CN118482817B - Digital textile equipment management method and system based on data analysis - Google Patents
Digital textile equipment management method and system based on data analysis Download PDFInfo
- Publication number
- CN118482817B CN118482817B CN202410933082.9A CN202410933082A CN118482817B CN 118482817 B CN118482817 B CN 118482817B CN 202410933082 A CN202410933082 A CN 202410933082A CN 118482817 B CN118482817 B CN 118482817B
- Authority
- CN
- China
- Prior art keywords
- color difference
- detection
- difference detector
- potential
- index
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 239000004753 textile Substances 0.000 title claims abstract description 90
- 238000007726 management method Methods 0.000 title claims abstract description 26
- 238000007405 data analysis Methods 0.000 title claims abstract description 25
- 238000001514 detection method Methods 0.000 claims abstract description 214
- 230000003595 spectral effect Effects 0.000 claims abstract description 104
- 230000005856 abnormality Effects 0.000 claims abstract description 82
- 238000004458 analytical method Methods 0.000 claims abstract description 69
- 238000000034 method Methods 0.000 claims abstract description 67
- 230000008569 process Effects 0.000 claims abstract description 38
- 238000010801 machine learning Methods 0.000 claims abstract description 31
- 238000012544 monitoring process Methods 0.000 claims abstract description 26
- 238000002310 reflectometry Methods 0.000 claims abstract description 23
- 230000008859 change Effects 0.000 claims description 65
- 230000002159 abnormal effect Effects 0.000 claims description 56
- 239000011159 matrix material Substances 0.000 claims description 22
- 238000012545 processing Methods 0.000 claims description 21
- 238000005516 engineering process Methods 0.000 claims description 17
- 238000012549 training Methods 0.000 claims description 17
- 230000006870 function Effects 0.000 claims description 16
- 238000010183 spectrum analysis Methods 0.000 claims description 11
- 230000004044 response Effects 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 7
- 230000004927 fusion Effects 0.000 claims description 6
- 238000012502 risk assessment Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 230000001133 acceleration Effects 0.000 claims description 3
- 238000009826 distribution Methods 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 abstract description 20
- 239000000047 product Substances 0.000 description 14
- 238000012423 maintenance Methods 0.000 description 6
- 238000004422 calculation algorithm Methods 0.000 description 5
- 238000004590 computer program Methods 0.000 description 4
- 238000003860 storage Methods 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 239000003086 colorant Substances 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000011156 evaluation Methods 0.000 description 3
- 230000007774 longterm Effects 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 238000010606 normalization Methods 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 238000003908 quality control method Methods 0.000 description 3
- 230000004075 alteration Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000004140 cleaning Methods 0.000 description 2
- 238000002790 cross-validation Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 230000002547 anomalous effect Effects 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000011109 contamination Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 239000012467 final product Substances 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000003449 preventive effect Effects 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Landscapes
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses a digital textile equipment management method and system based on data analysis, which relate to the technical field of textile equipment management and comprise the following steps: in the process of carrying out color detection on textiles by the online color difference detector, the online color difference detector is embedded with a high-precision sensor and a data acquisition system, and the operation parameter information of the online color difference detector is acquired in real time and stored in a central database. According to the invention, through monitoring the operation parameters and spectral reflectivity information of the online color difference detector in real time, the machine learning model is utilized to perform anomaly analysis and detection, potential problems are identified in time, an early warning signal is generated, the problem of inconsistent color difference of large batches of products is avoided, and the economic losses of reworking and scrapping are reduced. After potential abnormality is detected, potential risk indexes are generated and risk grades are divided through multi-level threshold comparison and comprehensive analysis, and targeted countermeasures are taken to ensure the product quality and improve the production efficiency and consistency.
Description
Technical Field
The invention relates to the technical field of textile equipment management, in particular to a digital textile equipment management method and system based on data analysis.
Background
The digital textile equipment management refers to a modern management mode for comprehensively monitoring, managing and optimizing textile production equipment through digital technology and tools. The management mode utilizes advanced technologies such as the Internet of things, sensors, big data analysis, cloud computing and the like to collect and analyze equipment operation data in real time, so that the automation of real-time monitoring, fault prediction, production efficiency optimization and maintenance management of equipment states is realized. Through digital textile equipment management, enterprises can improve production efficiency, reduce equipment failure rate and downtime, optimize resource allocation and reduce operation cost. Meanwhile, the management mode can help enterprises to better cope with the change of market demands, the production plan is quickly adjusted, the product quality and the market competitiveness are improved, and finally the aims of intelligent manufacturing and lean production are achieved.
An online color difference detector is used in the digital textile equipment management, and the online color difference detector has the main function of monitoring and controlling the color consistency in the textile production process in real time. Through high-precision spectrum analysis technology, the online color difference detector can detect the colors of textiles at different stages on a production line, capture tiny color difference changes and feed detection data back to a production control system in real time. When the color difference is detected to exceed the preset standard, the system can immediately adjust to ensure that the color of the final product is consistent with the design standard. The method not only effectively avoids the product quality reduction and customer complaints caused by the chromatic aberration problem, but also reduces the material waste and the reworking cost caused by the chromatic aberration problem. The on-line color difference detector can greatly improve the production efficiency and the product quality stability, and is an indispensable important device in the modern textile industry. By introducing the efficient online detection technology, textile enterprises can better cope with the demands of markets on high-quality and high-consistency products, and the competitiveness and market praise of the enterprises are improved. In addition, the online color difference detector can also help enterprises to accumulate a large amount of production data, provide data support for subsequent production optimization and intelligent manufacturing, and realize digital management and quality control of the whole process.
When the online color difference detector detects the colors of the textiles through a high-precision spectrum analysis technology, the prior art cannot find the potential problems existing in the detection precision of the online color difference detector in time, the potential problems can be known by related staff only when the textiles have obvious color difference changes, and the color difference problems cannot be detected and corrected in time, so that the color of the textiles produced in large quantity is inconsistent, and the quality standard of customers cannot be met by the whole batch of products, so that the large batch of products need to be reworked or scrapped, and huge economic loss is caused.
The above information disclosed in the background section is only for enhancement of understanding of the background of the disclosure and therefore it may include information that does not form the prior art that is already known to a person of ordinary skill in the art.
Disclosure of Invention
The invention aims to provide a digital textile equipment management method and system based on data analysis, which are used for monitoring the operation parameters and spectral reflectivity information of an online color difference detector in real time by embedding a high-precision sensor and a data acquisition system, carrying out anomaly analysis and detection by using a machine learning model, identifying potential problems in time, generating an early warning signal, avoiding the problem of inconsistent color difference of large batches of products and reducing the economic loss of reworking and scrapping. After potential abnormality is detected, potential risk indexes are generated and risk grades are divided through multi-level threshold comparison and comprehensive analysis, targeted treatment measures are taken, product quality is ensured, and production efficiency and consistency are improved, so that the problems in the background technology are solved.
