stretching process, such as frequency increase and amplitude change. The analyzed data of the spinning stretching audio signal changes are recorded as the basis for subsequent steps.

[0121] Step S24: Perform energy criticality calculation on the change data of the spinning stretching audio signal to obtain the energy critical data of the audio signal;

[0122] In this embodiment of the invention, the energy distribution of the audio signal is calculated based on the changes in the spinning stretching audio signal. This calculation can be performed using methods such as integrated energy or average energy. Based on the energy distribution, a method for calculating the energy threshold is determined, typically employing statistical methods or empirically based threshold settings. Using the selected calculation method, the energy threshold of the spinning stretching audio signal is calculated to distinguish between normal and abnormal audio signals.

[0123] Step S25: Based on the spinning stretching audio signal change data and the audio signal energy critical data, perform ring spinning yarn breakage audio signal condition analysis to obtain yarn breakage audio signal condition data.

[0124] In this embodiment of the invention, the data on changes in the spinning stretching acoustic signal and the critical energy data of the acoustic signal are comprehensively analyzed, taking into account factors such as frequency changes and energy distribution. Based on the analysis results, conditions for the yarn breakage acoustic signal are established, such as frequency range and energy threshold, for the detection and identification of the yarn breakage acoustic signal. The established yarn breakage acoustic signal condition data are recorded for use in subsequent yarn breakage detection processes.

[0125] This invention extracts spinning and stretching operation data from video data to obtain detailed information about the spinning machine's operating status and process, including the degree of stretching and speed. This helps monitor the spinning machine's operation. By optimizing the audio signal baseline and matching the spinning and stretching operation data with the audio signal, the sound-related parts of the operation can be identified, facilitating further analysis of the correlation between sound and operation. Change analysis of the matching audio signal data during spinning and stretching reveals how the sound changes during the process, potentially reflecting the sound characteristics at different stretching stages, which helps identify anomalies. Energy threshold calculations are performed on the changes in the spinning and stretching audio signal data to determine the sound energy threshold, helping to judge the sound intensity and changes, and further analyze the relationship between sound and yarn breakage. Based on the changes in the spinning and stretching audio signal data and the audio signal energy threshold data, conditional analysis of the yarn breakage audio signal can identify the sound characteristics at the time of yarn breakage, aiding in real-time monitoring and prevention of yarn breakage events.

[0126] Preferably, step S24 includes the following steps:

[0127] Step S241: Perform frequency domain transformation on the spinning stretching audio signal change data to obtain stretching audio signal frequency domain change data; perform frequency discrete structure analysis on the stretching audio signal frequency domain change data to obtain frequency discrete structure data.

[0128] Step S242: Divide the frequency domain variation data of the stretched audio signal into signal frequency structure regions based on the discrete structure data of the signal frequency to obtain signal frequency structure region data; extract the frequency extreme value intervals of different regions from the signal frequency structure region data to obtain the regional frequency extreme value intervals.

[0129] Step S243: Calculate the wavelet coefficients of different regions of the signal frequency structure region data according to the regional frequency extreme value intervals to obtain the regional frequency wavelet coefficients; perform multi-scale frequency energy analysis of different regions of the signal frequency structure region data according to the regional frequency wavelet coefficients to obtain the regional multi-scale frequency energy data.

[0130] Step S244: Extract frequency energy boundary intervals for different regions from the multi-scale frequency energy data of the region to obtain frequency energy boundary intervals; evaluate the frequency energy amplification rate for different regions from the frequency energy boundary intervals to obtain frequency energy amplification rate evaluation data.

[0131] Step S245: Calculate the variance of the frequency energy amplification rate assessment data to obtain the rate amplification variance data; perform frequency energy normal distribution structure analysis on the regional multi-scale frequency energy data based on the rate amplification variance data and the frequency energy boundary interval to obtain frequency energy normal distribution data; perform multi-stage sampling of the normal distribution structure on the regional multi-scale frequency energy data based on the frequency energy normal distribution data to obtain frequency energy multi-stage sampling data.

[0132] Step S246: Apply energy structure complexity constraints to the multi-stage sampling data of frequency energy to obtain energy structure complexity constraint data;

[0133] Step S247: Calculate the fractal dimension of frequency energy in different regions based on the multi-stage sampling data of frequency energy and the energy space complexity constraint data, and obtain the frequency energy fractal dimension data.

[0134] Step S248: Perform energy criticality calculation on the regional multi-scale frequency energy data based on the frequency energy fractal dimension data to obtain the audio signal energy criticality data.