In order to achieve the above object, the present invention provides the following technical solutions: a digital textile equipment management method based on data analysis comprises the following steps:
In the process of carrying out color detection on textiles by the online color difference detector, the online color difference detector is embedded with a high-precision sensor and a data acquisition system, and the operation parameter information of the online color difference detector is acquired in real time and stored in a central database for subsequent analysis;
Performing exception analysis on the preprocessed operation parameter information, detecting a potential exception mode in the operation parameters, and identifying parameter changes of the online color difference detector, which are different from the normal operation state;
Training a machine learning model by using a supervised learning method, deploying the trained machine learning model in an online color difference detection system based on an abnormal analysis result, monitoring an online color difference detector in real time, and identifying potential abnormality existing in the detection precision of the online color difference detector;
The online color difference detector with the potential abnormality of the detection precision is further comprehensively analyzed, the risk level of the potential abnormality of the detection precision is judged, and different countermeasures are adopted for the potential abnormality of the detection precision with different risk levels.
Preferably, the operation parameter information of the online color difference detector comprises spectral reflectivity information and electronic drift information, wherein the spectral reflectivity information refers to the proportion of reflected light of the textile surface under the irradiation of light with different wavelengths, and the electronic drift information refers to the output signal deviation condition caused by the change of the performance of electronic elements in the online color difference detector along with time.
Preferably, after the spectral reflectance information and the electronic drift information of the online color difference detector during color detection of the textile are obtained through a high-precision spectral analysis technology, the spectral reflectance abnormal index is generated after abnormal analysis processing is carried out on the spectral reflectance information, the electronic drift index is generated after abnormal analysis processing is carried out on the electronic drift information, and the parameter changes of the online color difference detector, which are different from the normal running state, are identified through the spectral reflectance abnormal index and the electronic drift index.
Preferably, after the spectral reflectance abnormality index and the electronic drift index generated when the online color difference detector detects the colors of the textiles through a high-precision spectral analysis technology are obtained, the spectral reflectance abnormality index and the electronic drift index are input into a pre-trained machine learning model, a detection precision coefficient is generated through the machine learning model, and the precision of the online color difference detector during the color detection of the textiles is evaluated through the detection precision coefficient.
Preferably, the detection precision coefficient generated in the fixed detection window when the online color difference detector detects the textile color is compared with a preset detection precision coefficient reference threshold value for analysis, and potential abnormality existing in the detection precision of the online color difference detector is identified, and the comparison analysis result is as follows:
If the detection precision coefficient is greater than or equal to the detection precision coefficient reference threshold, generating a high-efficiency detection signal, wherein the high-efficiency detection signal indicates that the high-efficiency detection can be realized when the online color difference detector detects the color of the textile;
if the detection precision coefficient is smaller than the detection precision coefficient reference threshold, generating a detection potential abnormal signal, which indicates that the online color difference detector has potential detection abnormality when detecting the color of the textile.
Preferably, when the online color difference detector detects the textile color, a detection potential abnormal signal is generated in a fixed detection window, an analysis set is established by continuously acquiring a plurality of detection precision coefficients generated in the fixed detection window when the online color difference detector detects the textile color, the detection precision coefficients in the analysis set are compared with a first gradient reference threshold value and a second gradient reference threshold value, wherein the second gradient reference threshold value is smaller than the detection precision coefficient reference threshold value, the detection precision coefficients are compared with the second gradient reference threshold value, the first gradient reference threshold value and the detection precision coefficient reference threshold value, the number of the detection precision coefficients which are larger than or equal to the first gradient reference threshold value and smaller than the detection precision coefficient reference threshold value is recorded as Ka, the number of the detection precision coefficients which are larger than or equal to the second gradient reference threshold value and smaller than the first gradient reference threshold value is recorded as Kb, and the number of the detection precision coefficients which are smaller than the second gradient reference threshold value is recorded as Kc.
Comprehensively analyzing Ka, kb and Kc to generate a potential risk indexThe formula according to is: In which, in the process, Weight coefficients of Ka, kb, kc, respectively, and。
Preferably, the on-line color difference detector with the potential abnormality of the detection precision is used for further comprehensively analyzing and comparing the generated potential risk index with a preset first potential risk index reference threshold value and a preset second potential risk index reference threshold value, judging the risk level of the potential abnormality of the detection precision, and comparing and analyzing results as follows:
If the potential risk index is smaller than the first potential risk index reference threshold, classifying the risk level of the detection precision potential abnormality into a low risk potential abnormality;
If the potential risk index is greater than or equal to the first potential risk index reference threshold value and less than the second potential risk index reference threshold value, classifying the risk level of the detection precision potential abnormality as a general risk potential abnormality;
And if the potential risk index is greater than or equal to the second potential risk index reference threshold, classifying the risk level of the detection precision potential abnormality into a high risk potential abnormality.
Preferably, after performing anomaly analysis processing on the spectral reflectance information, the specific steps of generating a spectral reflectance anomaly index are as follows:
In the fixed detection window, collecting spectral reflectance data of a plurality of wavelengths, and setting the collected spectral reflectance data as WhereinRepresents the i-th wavelength, i=1, 2,..n, N is the number of wavelengths collected;
Preprocessing the collected spectral reflectance data, and recording the preprocessed spectral reflectance data as ；
The rate of change of spectral reflectance with wavelength is calculated, and the expression is calculated as follows: In which, in the process, Representing the derivative of the spectral reflectance with respect to wavelength,Representing the rate of change of the spectral reflectance;
Calculating a second derivative of the rate of change to capture the rate of change acceleration of the reflectance curve, the calculated expression being: In which, in the process, A second derivative representing the rate of change of spectral reflectance;
defining a comprehensive anomaly detection function The method is characterized in that the change rate of the spectral reflectance and the second derivative of the change rate of the spectral reflectance are combined to capture the anomaly, and the expression of the comprehensive anomaly detection function is as follows: In which, in the process, Representing the function of the integrated anomaly detection,AndRates of change of spectral reflectance, respectivelyAnd the second derivative of the rate of change of spectral reflectanceControlling the rate of change of spectral reflectanceAnd the second derivative of the rate of change of spectral reflectanceIs a weight of influence of (1);
In the fixed detection window, the abnormal detection function values of all wavelengths are accumulated to generate a spectral reflectance abnormal index, and the generated expression is: In which, in the process, Indicating the spectral reflectance anomaly index.
Preferably, after the electronic drift information is subjected to the anomaly analysis processing, the specific steps of generating the electronic drift index are as follows:
In a fixed detection window, acquiring reference signal data of an electronic element in real time, wherein the reference signal data comprises a voltage signal V, a current signal I, a temperature T and a time T;
preprocessing the acquired signal data;
Extracting characteristic parameters from the preprocessed data, wherein the characteristic parameters comprise instantaneous voltage change rate, instantaneous current change rate, temperature change rate and time interval, and the extracted expression is as follows: ，， In which, in the process, AndRespectively representing the instantaneous voltage change rate, the instantaneous current change rate, the temperature change rate and the time interval;
The extracted characteristic parameters are subjected to multidimensional characteristic fusion to construct a multidimensional characteristic vector F, the multidimensional characteristic vector F comprises instantaneous voltage change rate, instantaneous current change rate, temperature change rate and time interval multidimensional characteristics, ；
The weight matrix W is calculated through a weighted least square method and used for weight distribution of multidimensional feature fusion, and the calculation expression of the weight matrix W is as follows: Wherein X is a feature matrix and Y is a historical drift index vector;
The multidimensional feature vector F and the weight matrix W are linearly combined, an electronic drift index is calculated, the electronic drift index reflects the comprehensive degree of electronic element drift, and the calculated expression is: In which, in the process, Representing the transpose of the weight matrix W,Representing the electron drift index.
A digital textile equipment management system based on data analysis comprises a data acquisition module, an anomaly analysis module, a machine learning monitoring module and a risk assessment and response module;
The data acquisition module is used for acquiring the operation parameter information of the online color difference detector in real time and storing the operation parameter information in a central database for subsequent analysis by embedding a high-precision sensor and a data acquisition system into the online color difference detector in the process of carrying out color detection on textiles by the online color difference detector;
the abnormality analysis module is used for carrying out abnormality analysis on the preprocessed operation parameter information, detecting potential abnormal modes in the operation parameters and identifying parameter changes of the online color difference detector, which are different from the normal operation state;
The machine learning monitoring module is used for training a machine learning model by using a supervised learning method, deploying the trained machine learning model in the online color difference detection system based on an abnormal analysis result, monitoring the online color difference detector in real time, and identifying potential abnormality existing in the detection precision of the online color difference detector;
and the risk assessment and response module is used for further comprehensively analyzing the online color difference detector with the potential abnormality of the detection precision, judging the risk level of the potential abnormality of the detection precision and taking different response measures for the potential abnormality of the detection precision of different risk levels.
In the technical scheme, the invention has the technical effects and advantages that:
According to the invention, the high-precision sensor and the data acquisition system are embedded, the operation parameters and the spectral reflectivity information of the online color difference detector are acquired in real time, the abnormal analysis and detection are carried out, the detection precision is monitored and evaluated in real time by using the machine learning model, the potential abnormality of the operation state of the detector can be identified in time, the high-efficiency detection signal or the potential abnormality signal is generated, the real-time monitoring and early warning mechanism can find problems before obvious color difference changes occur to textiles, the color inconsistency of large batches of products is avoided, and the economic loss of reworking or scrapping is reduced.
According to the method, after the potential abnormality is detected, the potential risk index is generated through multi-level threshold comparison and comprehensive analysis, the risk grades are divided, corresponding treatment measures are adopted according to different risk grades, the potential abnormality problem of the detection precision can be more accurately treated, especially for the potential abnormality with high risk, measures can be immediately adopted to correct, the product quality is ensured to meet the customer standard, and the consistency of the production efficiency and the product quality is improved.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments described in the present application, and other drawings may be obtained according to these drawings for those skilled in the art.
Fig. 1 is a flow chart of a method for managing digital textile equipment based on data analysis.
Fig. 2 is a schematic block diagram of a digital textile equipment management system based on data analysis according to the present invention.
Detailed Description
Example embodiments will now be described more fully with reference to the accompanying drawings. However, the exemplary embodiments may be embodied in many forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of the example embodiments to those skilled in the art.
The invention provides a digital textile equipment management method based on data analysis as shown in fig. 1, which comprises the following steps:
In the process of carrying out color detection on textiles by the online color difference detector, the online color difference detector is embedded with a high-precision sensor and a data acquisition system, and the operation parameter information of the online color difference detector is acquired in real time and stored in a central database for subsequent analysis;
The operation parameter information of the online color difference detector comprises spectral reflectivity information and electronic drift information, wherein the spectral reflectivity information refers to the proportion of reflected light of the textile surface under the irradiation of light with different wavelengths, and the electronic drift information refers to the output signal deviation condition caused by the change of the performance of electronic elements in the online color difference detector along with time.
The acquisition of spectral reflectance information by a high-precision sensor (this sensor is typically a spectral sensor or photometer, the spectral sensor can emit and receive light of specific wavelengths and measure the reflectance of the textile for these wavelengths to generate a spectrogram of the textile for color detection) and a data acquisition system can be performed as follows:
First, a high-precision sensor in the system emits light of a specific wavelength to the textile surface. The light reflected by the textile surface is then received by the spectral sensor and converted into an electrical signal. The data acquisition system then acquires and processes these electrical signals in real time, converting them into spectral reflectance data. Finally, the data is stored in a central database for subsequent analysis and processing. Through the steps, the spectral reflectivity information of the textile can be accurately obtained and used for color detection and quality control.
The acquisition of electronic drift information by a high-precision sensor (this sensor is typically a voltage reference sensor or an electronic drift detection sensor, which is used to measure and calibrate the voltage reference signal of an electronic component, monitoring its drift, the electronic drift detection sensor is dedicated to capturing the performance variations of the electronic component under different operating conditions, ensuring the long-term stability and accuracy of the device) and the data acquisition system can be performed according to the following steps:
First, the high-precision sensor periodically measures a reference signal of the electronic component when the detector is operating normally. The data acquisition system then records these reference signals in real time and compares them to the initial calibration values. Then, the system removes noise and interference through filtering and signal processing technology, and accurately calculates the drift amount of the reference signal. The system then converts these drift amount data to digital signals, generating an electron drift curve or drift index. Finally, these electronic drift information are stored in a central database for analysis and correction by maintenance personnel.