[0135] In this embodiment of the invention, the spinning stretching audio signal variation data is subjected to Discrete Fourier Transform (DFT) to convert the time-domain signal into a frequency-domain signal. Amplitude spectrum analysis is performed on the DFT-derived frequency-domain signal to obtain amplitude spectrum data, and phase spectrum analysis is performed to obtain phase spectrum data. The amplitude and phase spectrum data are integrated to obtain complete stretching audio signal frequency domain variation data. Using the discrete structure data of the signal frequency, the frequency domain variation data is analyzed to identify the discrete structure characteristics of the frequency. Based on the discrete structure data of the signal frequency, the frequency domain variation data is divided into different frequency structure regions according to the degree of frequency dispersion and distribution. Within each frequency structure region, the extreme value intervals of the frequency are extracted, i.e., the maximum and minimum values ​​of the frequency within that region are determined, forming regional frequency extreme value interval data. Based on the regional frequency extreme value intervals, the signal frequency structure region data is divided into different regions. Wavelet transform is performed on the data within each region to calculate the regional frequency wavelet coefficients. This can be achieved by selecting appropriate wavelet basis functions and scales. The obtained regional frequency wavelet coefficients are then used for multi-scale frequency energy analysis. This can be achieved by calculating the energy of wavelet coefficients or reconstructing the signal using wavelet coefficients and analyzing the energy distribution of the reconstructed signal, analyzing regional multi-scale frequency energy data, and determining the boundary intervals of frequency energy. This can be achieved by setting thresholds or using statistical methods to determine energy boundaries, extracting the boundary intervals of frequency energy within each region, obtaining frequency energy boundary interval data, evaluating the frequency energy boundary intervals, and calculating the rate of increase of frequency energy. This can be achieved by analyzing the changing trend and slope of the frequency energy boundaries. The variance of the rate of increase is calculated on the frequency energy increase rate evaluation data. This can be achieved by calculating the variance of the rate of increase data, and based on the calculated rate of increase variance data and the frequency energy boundary intervals, performing a normal distribution structure analysis of the regional multi-scale frequency energy data. This can be achieved by using statistical or fitting methods to determine the normal distribution characteristics of frequency energy, and performing multi-stage sampling of the normal distribution structure of the regional multi-scale frequency energy data based on the normal distribution data. This can be achieved by using sampling methods, such as Monte Carlo sampling, to obtain multi-stage sampled data of frequency energy, and constraining the energy structure complexity of the multi-stage sampled data of frequency energy. This can be achieved by setting a complexity threshold for the energy structure or by using a certain index to measure the complexity of the energy structure. Based on the constraints, energy structure complexity constraint data is obtained, which reflects the characteristics of the multi-stage frequency energy sampling data under the constraints. Using the multi-stage frequency energy sampling data and the energy space complexity constraint data, the frequency energy fractal dimension of the regional multi-scale frequency energy data is calculated. This can be done using methods from fractal geometry theory, such as the box dimension method and ergodic method, to calculate the fractal dimension of the frequency energy data in each region, thus obtaining the frequency energy fractal dimension data.These data reflect the fractal characteristics of frequency energy at different scales. Based on the calculated fractal dimension data of frequency energy, energy criticality calculations are performed on the multi-scale frequency energy data of the region. This can be achieved by analyzing the changing trend of the frequency energy fractal dimension to determine the energy critical point. By determining the energy critical point in each region, the critical energy data of the audio signal are obtained. These data represent the critical energy level of frequency energy at different scales; that is, energy exceeding this level can be considered the critical energy of the signal.

[0136] This invention decomposes audio signals into frequency structure regions through frequency domain transformation and discrete structure analysis, helping to identify different frequency components in the sound and providing a foundation for subsequent analysis. Wavelet coefficient calculation and multi-scale frequency energy analysis within different frequency structure regions provide a deeper understanding of the energy distribution of sound across different frequency ranges, aiding in the detection of frequency energy changes and anomalies. By extracting frequency energy boundary intervals and evaluating their amplification rate and variance, the changing trends and fluctuations of sound energy can be determined, helping to identify abnormal sounds and predict yarn breakage events. Structural analysis and multi-stage sampling of the normal distribution of frequency energy provide a more comprehensive understanding of the sound energy distribution and identify possible anomalies or trend changes. By constraining the complexity of the energy structure and calculating the fractal dimension of frequency energy, the complexity and regularity of the sound signal can be further analyzed, thereby better identifying anomalies. Finally, based on the frequency energy fractal dimension data, the energy threshold of the audio signal is calculated to determine the energy difference between normal and abnormal sounds, thus achieving effective identification and prediction of yarn breakage sounds.

[0137] Preferably, step S25 includes the following steps:

[0138] Step S251: Perform periodicity analysis on the spinning stretching audio signal change data to obtain periodicity data; perform signal abrupt change point analysis on the spinning stretching audio signal change data based on the periodicity data to obtain stretching signal abrupt change point data.

[0139] Step S252: Evaluate the instantaneous intensity of the signal abrupt change at the tensile signal abrupt change point data to obtain the instantaneous intensity data at the signal abrupt change point; calculate the difference in instantaneous intensity abrupt change rate at the instantaneous intensity data at the signal abrupt change point to obtain the difference in instantaneous intensity abrupt change rate data.

[0140] Step S253: Based on the instantaneous intensity increase rate difference data and the critical data of audio signal energy, perform audio signal condition analysis on the yarn breakage in ring spinning to obtain the audio signal condition data for yarn breakage.