Performing exception analysis on the preprocessed operation parameter information, detecting a potential exception mode in the operation parameters, and identifying parameter changes of the online color difference detector, which are different from the normal operation state;
After spectral reflectance information and electronic drift information of the online color difference detector during color detection of textiles are obtained through a high-precision spectral analysis technology, abnormal analysis processing is carried out on the spectral reflectance information, a spectral reflectance abnormal index is generated, abnormal analysis processing is carried out on the electronic drift information, an electronic drift index is generated, and parameter changes of the online color difference detector, which are different from normal operation states, are identified through the spectral reflectance abnormal index and the electronic drift index.
When the online color difference detector detects the color of the textile through a high-precision spectrum analysis technology, if the spectral reflectivity is abnormal, the detection precision of the detector during the color detection of the textile is abnormal. This is because spectral reflectance is an important parameter in assessing the color of a textile, reflecting the reflective properties of the textile surface for light of different wavelengths. If the spectral reflectance data is anomalous, such as an unreasonable peak, deviation or fluctuation, this may be due to sensor failure, unstable light sources, ambient light interference or contamination of the textile surface. These abnormal data can directly affect the accuracy of color analysis, so that the detector cannot correctly evaluate the actual color of the textile, and further cause errors in color difference detection results. Accurate spectral reflectance data is the basis for ensuring color detection accuracy, and any anomalies can lead to false color matching and quality control problems. Therefore, the abnormal spectral reflectivity is found and corrected in time, and is very important to the detection precision of the on-line color difference detector.
After carrying out anomaly analysis processing on the spectral reflectance information, the specific steps for generating the spectral reflectance anomaly index are as follows:
In the fixed detection window, collecting spectral reflectance data of a plurality of wavelengths, and setting the collected spectral reflectance data as WhereinRepresents the i-th wavelength, i=1, 2,..n, N is the number of wavelengths collected;
Preprocessing the collected spectral reflectance data, and recording the preprocessed spectral reflectance data as ；
The rate of change of spectral reflectance with wavelength is calculated, and the expression is calculated as follows: In which, in the process, Representing the derivative of spectral reflectance with respect to wavelength, may be calculated by numerical differentiation,Representing the rate of change of the spectral reflectance;
Calculating a second derivative of the rate of change to capture the rate of change acceleration of the reflectance curve, the calculated expression being: In which, in the process, The second derivative representing the rate of change of the spectral reflectance can be used to determine the rate of change by numerical differentiationPerforming difference again to obtain;
defining a comprehensive anomaly detection function The method is characterized in that the change rate of the spectral reflectance and the second derivative of the change rate of the spectral reflectance are combined to capture the anomaly, and the expression of the comprehensive anomaly detection function is as follows: In which, in the process, Representing the function of the integrated anomaly detection,AndRates of change of spectral reflectance, respectivelyAnd the second derivative of the rate of change of spectral reflectanceControlling the rate of change of spectral reflectanceAnd the second derivative of the rate of change of spectral reflectanceIs used for the influence weight of the (c) in the (c),AndThe value of (2) is not particularly limited herein, and can be adjusted according to actual conditions and requirements;
In the fixed detection window, the abnormal detection function values of all wavelengths are accumulated to generate a spectral reflectance abnormal index, and the generated expression is: In which, in the process, Indicating the spectral reflectance anomaly index.
When the online color difference detector detects the color of the textile by a high-precision spectrum analysis technology, the appearance value of the abnormal index of the spectral reflectivity reflects the quality of the detection precision. Specifically, in the fixed detection window, the larger the value of the spectral reflectance abnormality index generated after the abnormal analysis processing of the spectral reflectance information is, the more abnormality is shown in the spectral reflectance data, which means that larger errors and instability may exist in the detection process, resulting in the poorer detection accuracy. In contrast, the smaller the value of the spectral reflectance anomaly index, the more stable and normal the spectral reflectance data, and the smaller the error in the detection process, and therefore the higher the detection accuracy.
The electronic drift of the online color difference detector can cause the abnormal detection precision of the online color difference detector when the online color difference detector detects the color of the textile. Electronic drift refers to a small signal shift that occurs with time or environmental condition changes of electronic components inside the online color difference detector. Such drift can cause the reference signal of the measurement circuit to change, thereby affecting the accuracy of the spectral data acquired by the spectral sensor. When electron drift occurs, otherwise stable spectral reflectance data may deviate, and these deviations may directly lead to inaccuracy in the color detection result. Since color difference detection requires extremely high accuracy, any small electron drift may be amplified to a significant detection error, resulting in a final color measurement that does not match the actual color. Therefore, monitoring and correcting the electron drift is an important measure to ensure the long-term stability and detection accuracy of the online color difference detector. Through regular reference signal measurement and data calibration, detection abnormality caused by electronic drift can be effectively prevented and corrected, and high precision and reliability of textile color detection are ensured.
After the electronic drift information is subjected to abnormal analysis processing, the specific steps of generating the electronic drift index are as follows:
In a fixed detection window, acquiring reference signal data of an electronic element in real time, wherein the reference signal data comprises a voltage signal V, a current signal I, a temperature T and a time T;
These parameters are key indicators of the reference signal data, reflecting the operating state of the electronic component.
Preprocessing the acquired signal data;
The preprocessing comprises denoising and normalization processing, wherein random noise is removed by the denoising processing through a filter, and the normalization processing converts data with different orders into dimensionless values, so that subsequent calculation is facilitated.
Extracting characteristic parameters from the preprocessed data, wherein the characteristic parameters comprise instantaneous voltage change rate, instantaneous current change rate and temperature change rate and time interval (the characteristic parameters can reflect dynamic change conditions of electron drift), and the extracted expression is as follows:，， In which, in the process, AndRespectively representing the instantaneous voltage change rate, the instantaneous current change rate, the temperature change rate and the time interval;
The extracted characteristic parameters are subjected to multidimensional characteristic fusion to construct a multidimensional characteristic vector F, the multidimensional characteristic vector F comprises instantaneous voltage change rate, instantaneous current change rate, temperature change rate and time interval multidimensional characteristics, ；
The weight matrix W is calculated through a weighted least square method and used for weight distribution of multidimensional feature fusion, and the calculation expression of the weight matrix W is as follows: Wherein X is a feature matrix and Y is a historical drift index vector;
The weight matrix is calculated by a weighted least square method according to the importance and the historical data of each characteristic parameter, and the characteristic matrix And T in (a) represents the transpose of the matrix. The transposed matrix is a matrix obtained by interchanging rows and columns of the original matrix. Transpose operations are very common in matrix operations, especially in the computation of weighted least squares and the like, to ensure that the dimensions of the matrices match.
The multidimensional feature vector F and the weight matrix W are linearly combined, an electronic drift index is calculated, the electronic drift index reflects the comprehensive degree of electronic element drift, and the calculated expression is: In which, in the process, Representing the transpose of the weight matrix W,Representing the electron drift index.
The larger the expression value of the electronic drift index is, the worse the detection precision of the online color difference detector is, and the higher the detection precision of the online color difference detector is. The electronic drift index is a comprehensive index generated by carrying out abnormal analysis processing on the electronic drift information acquired in the detection process, and reflects the performance drift degree of electronic elements in the detector. The higher electron drift index indicates that the drift of the electronic element is larger, and the stability of the signal is affected, so that the color detection result is inaccurate; the lower electron drift index indicates that the electronic element works stably, the drift is smaller, and the detection precision is higher.
Training a machine learning model by using a supervised learning method, deploying the trained machine learning model in an online color difference detection system based on an abnormal analysis result, monitoring an online color difference detector in real time, and identifying potential abnormality existing in the detection precision of the online color difference detector;
the specific steps for training the machine learning model using the supervised learning method are as follows:
First, a training dataset is collected and sorted, ensuring that the dataset contains enough labeled samples, each sample comprising a feature vector (input) and a corresponding target value (output). The data is preprocessed, including cleaning, denoising, normalizing and segmenting training and testing sets. The data cleaning is to remove abnormal values and missing values, the denoising process eliminates random noise in the data, and the normalization process ensures that different features are in the same scale. Finally, the data is divided into training and testing sets, typically in a ratio of 70:30 or 80:20, for model training and evaluation.
Machine learning algorithms suitable for the task, such as linear regression, support Vector Machines (SVMs), decision trees, random forests, neural networks, etc., are selected. Based on the task type (regression or classification) and the data characteristics, an appropriate algorithm is selected. Initializing model parameters, and for complex models such as neural networks, setting network structures (such as layer numbers, the number of neurons per layer) and super-parameters (such as learning rate, regularization parameters and the like) are required. If it is not determined which algorithm is most appropriate, multiple models can be tried and the best performing model selected by cross-validation.
The training set is input into the selected model, and model parameters are adjusted through the training process of the algorithm, so that the prediction result of the model is as close to the actual target value as possible. The training process typically involves iterative optimization, such as minimizing a loss function (e.g., mean square error, cross entropy, etc.) using a gradient descent method. In each iteration, the model calculates the prediction error from the training samples, and then updates the weights and biases by back propagation. For some complex models, it is also necessary to adjust the superparameters, and to evaluate the effect of different superparameter combinations using cross-validation to find the optimal parameter settings.
The test set is used to evaluate the trained model and calculate the behavior of the model on unseen data. Common evaluation indexes include accuracy, precision, recall, F1 fraction, mean square error, etc. Based on the evaluation result, the model is further tuned, for example, by adjusting super parameters, adding training data, adopting data enhancement technology, changing the model structure, and the like, to improve the model performance. If the model performs well on the test set and the risk of overfitting is low, the model can be considered to achieve the desired effect, ready for practical use. Otherwise, it may be necessary to return to the first few steps to readjust or select a different method.
Acquiring abnormal index of spectral reflectivity generated by the online color difference detector when the textile is subjected to color detection by high-precision spectral analysis technologyAnd electron drift indexThen, the spectral reflectance anomaly index is calculatedAnd electron drift indexInputting the detection precision coefficient into a pre-trained machine learning model, and generating the detection precision coefficient through the machine learning modelBy detecting the precision coefficientAnd evaluating the precision of the online color difference detector during textile color detection.
The machine learning model is not particularly limited herein, and can realize the index of spectral reflectance anomalyAnd electron drift indexComprehensive analysis is carried out to generate a detection precision coefficientIn order to realize the technical scheme of the invention, the invention provides a specific implementation mode;
Detection precision coefficient The generated calculation formula is as follows: In which, in the process, Respectively, spectral reflectance anomaly indexesAnd electron drift indexIs a preset proportionality coefficient of (1), andAre all greater than 0.
According to the detection precision coefficient, when the online color difference detector detects the color of the textile through a high-precision spectrum analysis technology in a fixed detection window, the larger the expression value of the spectrum reflectivity abnormal index generated after the abnormal analysis processing of the spectrum reflectivity information is, the larger the expression value of the electron drift index generated after the abnormal analysis processing of the electron drift information is, namely the smaller the expression value of the generated detection precision coefficient is, the worse the monitoring precision of the online color difference detector is, and otherwise, the better the monitoring precision of the online color difference detector is.
Comparing and analyzing the detection precision coefficient generated in the fixed detection window when the online color difference detector detects the textile color with a preset detection precision coefficient reference threshold value, and identifying potential abnormality of the detection precision of the online color difference detector, wherein the comparison and analysis result is as follows:
If the detection precision coefficient is greater than or equal to the detection precision coefficient reference threshold, generating a high-efficiency detection signal, wherein the high-efficiency detection signal indicates that the high-efficiency detection can be realized when the online color difference detector detects the color of the textile;
if the detection precision coefficient is smaller than the detection precision coefficient reference threshold, generating a detection potential abnormal signal, which indicates that the online color difference detector has potential detection abnormality when detecting the color of the textile.
Further comprehensively analyzing the online color difference detector with the potential abnormality of the detection precision, judging the risk level of the potential abnormality of the detection precision, and taking different countermeasures for the potential abnormality of the detection precision of different risk levels;
When a detection potential abnormal signal is generated in a fixed detection window when the online color difference detector detects the textile color, continuously acquiring a plurality of detection precision coefficients generated in the fixed detection window when the online color difference detector detects the textile color, establishing an analysis set, comparing the detection precision coefficients in the analysis set with a first gradient reference threshold value and a second gradient reference threshold value, wherein the second gradient reference threshold value is smaller than the first gradient reference threshold value and smaller than the detection precision coefficient reference threshold value, comparing the detection precision coefficients with the second gradient reference threshold value, the first gradient reference threshold value and the detection precision coefficient reference threshold value, recording the number of the detection precision coefficients which are larger than or equal to the first gradient reference threshold value and smaller than the detection precision coefficient reference threshold value as Ka, recording the number of the detection precision coefficients which are larger than or equal to the second gradient reference threshold value and smaller than the first gradient reference threshold value as Kb, and recording the number of the detection precision coefficients which are smaller than the second gradient reference threshold value as Kc.
Comprehensively analyzing Ka, kb and Kc to generate a potential risk indexThe formula according to is: In which, in the process, Weight coefficients of Ka, kb, kc, respectively, and；
From the risk potential index, it is known that the larger the expression value of Kc is, the smaller the expression values of Ka and Kb are, that is, the generated risk potential indexThe larger the expression value of the color difference detector is, the more serious the potential abnormality of the detection precision is, which occurs when the online color difference detector detects the textile color, and the more slight the potential abnormality of the detection precision is, which occurs when the online color difference detector detects the textile color.