[0141] In this embodiment of the invention, periodicity analysis of the spinning stretching audio signal variation data is performed. This can be achieved using periodic signal processing methods, such as autocorrelation function and Fourier transform, to determine the periodicity of the signal. Based on the periodicity data of the signal variation, signal abrupt change point analysis is performed on the spinning stretching audio signal variation data. This can be done by detecting signal abrupt change points, i.e., sudden changes or jumps in signal variation, to determine the location and characteristics of the abrupt change points. The instantaneous intensity of the abrupt change is evaluated on the stretching signal abrupt change point data. This can be done by analyzing the changes in signal amplitude or slope near the abrupt change point to evaluate the intensity of the abrupt change point. The instantaneous intensity amplification rate difference is calculated on the instantaneous intensity intensity abrupt change point data. This can be done by calculating the difference in intensity amplification rate between adjacent abrupt change points to evaluate the rate of change of the abrupt change point. Based on the instantaneous intensity amplification rate difference data and the audio signal energy critical data, the conditions for the ring spinning yarn breakage audio signal are analyzed. This can be done by comparing the instantaneous intensity amplification rate difference and the energy critical data to determine the conditions for the yarn breakage audio signal. The obtained yarn breakage audio signal condition data reflects the conditions under which yarn breakage occurs and can be used to predict or control the occurrence of yarn breakage events.

[0142] This invention identifies periodic changes in spinning stretching audio signal data, such as periodic oscillations or fluctuations, through periodic analysis. Abrupt point analysis detects sudden changes or events in the signal, which may be caused by machine malfunctions or production process anomalies. These analyses help to promptly identify abnormal changes in the signal and facilitate subsequent analysis and processing. Instantaneous intensity assessment of abrupt change points in the stretching signal quantifies their intensity or importance, thus identifying which abrupt changes pose a higher risk or impact. Calculating the difference in instantaneous intensity amplification rates helps assess the rate of change in abrupt change point intensity, which is crucial for determining the severity and urgency of the abrupt change. Combining the instantaneous intensity amplification rate difference data with audio signal energy critical data allows for conditional analysis of ring spinning yarn breakage audio signals. This means that by comparing signal characteristics and critical values, the probability and severity of yarn breakage events can be determined. This facilitates timely measures to prevent or reduce yarn breakage events, improving production efficiency and product quality.

[0143] Preferably, step S3 includes the following steps:

[0144] Step S31: Based on the yarn breakage audio signal condition data, perform audio signal energy mapping processing on the audio signal energy critical data under different yarn breakage audio signal conditions to obtain yarn breakage audio signal energy mapping data;

[0145] Step S32: Analyze the factors of yarn breakage based on the energy mapping data of the yarn breakage audio signal to obtain yarn breakage factor data, which includes yarn breakage tension factor data and yarn breakage speed factor data;

[0146] Step S33: Optimize spinning behavior control based on the energy mapping data of the yarn breakage audio signal and the yarn breakage factor data to obtain optimized spinning behavior control data.

[0147] As an example of the present invention, reference is made to... Figure 3 As shown, step S3 in this example includes:

[0148] Step S31: Based on the yarn breakage audio signal condition data, perform audio signal energy mapping processing on the audio signal energy critical data under different yarn breakage audio signal conditions to obtain yarn breakage audio signal energy mapping data;

[0149] In this embodiment of the invention, based on the yarn breakage audio signal condition data, the audio signal energy critical data under different yarn breakage audio signal conditions are processed by mapping the audio signal energy. This can be achieved by establishing a mapping function or by interpolating based on the condition data. For each yarn breakage audio signal condition, corresponding audio signal energy mapping data is obtained, reflecting the distribution of audio signal energy under different conditions.

[0150] Step S32: Analyze the factors of yarn breakage based on the energy mapping data of the yarn breakage audio signal to obtain yarn breakage factor data, which includes yarn breakage tension factor data and yarn breakage speed factor data;

[0151] In this embodiment of the invention, yarn breakage factor analysis is performed based on the energy mapping data of the yarn breakage audio signal. This can be achieved by statistical analysis of the energy mapping data or by pattern recognition methods to determine the relevant factors of yarn breakage. The results of the yarn breakage factor analysis are divided into yarn breakage tension factor data and yarn breakage speed factor data, whereby the yarn breakage tension factor data reflects factors related to yarn tension, and the yarn breakage speed factor data reflects factors related to yarn speed.

[0152] Step S33: Optimize spinning behavior control based on the energy mapping data of the yarn breakage audio signal and the yarn breakage factor data to obtain optimized spinning behavior control data.

[0153] In this embodiment of the invention, spinning behavior control optimization is performed based on the energy mapping data of the yarn breakage audio signal and the yarn breakage factor data. This can be achieved by adjusting the parameters of the spinning equipment or by taking control measures to optimize the spinning behavior. The resulting spinning behavior control optimization data reflects the spinning behavior control optimization strategies under different yarn breakage factors and can be used to improve the stability and efficiency of the spinning process.