The potential risk index generated by further comprehensive analysis of the online color difference detector with the potential abnormality of the detection precision is compared with a preset first potential risk index reference threshold value and a preset second potential risk index reference threshold value, the risk level of the potential abnormality of the detection precision is judged, and the comparison analysis result is as follows:
If the potential risk index is smaller than the first potential risk index reference threshold, classifying the risk level of the detection precision potential abnormality into a low risk potential abnormality;
If the potential risk index is greater than or equal to the first potential risk index reference threshold value and less than the second potential risk index reference threshold value, classifying the risk level of the detection precision potential abnormality as a general risk potential abnormality;
And if the potential risk index is greater than or equal to the second potential risk index reference threshold, classifying the risk level of the detection precision potential abnormality into a high risk potential abnormality.
For situations where the detection accuracy potential abnormal risk level is low, regular monitoring and preventive maintenance measures may be generally taken. While this risk level is low, it is still necessary to ensure equipment operational stability and detection accuracy. Periodic equipment calibration and checks may be scheduled to monitor changes in key parameters to ensure that any minor anomalies are discovered and handled initially. The operator should remain alert and record all monitoring data for later reference and analysis. The measures can effectively prevent the small problem from developing into a large fault, and ensure the long-term stable operation of the online color difference detector.
When the risk level of the detection accuracy potential abnormality is classified as a general risk, more aggressive countermeasures need to be taken. The method comprises the steps of performing detailed fault detection and diagnosis on the online color difference detector, and finding out specific reasons of potential abnormality. Comprehensive inspection and maintenance of the device, replacement of components that may lead to reduced detection accuracy, recalibration of the sensors and optical elements may be required. In addition, training of operators should be enhanced to better identify and handle anomalies. By means of the measures, the influence of potential abnormality of general risks on production quality can be effectively reduced, and the equipment is ensured to operate in an optimal state.
For the situation that the detection precision potential abnormal risk level is high risk, emergency measures need to be immediately taken. Firstly, the use of the on-line color difference detector should be stopped immediately, so as to prevent further influence on the production quality. Then, the calling professional technician performs emergency maintenance and deep fault analysis to find out and repair all possible abnormal reasons. This may include replacing critical components, upgrading a software system, or making a full device reset. In addition, the whole detection process should be re-evaluated to ensure that no potential problems are missed. After the equipment resumes normal operation, the monitoring frequency and the maintenance force should be increased to ensure that similar problems no longer occur. Such comprehensive and rigorous countermeasures may minimize the negative impact of high risk potential anomalies on production and quality.
According to the invention, the high-precision sensor and the data acquisition system are embedded, the operation parameters and the spectral reflectivity information of the online color difference detector are acquired in real time, the abnormal analysis and detection are carried out, the detection precision is monitored and evaluated in real time by using the machine learning model, the potential abnormality of the operation state of the detector can be identified in time, the high-efficiency detection signal or the potential abnormality signal is generated, the real-time monitoring and early warning mechanism can find problems before obvious color difference changes occur to textiles, the color inconsistency of large batches of products is avoided, and the economic loss of reworking or scrapping is reduced.
According to the method, after the potential abnormality is detected, the potential risk index is generated through multi-level threshold comparison and comprehensive analysis, the risk grades are divided, corresponding treatment measures are adopted according to different risk grades, the potential abnormality problem of the detection precision can be more accurately treated, especially for the potential abnormality with high risk, measures can be immediately adopted to correct, the product quality is ensured to meet the customer standard, and the consistency of the production efficiency and the product quality is improved.
The invention provides a digital textile equipment management system based on data analysis as shown in fig. 2, which comprises a data acquisition module, an anomaly analysis module, a machine learning monitoring module and a risk assessment and response module;
The data acquisition module is used for acquiring the operation parameter information of the online color difference detector in real time and storing the operation parameter information in a central database for subsequent analysis by embedding a high-precision sensor and a data acquisition system into the online color difference detector in the process of carrying out color detection on textiles by the online color difference detector;
the abnormality analysis module is used for carrying out abnormality analysis on the preprocessed operation parameter information, detecting potential abnormal modes in the operation parameters and identifying parameter changes of the online color difference detector, which are different from the normal operation state;
The machine learning monitoring module is used for training a machine learning model by using a supervised learning method, deploying the trained machine learning model in the online color difference detection system based on an abnormal analysis result, monitoring the online color difference detector in real time, and identifying potential abnormality existing in the detection precision of the online color difference detector;
The risk assessment and response module is used for further comprehensively analyzing the online color difference detector with the potential abnormality of the detection precision, judging the risk level of the potential abnormality of the detection precision and adopting different response measures for the potential abnormality of the detection precision of different risk levels;
the embodiment of the invention provides a method for managing digital textile equipment based on data analysis, which is realized by the digital textile equipment management system based on data analysis, and a specific method and a flow of the digital textile equipment management system based on data analysis are detailed in the embodiment of the digital textile equipment management method based on data analysis, and are not repeated herein.
The above formulas are all formulas with dimensions removed and numerical values calculated, the formulas are formulas with a large amount of data collected for software simulation to obtain the latest real situation, and preset parameters in the formulas are set by those skilled in the art according to the actual situation.
The above embodiments may be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented in software, the above-described embodiments may be implemented in whole or in part in the form of a computer program product. The computer program product comprises one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the processes or functions described in accordance with embodiments of the present application are produced in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from one website site, computer, server, or data center to another website site, computer, server, or data center by wired or wireless means (e.g., infrared, wireless, microwave, etc.). The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains one or more sets of available media. The usable medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium may be a solid state disk.
It should be understood that, in various embodiments of the present application, the sequence numbers of the foregoing processes do not mean the order of execution, and the order of execution of the processes should be determined by the functions and internal logic thereof, and should not constitute any limitation on the implementation process of the embodiments of the present application.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
The foregoing is merely illustrative of the present application, and the present application is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (8)
1. The digital textile equipment management method based on data analysis is characterized by comprising the following steps of:
In the process of carrying out color detection on textiles by the online color difference detector, the online color difference detector is embedded with a high-precision sensor and a data acquisition system, and the operation parameter information of the online color difference detector is acquired in real time and stored in a central database for subsequent analysis;
Performing exception analysis on the preprocessed operation parameter information, detecting a potential exception mode in the operation parameters, and identifying parameter changes of the online color difference detector, which are different from the normal operation state;
Training a machine learning model by using a supervised learning method, deploying the trained machine learning model in an online color difference detection system based on an abnormal analysis result, monitoring an online color difference detector in real time, and identifying potential abnormality existing in the detection precision of the online color difference detector;
further comprehensively analyzing the online color difference detector with the potential abnormality of the detection precision, judging the risk level of the potential abnormality of the detection precision, and taking different countermeasures for the potential abnormality of the detection precision of different risk levels;
the operation parameter information of the online color difference detector comprises spectral reflectivity information and electronic drift information, wherein the spectral reflectivity information refers to the proportion of reflected light of the textile surface under the irradiation of light with different wavelengths, and the electronic drift information refers to the output signal deviation condition caused by the change of the performance of electronic elements in the online color difference detector along with time;
After spectral reflectance information and electronic drift information of the online color difference detector during color detection of textiles are obtained through a high-precision spectral analysis technology, abnormal analysis processing is carried out on the spectral reflectance information, a spectral reflectance abnormal index is generated, abnormal analysis processing is carried out on the electronic drift information, an electronic drift index is generated, and parameter changes of the online color difference detector, which are different from normal operation states, are identified through the spectral reflectance abnormal index and the electronic drift index.
2. The method for managing digital textile equipment based on data analysis according to claim 1, wherein after obtaining the abnormal spectral reflectance index and the electronic drift index generated when the online color difference detector detects the color of the textile by a high-precision spectral analysis technology, the abnormal spectral reflectance index and the electronic drift index are input into a pre-trained machine learning model, a detection precision coefficient is generated by the machine learning model, and the precision of the online color difference detector during textile color detection is evaluated by the detection precision coefficient.
3. The method for managing digital textile equipment based on data analysis according to claim 2, wherein the detection precision coefficient generated in the fixed detection window when the online color difference detector detects the textile color is compared with a preset detection precision coefficient reference threshold value for analysis, and potential anomalies of the detection precision of the online color difference detector are identified, and the comparison analysis results are as follows:
If the detection precision coefficient is greater than or equal to the detection precision coefficient reference threshold, generating a high-efficiency detection signal, wherein the high-efficiency detection signal indicates that the high-efficiency detection can be realized when the online color difference detector detects the color of the textile;
if the detection precision coefficient is smaller than the detection precision coefficient reference threshold, generating a detection potential abnormal signal, which indicates that the online color difference detector has potential detection abnormality when detecting the color of the textile.
4. The method for managing digital textile equipment based on data analysis according to claim 3, wherein when a detection potential abnormal signal is generated in a fixed detection window when an online color difference detector detects textile color, an analysis set is established by continuously acquiring a plurality of detection precision coefficients generated in the fixed detection window when the online color difference detector detects textile color, the detection precision coefficients in the analysis set are compared with a first gradient reference threshold value and a second gradient reference threshold value, wherein the second gradient reference threshold value is smaller than the first gradient reference threshold value, the detection precision coefficients are compared with the second gradient reference threshold value, the first gradient reference threshold value and the detection precision coefficient reference threshold value, the number of the detection precision coefficients which are larger than or equal to the first gradient reference threshold value and smaller than the detection precision coefficient reference threshold value is denoted as Ka, the number of the detection precision coefficients which are larger than or equal to the second gradient reference threshold value and smaller than the first gradient reference threshold value is denoted as Kb, and the number of the detection precision coefficients which are smaller than the second gradient reference threshold value is denoted as Kc;
Comprehensively analyzing Ka, kb and Kc to generate Potential risk index Potential Risk according to the following formula: where u 1、u2、u3 is the weight coefficient of Ka, kb, kc, respectively, and 0<u 1<u2<u3.
5. The method for managing digital textile equipment based on data analysis according to claim 4, wherein the risk indexes generated by further comprehensive analysis of the online color difference detector with the potential abnormality of detection precision are compared with a preset first potential risk index reference threshold value and a preset second potential risk index reference threshold value, and the risk level of the potential abnormality of detection precision is judged, and the comparison analysis results are as follows:
If the potential risk index is smaller than the first potential risk index reference threshold, classifying the risk level of the detection precision potential abnormality into a low risk potential abnormality;
If the potential risk index is greater than or equal to the first potential risk index reference threshold value and less than the second potential risk index reference threshold value, classifying the risk level of the detection precision potential abnormality as a general risk potential abnormality;
And if the potential risk index is greater than or equal to the second potential risk index reference threshold, classifying the risk level of the detection precision potential abnormality into a high risk potential abnormality.
6. The method for managing digital textile equipment based on data analysis according to claim 1, wherein the specific step of generating the abnormal index of spectral reflectance after performing the abnormal analysis processing on the information of spectral reflectance is as follows:
Collecting spectral reflectance data of a plurality of wavelengths in a fixed detection window, and setting the collected spectral reflectance data as R (lambda i), wherein lambda i represents the ith wavelength, i=1, 2, N is the number of the collected wavelengths;
Preprocessing the collected spectral reflectance data, and recording the preprocessed spectral reflectance data as
The rate of change of spectral reflectance with wavelength is calculated, and the expression is calculated as follows: in the method, in the process of the invention, Represents the derivative of the spectral reflectance with respect to wavelength, ΔR (λ i) represents the rate of change of the spectral reflectance;
Calculating a second derivative of the rate of change to capture the rate of change acceleration of the reflectance curve, the calculated expression being: Where Δ 2R(λi) represents the second derivative of the spectral reflectance change rate;
Defining a comprehensive anomaly detection function A (lambda i), combining the change rate of the spectral reflectivity and the second derivative of the change rate of the spectral reflectivity to capture anomalies, wherein the expression of the comprehensive anomaly detection function is as follows: a (lambda i)＝|ΔR(λi)α·|Δ2R(λi)|β, wherein a (lambda i) represents a comprehensive anomaly detection function, and alpha and beta are respectively the change rate delta R (lambda i) of the spectral reflectivity and the regulation parameter of the second derivative delta 2R(λi of the spectral reflectivity, and the influence weights of the change rate delta R (lambda i) of the spectral reflectivity and the second derivative delta 2R(λi of the spectral reflectivity are controlled;
In the fixed detection window, the abnormal detection function values of all wavelengths are accumulated to generate a spectral reflectance abnormal index, and the generated expression is: Wherein SP spc represents the spectral reflectance abnormality index.