[0154] This invention processes the critical energy data of the audio signal based on the conditions of the yarn breakage audio signal to obtain yarn breakage audio signal energy mapping data. The main purpose of this step is to map the energy data of the audio signal to different yarn breakage conditions, so as to further analyze and understand the characteristics and changes of the audio signal under different conditions. Through this mapping process, the characteristics and patterns of yarn breakage events can be more clearly identified. Based on the yarn breakage audio signal energy mapping data, yarn breakage factor analysis is performed to obtain yarn breakage factor data, including yarn breakage tension factor data and yarn breakage speed factor data. The purpose of this step is to identify the key factors affecting the occurrence of yarn breakage events, such as spinning tension and spinning speed. By analyzing these factors, the root causes of yarn breakage events can be understood in depth, and corresponding improvement measures can be proposed. Combining the yarn breakage audio signal energy mapping data and the yarn breakage factor data, spinning behavior control optimization is performed. The purpose of this step is to adjust the parameters and control strategies in the spinning process according to the analysis results to minimize the occurrence of yarn breakage events and optimize the stability and efficiency of spinning behavior. Through real-time monitoring and adjustment, changes and anomalies in the production process can be responded to in a timely manner, thereby improving the stability and output of the production line.

[0155] Preferably, step S33 includes the following steps:

[0156] Step S331: Based on the yarn breakage audio signal energy mapping data, perform yarn breakage tension factor data and yarn breakage speed factor data to obtain yarn breakage tension signal energy matching data and yarn breakage speed signal energy matching data;

[0157] Step S332: Perform correlation and coupling analysis on the energy matching data of the yarn breakage speed signal and the energy matching data of the yarn breakage tension signal to obtain the correlation and coupling data of the yarn breakage speed;

[0158] Step S333: Perform deviation adjustment factor analysis on the yarn breakage speed-related tension coupling data to obtain speed-related tension potential factor data;

[0159] Step S334: Use the decision tree model to construct a potential factor association model for the potential factor data of speed-related tension, and obtain the speed-tension factor association model; learn the potential factor linear relationship model based on the yarn breakage tension signal energy matching data and the yarn breakage speed signal energy matching data to obtain the potential factor linear relationship model.

[0160] Step S335: Dynamically adjust the spinning speed according to the linear relationship model of potential factors to obtain dynamic adjustment data of spinning speed; perform tension intensity matching on the dynamic adjustment data of spinning speed to obtain tension intensity matching data;

[0161] Step S336: Optimize spinning behavior control based on dynamic adjustment data of spinning speed and tension intensity matching data to obtain optimized spinning behavior control data.

[0162] In this embodiment of the invention, yarn breakage energy matching is performed on yarn breakage tension factor data and yarn breakage speed factor data based on the yarn breakage audio signal energy mapping data. This is achieved by matching the energy mapping data with the tension factor data and speed factor data to determine the energy matching situation under different yarn breakage conditions. Based on the matching results, yarn breakage tension signal energy matching data and yarn breakage speed signal energy matching data are obtained, reflecting the matching situation between the yarn breakage tension signal and speed signal and energy, respectively. Correlation coupling analysis is performed on the yarn breakage tension signal energy matching data based on the yarn breakage speed signal energy matching data. This can be achieved by using statistical methods or correlation analysis techniques to determine the degree of correlation between the tension signal and the speed signal. Yarn breakage speed-related tension coupling data is obtained, reflecting the correlation between the speed signal and the tension signal under different yarn breakage conditions. Deviation adjustment factor analysis is performed on the yarn breakage speed-related tension coupling data. This can be achieved by analyzing the deviation or residual of the correlation data to determine the potential factors affecting the correlation. Speed-related tension potential factor data is obtained, reflecting the potential factors affecting the correlation between the speed signal and the tension signal, which can be used to further optimize the control strategy of the yarn breakage process. A potential factor correlation model is constructed using a decision tree model for the speed-related tension potential factor data. This can be achieved by employing a decision tree algorithm to construct a predictive model using speed-related tension potential factor data. Based on the yarn breakage tension signal energy matching data and yarn breakage speed signal energy matching data, the linear relationship of potential factors in the speed-tension factor correlation model is learned. This can be done using methods such as linear regression to determine the linear relationship between potential factors. Based on the linear relationship model of potential factors, the spinning speed is dynamically adjusted. This can be done dynamically based on real-time yarn breakage tension signal energy matching data and yarn breakage speed signal energy matching data, combined with the linear relationship model of potential factors. Tension intensity matching is then performed on the dynamically adjusted spinning speed data. This can be done by matching and adjusting the tension intensity based on the adjusted spinning speed data and yarn breakage tension signal energy matching data. Spinning behavior control optimization is then performed based on the dynamically adjusted spinning speed data and tension intensity matching data. This can be done by comprehensively considering speed and tension factors and adopting appropriate control strategies to optimize spinning behavior. The resulting optimized spinning behavior control data reflects the optimization effect of spinning behavior after dynamically adjusting the spinning speed and matching the tension intensity, and can be used to improve the stability and efficiency of the spinning process.

[0163] In this step, the invention performs energy matching on yarn breakage tension factor data and yarn breakage speed factor data based on the yarn breakage audio signal energy mapping data, obtaining yarn breakage tension signal energy matching data and yarn breakage speed signal energy matching data. This process correlates energy data with yarn breakage events, providing crucial data support for subsequent analysis. By using the yarn breakage speed signal energy matching data to perform correlation coupling analysis on the yarn breakage tension signal energy matching data, yarn breakage speed-related tension coupling data is obtained. This helps identify the correlation between speed and tension in yarn breakage events, further revealing the occurrence mechanism of yarn breakage events. In this step, deviation adjustment factor analysis is performed on the yarn breakage speed-related tension coupling data to obtain speed-related tension potential factor data. This analysis helps identify potential influencing factors between speed and tension, thereby better understanding the root cause of yarn breakage events. A decision tree model is used to construct a potential factor correlation model for the speed-related tension potential factor data, obtaining a speed-tension factor correlation model. This model helps understand the correlation between different factors and provides a foundation for subsequent analysis. Based on the potential factor linear relationship model, spinning speed is dynamically adjusted to obtain spinning speed dynamic adjustment data. This allows for timely adjustments to the spinning speed based on changes in potential factors during the spinning process, minimizing yarn breakage. Tension matching is then performed on the dynamic adjustment data of the spinning speed to obtain tension matching data. This step helps ensure that an appropriate level of spinning tension is maintained while adjusting the spinning speed, further reducing the risk of yarn breakage. Finally, spinning behavior control optimization is performed based on the dynamic adjustment data of the spinning speed and the tension matching data to obtain optimized spinning behavior control data. This step optimizes the parameters and control strategies in the spinning process by comprehensively considering factors such as speed and tension, thereby improving production efficiency, reducing costs, and minimizing yarn breakage.

[0164] Preferably, the deviation adjustment factor analysis of the yarn breakage speed-related tension coupling data includes the following steps:

[0165] Data standardization processing is performed on the yarn breakage speed-related tension coupling data to obtain the associated standard coupling data;

[0166] Based on the associated standard coupling data, the speed-tension coupling data of yarn breakage speed is processed by speed-tension correlation normal distribution to obtain speed-tension correlation normal distribution data;

[0167] The dispersion of the velocity-tension correlated normal distribution data is calculated to obtain the dispersion data.

[0168] Based on the data on the degree of dispersion of the velocity-tension correlation normal distribution, the normal discrete distribution AIC deviation is adjusted to obtain the normal distribution AIC deviation adjusted data;

[0169] Based on the normal distribution AIC deviation adjustment data, deviation adjustment factor analysis was performed on the yarn breakage speed-related tension coupling data to obtain speed-related tension potential factor data.

[0170] In this embodiment of the invention, the yarn breakage speed-related tension coupling data undergoes data standardization. This can be achieved using standardization methods such as z-score standardization or min-max standardization, converting the data into a standard normal distribution with a mean of 0 and a standard deviation of 1, resulting in standardized yarn breakage speed-related tension coupling data. Based on the standardized coupling data, the yarn breakage speed-related tension coupling data is then processed using a speed-tension correlation normal distribution. This can be achieved using statistical methods, such as fitting a normal distribution curve, converting the data into a normal distribution, resulting in speed-tension correlation normal distribution data, i.e., the normal distribution form of the speed-tension correlation data obtained from the standardized coupling data. The dispersion of the speed-tension correlation normal distribution data is then calculated. This can be achieved using statistical methods, such as calculating the standard deviation or coefficient of variation, to assess the dispersion of the data. The obtained dispersion data reflects the dispersion of the speed-tension correlation data and can be used for further analysis and optimization of the spinning process. Based on the dispersion data, the normal discrete distribution AIC deviation adjustment is performed on the speed-tension correlation normal distribution data. This can be achieved by using information criteria (such as the Akaike information criterion) to adjust the parameters of the normal distribution model to better fit the data, resulting in adjusted normal distribution AIC deviation data, i.e., adjusted speed-tension-related normal distribution data. Based on the adjusted normal distribution AIC deviation data, deviation adjustment factor analysis is performed on the yarn breakage speed-related tension coupling data. This can be done by comparing the differences between the adjusted data and the original data to determine the influencing factors. The resulting data on potential factors affecting speed-tension correlation reflects the potential factors influencing the speed-tension correlation and can be used for further analysis and optimization of the spinning process.

[0171] This invention standardizes the speed-tension coupling data related to yarn breakage. Standardization ensures the data has a uniform scale, eliminating the influence of different dimensions and making the data easier to compare and analyze. This helps ensure that subsequent analysis accurately reflects the relationships between the variables. The speed-tension coupling data is then processed according to a normal distribution. Converting the data to a normal distribution makes it more consistent with the assumptions of statistical analysis, leading to more accurate subsequent analysis. The dispersion of the speed-tension coupling normal distribution data is calculated. This step aims to assess the dispersion of the data, i.e., the degree of data variation. This helps determine the stability and consistency of the data, providing a basis for further analysis. Based on the dispersion data, the speed-tension coupling normal distribution data undergoes AIC (Akaike Information Criterion) bias adjustment. AIC is a commonly used model selection criterion that penalizes the model to balance goodness of fit and complexity. This step further optimizes the data distribution and improves the model's accuracy. Finally, based on the AIC bias-adjusted data, bias adjustment factor analysis is performed on the speed-tension coupling data to obtain potential factors related to speed-tension. This step can identify and assess potential factors affecting the relationship between yarn breakage speed and tension, thereby providing a better understanding of the mechanisms and influencing factors of yarn breakage events.

[0172] Preferably, the present invention also provides a signal processing-based ring spinning yarn breakage detection system for performing the signal processing-based ring spinning yarn breakage detection method described above. The signal processing-based ring spinning yarn breakage detection system includes:

[0173] The signal baseline optimization module is used to acquire video data of the ring spinning process through electronic monitoring equipment to obtain video data of the ring spinning process; extract audio signals from the video data of the ring spinning process to obtain audio signals of the ring spinning process; and perform parallel baseline drift removal processing on the audio signals of the ring spinning process to obtain baseline optimized data of the audio signals.

[0174] The yarn breakage audio signal condition analysis module is used to perform yarn stretching audio signal matching based on the audio signal baseline optimization data to obtain yarn stretching audio signal matching data; to perform yarn stretching audio signal change analysis on the yarn stretching audio signal matching data to obtain yarn stretching audio signal change data; to perform energy criticality calculation on the yarn stretching audio signal change data to obtain audio signal energy critical data; and to perform ring spinning yarn breakage audio signal condition analysis based on the audio signal energy critical data to obtain yarn breakage audio signal condition data.

[0175] The spinning behavior control optimization module is used to perform sound frequency signal energy mapping processing under different yarn breakage sound frequency signal conditions based on the yarn breakage sound frequency signal condition data, to obtain yarn breakage sound frequency signal energy mapping data; to perform yarn breakage factor analysis based on the yarn breakage sound frequency signal energy mapping data, to obtain yarn breakage factor data; and to optimize spinning behavior control based on the yarn breakage factor data, to obtain spinning behavior control optimization data.

[0176] The signal condition time period intervention coding module is used to intervene in the control time period of yarn breakage factor data based on the spinning behavior control optimization data to obtain signal condition time period control intervention data; the signal condition time period control intervention data is encoded to obtain signal condition time period intervention coded data, and the signal condition time period intervention coded data is sent to the cloud platform to perform ring spinning yarn breakage detection.

[0177] The beneficial effects of this invention are that it provides a signal processing-based method for detecting yarn breakage in ring spinning, which is an optimization of a traditional signal processing-based method for detecting yarn breakage in ring spinning. This method solves the problems of signal baseline drift affecting the signal and inaccurate yarn breakage signal analysis in the traditional method, reduces the impact of signal baseline drift on the signal, and improves the accuracy of yarn breakage signal analysis.

[0178] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0179] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A method for detecting yarn breakage in ring spinning based on signal processing, characterized in that, Includes the following steps: Step S1: Collect video data of the ring spinning process using electronic monitoring equipment to obtain video data of the ring spinning process; Audio signals were extracted from the video data of the ring spinning process to obtain the ring spinning operation audio signal; Parallel baseline drift removal processing was performed on the acoustic signals of ring spinning operation to obtain baseline optimization data for the acoustic signals. Step S2: Match the spinning stretching audio signal based on the baseline optimization data of the audio signal to obtain the spinning stretching audio signal matching data; perform spinning stretching audio signal change analysis on the matching data to obtain the spinning stretching audio signal change data; perform energy criticality calculation on the spinning stretching audio signal change data to obtain the audio signal energy critical data; perform ring spinning yarn breakage audio signal condition analysis based on the audio signal energy critical data to obtain the yarn breakage audio signal condition data. Step S3: Perform audio signal energy mapping processing under different yarn breakage audio signal conditions based on the yarn breakage audio signal condition data to obtain yarn breakage audio signal energy mapping data; perform yarn breakage factor analysis based on the yarn breakage audio signal energy mapping data to obtain yarn breakage factor data; Based on the yarn breakage factor data, the spinning behavior control optimization was carried out to obtain the spinning behavior control optimization data; Step S4: Based on the spinning behavior control optimization data, the control period intervention of the yarn breakage factor data is applied to obtain the signal condition period intervention data; The signal condition period control intervention data is encoded to obtain signal condition period intervention coded data, which is then sent to the cloud platform to perform ring spinning yarn breakage detection.

2. The ring spinning yarn breakage detection method based on signal processing according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: Collect video data of the ring spinning process using electronic monitoring equipment to obtain video data of the ring spinning process; Step S12: Extract audio signals from the video data of the ring spinning process to obtain the ring spinning operation audio signal; Step S13: Perform signal filtering on the spinning operation audio signal to obtain the spinning operation audio signal filtering data; Step S14: Perform parallel baseline drift removal processing on the ring spinning operation audio signal based on the filtered data of the spinning operation audio signal to obtain the optimized baseline data of the audio signal.

3. The ring spinning yarn breakage detection method based on signal processing according to claim 2, characterized in that, Step S14 includes the following steps: Step S141: Perform time-series annotation on the ring spinning operation audio signal to obtain audio signal time-series annotation data; perform signal segmentation on the audio signal time-series annotation data based on the filtered data of the spinning operation audio signal to obtain audio signal segmentation annotation data; Step S142: Perform time-domain feature transformation on the segmented labeled data of the audio signal to obtain time-domain feature data of the audio signal; perform sparse structure analysis of the signal frequency of different signal segments on the time-domain feature data of the audio signal to obtain sparse structure data of the signal frequency. Step S143: Based on the sparse structure data of the signal frequency, perform baseline estimation for different signal segments on the time-domain feature data of the audio signal to obtain baseline estimation data of the audio signal; based on the baseline estimation data of the audio signal, perform signal baseline matrix transformation on the time-domain feature data of the audio signal to obtain the signal baseline estimation matrix; Step S144: Perform low-rank approximation on the signal baseline estimation matrix to obtain low-rank approximation data of the signal baseline; perform parallel interpolation on the time-domain feature data of the audio signal based on the low-rank approximation data of the signal baseline to obtain low-rank parallel interpolation data of the signal baseline. Step S145: Perform parallel baseline drift removal processing on the audio signal segmentation labeling data based on the low-rank approximation data of the signal baseline and the parallel interpolation data of the low-rank signal baseline to obtain the optimized data of the audio signal baseline.

4. The ring spinning yarn breakage detection method based on signal processing according to claim 1, characterized in that, Step S2 includes the following steps: Step S21: Extract the spinning stretching operation process data from the video data of the ring spinning operation to obtain the spinning stretching operation process data; Step S22: Based on the baseline optimization data of the audio signal, perform spinning stretching audio signal matching on the yarn stretching operation data to obtain spinning stretching audio signal matching data; Step S23: Perform spinning stretching audio frequency change analysis on the matching data of the spinning stretching audio frequency signal to obtain the spinning stretching audio frequency signal change data; Step S24: Perform energy criticality calculation on the change data of the spinning stretching audio signal to obtain the energy critical data of the audio signal; Step S25: Based on the spinning stretching audio signal change data and the audio signal energy critical data, perform ring spinning yarn breakage audio signal condition analysis to obtain yarn breakage audio signal condition data.

5. The ring spinning yarn breakage detection method based on signal processing according to claim 4, characterized in that, Step S24 includes the following steps: Step S241: Perform frequency domain transformation on the spinning stretching audio signal change data to obtain stretching audio signal frequency domain change data; perform frequency discrete structure analysis on the stretching audio signal frequency domain change data to obtain frequency discrete structure data. Step S242: Divide the frequency domain variation data of the stretched audio signal into signal frequency structure regions based on the discrete structure data of the signal frequency to obtain signal frequency structure region data; extract the frequency extreme value intervals of different regions from the signal frequency structure region data to obtain the regional frequency extreme value intervals. Step S243: Calculate the wavelet coefficients of different regions of the signal frequency structure region data according to the regional frequency extreme value intervals to obtain the regional frequency wavelet coefficients; perform multi-scale frequency energy analysis of different regions of the signal frequency structure region data according to the regional frequency wavelet coefficients to obtain the regional multi-scale frequency energy data. Step S244: Extract frequency energy boundary intervals for different regions from the multi-scale frequency energy data of the region to obtain frequency energy boundary intervals; evaluate the frequency energy amplification rate for different regions from the frequency energy boundary intervals to obtain frequency energy amplification rate evaluation data. Step S245: Calculate the variance of the frequency energy amplification rate assessment data to obtain the rate amplification variance data; perform frequency energy normal distribution structure analysis on the regional multi-scale frequency energy data based on the rate amplification variance data and the frequency energy boundary interval to obtain frequency energy normal distribution data; perform multi-stage sampling of the normal distribution structure on the regional multi-scale frequency energy data based on the frequency energy normal distribution data to obtain frequency energy multi-stage sampling data. Step S246: Apply energy structure complexity constraints to the multi-stage sampling data of frequency energy to obtain energy structure complexity constraint data; Step S247: Calculate the fractal dimension of frequency energy in different regions based on the multi-stage sampling data of frequency energy and the energy space complexity constraint data, and obtain the frequency energy fractal dimension data. Step S248: Perform energy criticality calculation on the regional multi-scale frequency energy data based on the frequency energy fractal dimension data to obtain the audio signal energy criticality data.

6. The ring spinning yarn breakage detection method based on signal processing according to claim 4, characterized in that, Step S25 includes the following steps: Step S251: Perform periodicity analysis on the spinning stretching audio signal change data to obtain periodicity data; perform signal abrupt change point analysis on the spinning stretching audio signal change data based on the periodicity data to obtain stretching signal abrupt change point data. Step S252: Evaluate the instantaneous intensity of the signal abrupt change at the tensile signal abrupt change point data to obtain the instantaneous intensity data at the signal abrupt change point; calculate the difference in instantaneous intensity abrupt change rate at the instantaneous intensity data at the signal abrupt change point to obtain the difference in instantaneous intensity abrupt change rate data. Step S253: Based on the instantaneous intensity increase rate difference data and the critical data of audio signal energy, perform audio signal condition analysis on the yarn breakage in ring spinning to obtain the audio signal condition data for yarn breakage.

7. The ring spinning yarn breakage detection method based on signal processing according to claim 1, characterized in that, Step S3 includes the following steps: Step S31: Based on the yarn breakage audio signal condition data, perform audio signal energy mapping processing on the audio signal energy critical data under different yarn breakage audio signal conditions to obtain yarn breakage audio signal energy mapping data; Step S32: Analyze the factors of yarn breakage based on the energy mapping data of the yarn breakage audio signal to obtain yarn breakage factor data, which includes yarn breakage tension factor data and yarn breakage speed factor data; Step S33: Optimize spinning behavior control based on the energy mapping data of the yarn breakage audio signal and the yarn breakage factor data to obtain optimized spinning behavior control data.

8. The ring spinning yarn breakage detection method based on signal processing according to claim 7, characterized in that, Step S33 includes the following steps: Step S331: Based on the yarn breakage audio signal energy mapping data, perform yarn breakage tension factor data and yarn breakage speed factor data to obtain yarn breakage tension signal energy matching data and yarn breakage speed signal energy matching data; Step S332: Perform correlation and coupling analysis on the energy matching data of the yarn breakage speed signal and the energy matching data of the yarn breakage tension signal to obtain the correlation and coupling data of the yarn breakage speed; Step S333: Perform deviation adjustment factor analysis on the yarn breakage speed-related tension coupling data to obtain speed-related tension potential factor data; Step S334: Use the decision tree model to construct a potential factor association model for the potential factor data of speed-related tension, and obtain the speed-tension factor association model; learn the potential factor linear relationship model based on the yarn breakage tension signal energy matching data and the yarn breakage speed signal energy matching data to obtain the potential factor linear relationship model. Step S335: Dynamically adjust the spinning speed according to the linear relationship model of potential factors to obtain dynamic adjustment data of spinning speed; perform tension intensity matching on the dynamic adjustment data of spinning speed to obtain tension intensity matching data; Step S336: Optimize spinning behavior control based on dynamic adjustment data of spinning speed and tension intensity matching data to obtain optimized spinning behavior control data.

9. The ring spinning yarn breakage detection method based on signal processing according to claim 7, characterized in that, The deviation adjustment factor analysis of the yarn breakage speed-related tension coupling data includes the following steps: Data standardization processing is performed on the yarn breakage speed-related tension coupling data to obtain the associated standard coupling data; Based on the associated standard coupling data, the speed-tension coupling data of yarn breakage speed is processed by speed-tension correlation normal distribution to obtain speed-tension correlation normal distribution data; The dispersion of the velocity-tension correlated normal distribution data is calculated to obtain the dispersion data. Based on the data on the degree of dispersion of the velocity-tension correlation normal distribution, the normal discrete distribution AIC deviation is adjusted to obtain the normal distribution AIC deviation adjusted data; Based on the normal distribution AIC deviation adjustment data, deviation adjustment factor analysis was performed on the yarn breakage speed-related tension coupling data to obtain speed-related tension potential factor data.

10. A ring spinning yarn breakage detection system based on signal processing, characterized in that, For performing the signal processing-based ring spinning yarn breakage detection method as described in claim 1, the signal processing-based ring spinning yarn breakage detection system comprises: The signal baseline optimization module is used to acquire video data of the ring spinning process through electronic monitoring equipment to obtain video data of the ring spinning process; extract audio signals from the video data of the ring spinning process to obtain audio signals of the ring spinning process; and perform parallel baseline drift removal processing on the audio signals of the ring spinning process to obtain baseline optimized data of the audio signals. The yarn breakage audio signal condition analysis module is used to perform yarn stretching audio signal matching based on the audio signal baseline optimization data to obtain yarn stretching audio signal matching data; to perform yarn stretching audio signal change analysis on the yarn stretching audio signal matching data to obtain yarn stretching audio signal change data; to perform energy criticality calculation on the yarn stretching audio signal change data to obtain audio signal energy critical data; and to perform ring spinning yarn breakage audio signal condition analysis based on the audio signal energy critical data to obtain yarn breakage audio signal condition data. The spinning behavior control optimization module is used to perform sound frequency signal energy mapping processing under different yarn breakage sound frequency signal conditions based on the yarn breakage sound frequency signal condition data, to obtain yarn breakage sound frequency signal energy mapping data; to perform yarn breakage factor analysis based on the yarn breakage sound frequency signal energy mapping data, to obtain yarn breakage factor data; and to optimize spinning behavior control based on the yarn breakage factor data, to obtain spinning behavior control optimization data. The signal condition time period intervention coding module is used to intervene in the control time period of yarn breakage factor data based on the spinning behavior control optimization data to obtain signal condition time period control intervention data; the signal condition time period control intervention data is encoded to obtain signal condition time period intervention coded data, and the signal condition time period intervention coded data is sent to the cloud platform to perform ring spinning yarn breakage detection.

Images