7. The method for managing digital textile equipment based on data analysis according to claim 1, wherein the specific step of generating the electronic drift index after performing the anomaly analysis processing on the electronic drift information is:
In a fixed detection window, acquiring reference signal data of an electronic element in real time, wherein the reference signal data comprises a voltage signal V, a current signal I, a temperature T and a time T;
preprocessing the acquired signal data;
Extracting characteristic parameters from the preprocessed data, wherein the characteristic parameters comprise instantaneous voltage change rate, instantaneous current change rate, temperature change rate and time interval, and the extracted expression is as follows: in the method, in the process of the invention, And Δt represents an instantaneous voltage change rate, an instantaneous current change rate, a temperature change rate, and a time interval, respectively;
The extracted characteristic parameters are subjected to multidimensional characteristic fusion to construct a multidimensional characteristic vector F, the multidimensional characteristic vector F comprises instantaneous voltage change rate, instantaneous current change rate, temperature change rate and time interval multidimensional characteristics,
The weight matrix W is calculated through a weighted least square method and used for weight distribution of multidimensional feature fusion, and the calculation expression of the weight matrix W is as follows: w= (X TX)-1XT Y, where X is the feature matrix and Y is the historical drift index vector;
The multidimensional feature vector F and the weight matrix W are linearly combined, an electronic drift index is calculated, the electronic drift index reflects the comprehensive degree of electronic element drift, and the calculated expression is: elec Drift＝F·WT, where W T represents the transpose of the weight matrix W and Elec Drift represents the electron drift index.
8. A digital textile equipment management system based on data analysis, for implementing a digital textile equipment management method based on data analysis as claimed in any one of claims 1 to 7, characterized by comprising a data acquisition module, an anomaly analysis module, a machine learning monitoring module and a risk assessment and response module;
The data acquisition module is used for acquiring the operation parameter information of the online color difference detector in real time and storing the operation parameter information in a central database for subsequent analysis by embedding a high-precision sensor and a data acquisition system into the online color difference detector in the process of carrying out color detection on textiles by the online color difference detector;
the abnormality analysis module is used for carrying out abnormality analysis on the preprocessed operation parameter information, detecting potential abnormal modes in the operation parameters and identifying parameter changes of the online color difference detector, which are different from the normal operation state;
The machine learning monitoring module is used for training a machine learning model by using a supervised learning method, deploying the trained machine learning model in the online color difference detection system based on an abnormal analysis result, monitoring the online color difference detector in real time, and identifying potential abnormality existing in the detection precision of the online color difference detector;
and the risk assessment and response module is used for further comprehensively analyzing the online color difference detector with the potential abnormality of the detection precision, judging the risk level of the potential abnormality of the detection precision and taking different response measures for the potential abnormality of the detection precision of different risk levels.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410933082.9A CN118482817B (en) | 2024-07-12 | 2024-07-12 | Digital textile equipment management method and system based on data analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410933082.9A CN118482817B (en) | 2024-07-12 | 2024-07-12 | Digital textile equipment management method and system based on data analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118482817A CN118482817A (en) | 2024-08-13 |
| CN118482817B true CN118482817B (en) | 2024-09-03 |
Family
ID=92198393
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410933082.9A Active CN118482817B (en) | 2024-07-12 | 2024-07-12 | Digital textile equipment management method and system based on data analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118482817B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107766603A (en) * | 2017-04-24 | 2018-03-06 | 东华大学 | A kind of colour-spun yarns computer is measured color method |
| CN114486768A (en) * | 2021-12-24 | 2022-05-13 | 石河子大学 | Seed cotton color grade detection device for seed cotton purchasing link and grading method thereof |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111724372A (en) * | 2020-06-19 | 2020-09-29 | 深圳新视智科技术有限公司 | Method, terminal and storage medium for detecting cloth defects based on antagonistic neural network |
| CN117557166B (en) * | 2024-01-02 | 2024-05-07 | 福建省政务门户网站运营管理有限公司 | Multi-mode and real-time data user data environment intelligent monitoring system |
| CN118034245A (en) * | 2024-02-08 | 2024-05-14 | 浙江佳良智动科技有限公司 | Multifunctional portable engine state analyzer |
-
2024
- 2024-07-12 CN CN202410933082.9A patent/CN118482817B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107766603A (en) * | 2017-04-24 | 2018-03-06 | 东华大学 | A kind of colour-spun yarns computer is measured color method |
| CN114486768A (en) * | 2021-12-24 | 2022-05-13 | 石河子大学 | Seed cotton color grade detection device for seed cotton purchasing link and grading method thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118482817A (en) | 2024-08-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10739752B2 (en) | Computer system and method for causality analysis using hybrid first-principles and inferential model | |
| US7275018B2 (en) | Method of condition monitoring | |
| CN116757534B (en) | Intelligent refrigerator reliability analysis method based on neural training network | |
| JP5735499B2 (en) | Method and apparatus for automatically detecting and classifying fault conditions during substrate processing | |
| JP4911055B2 (en) | Batch process data analysis apparatus and abnormality detection / quality estimation apparatus using the same | |
| CN112284440B (en) | Sensor data deviation self-adaptive correction method | |
| CN118154174B (en) | Intelligent operation and maintenance cloud platform for industrial equipment | |
| KR102408426B1 (en) | Method for detecting anomaly using equipment age index and apparatus thereof | |
| CN112200327B (en) | MES equipment maintenance early warning method and system | |
| CN109523171B (en) | SVDD-based gas turbine air inlet system health degree evaluation method | |
| CN117193222A (en) | Intelligent quality control system based on industrial Internet of things and big data and control method thereof | |
| WO2020166236A1 (en) | Work efficiency evaluating method, work efficiency evaluating device, and program | |
| CN118243737B (en) | Method, device, equipment and storage medium for detecting concentration of negative ions | |
| JP2001526418A (en) | Processing unit monitoring method | |
| JP7248188B2 (en) | Abnormality diagnosis model construction method, abnormality diagnosis method, abnormality diagnosis model construction device, and abnormality diagnosis device | |
| CN110751217A (en) | Equipment energy consumption ratio early warning analysis method based on principal component analysis | |
| CN118482817B (en) | Digital textile equipment management method and system based on data analysis | |
| CN118014115A (en) | Production line loss monitoring method based on long-short-term memory network | |
| CN117977536A (en) | Smart power grid load identification method | |
| Ison et al. | Robust fault detection and fault classification of semiconductor manufacturing equipment | |
| CN113780852A (en) | Diagnosis method for quality defects in plate and strip rolling process | |
| CN118521222B (en) | Liquid crystal glass thickness online detection method, system, equipment and medium | |
| CN117871771B (en) | Big data-based gas energy monitoring method | |
| CN117836734A (en) | Comprehensive analysis module for determining performance of processing equipment | |
| CN118521222A (en) | Liquid crystal glass thickness online detection method, system, equipment and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |