Paper flatness detection system and method and storage medium

By real-time detection of multi-source data on paper surface and comprehensive analysis using neural network models, the problems of unstable and poor real-time performance of paper flatness detection results in the existing technology are solved, and more accurate and timely monitoring of the production process is achieved, ensuring that Paper quality.

CN118980339AInactive Publication Date: 2024-11-19CHANGZHOU WEIYIMEIJIA NEW MATERIALS CO LTD
View PDF 9 Cites 0 Cited by

Patent Information

Application Number
CN202411286601.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-13
Publication Date
2024-11-19
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing paper flatness detection methods rely on a single data point and are susceptible to environmental factors and errors in testing equipment, resulting in unstable detection results and poor real-time performance, which cannot promptly reflect changes in the production process.

Method used

By detecting the finish and related production data of the paper surface in real time during paper transfer, such as movement speed, tension, temperature and humidity, using multi-source data for comprehensive analysis, drawing the electrical signal change curve, and combining with neural network models for judgment and correction, Transfer and process results in real time.

Benefits of technology

It improves the accuracy and reliability of paper flatness detection, realizes real-time monitoring of the production process, timely discovery and adjustment, and ensures production quality.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118980339A_ABST
    Figure CN118980339A_ABST
Patent Text Reader

Abstract

The invention belongs to the technical field of data communication, and particularly relates to a paper flatness detection system and method and a storage medium, and the system comprises a detection data recognition module which recognizes the detection data of the paper flatness; the data transmission module is used for respectively transmitting the detection data and other set data through two transmission paths; the signal conversion module performs electric signal conversion on the transmission data to obtain a first electric signal and a second electric signal; the control module is used for drawing change curves and judging the flatness of the paper according to the relative relation of the two curves; and the communication module is used for sending a judgment result of the control module to a preset server. Multi-source data is utilized for comprehensive analysis in the data communication process, the flatness of the paper surface can be judged more accurately, the method is different from a traditional method which only depends on flatness detection data, the precision and reliability of a detection result are improved, and in the data transmission process, the data transmission efficiency is improved through real-time transmission and processing. The system can find and feed back the flatness problem in production in time.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of data communication, and in particular relates to a paper flatness detection system, method and storage medium. Background Art

[0002] In the production process of various types of paper, the flatness of the paper surface is one of the important factors affecting the quality of paper. Paper with poor flatness may cause a series of problems such as poor printing effects and copy jams during subsequent processing and use. Therefore, how to effectively detect and control the flatness of the paper surface is the focus of attention of relevant paper production companies.

[0003] At present, the detection of paper flatness mainly relies on the traditional method of analyzing flatness detection data. Usually, sensors or other detection equipment are used to directly obtain the flatness detection data of the paper surface, and then these data are processed and analyzed through statistical analysis methods, such as average value, standard deviation, etc. However, this traditional detection method has the following main problems: Existing technologies usually rely only on single flatness test data and lack a comparison benchmark for other production data. This single data point analysis method is easily affected by various environmental factors and detection equipment errors, resulting in unstable test results. In the actual production process, various production parameters such as temperature, humidity, and production speed will affect the flatness of the paper. However, existing technologies often fail to fully consider the impact of these factors, resulting in large fluctuations in test data and affecting the accuracy of test results. Traditional detection methods require data collection first, and then offline analysis and processing. It is difficult to reflect changes in the production process in a timely manner, and the real-time performance is poor. As a result, when flatness problems are discovered, they often have affected large batches of products and cannot be adjusted and corrected in a timely manner. Summary of the invention

[0004] The present invention provides a paper flatness detection system, method and storage medium, which can effectively solve the problems in the background technology.

[0005] To achieve the above purpose, the technical solution of the present invention is as follows: The paper flatness detection system detects the flatness of the paper during its transportation, including: A detection data identification module, identifying a data combination in the production data of the paper, each of the data combinations includes detection data of the flatness of the paper and other related set data within a set time range, and different data combinations are arranged in chronological order; A data transmission module, transmitting the identified detection data and other setting data through a first transmission path and a second transmission path respectively; A signal conversion module, which converts the transmission data of the first transmission path and the second transmission path into electrical signals respectively, to obtain a first electrical signal and a second electrical signal; A control module, plotting a curve of the first electrical signal and the second electrical signal changing with time, and judging the flatness of the paper according to the relative relationship between the two curves; The communication module sends the judgment result of the control module to a preset server.

[0006] Furthermore, the detection data is a data string of the surface finish of the paper, which is detected in real time by a fixed sensor matrix during the paper conveying process, and the sensor is a photoelectric sensor or a laser sensor; The sensor emits a light beam or a laser beam to illuminate the surface of the paper, and measures the surface finish of the paper according to the intensity and / or scattering angle of the reflected light; The other setting data is a data string formatted into a unified data structure, including the moving speed, tension, temperature and humidity of the paper detected in real time.

[0007] Furthermore, the signal conversion module includes: a data receiving unit, each provided with an independent input port, for receiving data transmitted through the first transmission path and the second transmission path; The analog-to-digital conversion (ADC) unit processes the received analog signal and converts it into a digital signal; The digital-to-analog conversion DAC unit processes the received digital signal and converts it into an analog electrical signal; The signal modulation unit modulates the analog electrical signal to meet predetermined signal standards and characteristics.

[0008] Furthermore, the control module includes: An input interface unit, receiving the first electrical signal and the second electrical signal; a curve drawing unit for drawing a first variation curve and a second variation curve respectively according to the variation of the first electrical signal and the second electrical signal over time; The judgment logic unit executes a set logic judgment algorithm on the first change curve and the second change curve to judge the surface flatness of the paper.

[0009] Furthermore, the logic judgment algorithm includes: Acquire values ​​of the first change curve and the second change curve at the same time point to form a data pair; Inputting the data pair into a first neural network model to identify abnormal situations; The surface flatness of the paper is determined based on the recognition result.

[0010] Furthermore, it also includes an abnormal revision module, which corrects the other setting data before the data transmission module transmits it, including: an abnormal data identification unit, which continuously predicts the other setting data through a second neural network model to obtain a current prediction value, compares the current other setting data with the current prediction value, and determines whether the current other setting data is abnormal data; The data correction unit corrects the abnormal data according to a set rule.

[0011] Furthermore, the abnormality revision module also includes: The abnormal alarm unit sets an alarm threshold, and outputs an alarm message when the degree and / or frequency of occurrence of the abnormal data exceeds the alarm threshold.

[0012] Furthermore, the setting rules are: Replacing the abnormal data with the mean of a number of normal data; Alternatively, the abnormal data is replaced by the current predicted value.

[0013] The paper flatness detection method performs flatness detection during the paper conveying process, including: Acquire paper production data, identify data combinations therefrom, and arrange different data combinations in chronological order, wherein each data combination includes detection data on paper flatness and other related setting data within a set time range; Transmitting the identified detection data and other setting data through a first transmission path and a second transmission path respectively; Respectively converting the transmission data of the first transmission path and the second transmission path into electrical signals to obtain a first electrical signal and a second electrical signal; Plotting a curve of the first electrical signal and the second electrical signal changing with time, and judging the flatness of the paper according to the relative relationship between the two curves; The determination result of the control module is sent to a preset server.

[0014] A storage medium stores a paper flatness detection program. When the paper flatness detection program is executed by a processor, the paper flatness detection method as described above is implemented.

[0015] The technical solution of the present invention can achieve the following technical effects: The present invention utilizes multi-source data for comprehensive analysis during the data communication process. Based on the relative relationship between the two curves, the system can more accurately judge the flatness of the paper surface. Unlike the traditional method that only relies on flatness detection data, the accuracy and reliability of the detection results are improved, and the factors that cause flatness changes in the production process can be more comprehensively analyzed; during the data transmission process, through real-time transmission and processing, the system can promptly discover and feedback flatness problems in production, and realize real-time monitoring of the production process. The communication module sends the judgment results of the control module to the preset server in real time, so that production management personnel can monitor the changes in paper flatness in real time, make timely adjustments, and ensure production quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 This is the framework diagram of the paper flatness detection system; Figure 2 It is a schematic diagram of the composition of data combination; Figure 3 This is the framework diagram of the control module; Figure 4 It is a flowchart of the logic judgment algorithm; Figure 5 This is a flow chart of paper flatness detection. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0020] Embodiment 1 The paper flatness detection system detects the flatness of the paper during its transmission. Figure 1 As shown, including: The detection data identification module identifies data combinations in the production data of the paper. Each data combination is included in the set time range. For the detection data of the flatness of the paper and other related set data, different data combinations are arranged in chronological order; the data in the same data combination may be acquired based on the same time point, or there may be differences in the acquisition time; The data transmission module transmits the identified detection data and other set data through the first transmission path and the second transmission path respectively; this separate transmission method can reduce interference and noise in the data transmission process and ensure the integrity and accuracy of the data; before transmission, each group of data transmitted through different paths can be stamped with the same timestamp, and the timestamp can ensure that the data is arranged and processed in the order of generation time at the receiving end, thereby ensuring synchronization, and the assignment of timestamps can be completed in the detection data identification module; based on the assignment of timestamps, different priorities can also be assigned to the first transmission path and the second transmission path. For example, the flatness detection data is marked as high priority, and the high-priority data transmission channel can preferentially obtain better network resources and bandwidth allocation, thereby reducing the data packet loss rate and transmission error rate; A signal conversion module, which converts the transmission data of the first transmission path and the second transmission path into electrical signals respectively, to obtain a first electrical signal and a second electrical signal; A control module, plotting a curve of the first electrical signal and the second electrical signal changing with time, and judging the flatness of the paper according to the relative relationship between the two curves; The communication module sends the judgment result of the control module to the preset server.

[0021] In the present invention, the flatness detection data and the corresponding other setting data are transmitted through independent transmission paths respectively through the detection data identification module and the data transmission module. After the signal conversion module converts the data into an electrical signal, the control module draws two change curves. Based on the relative relationship of these curves, the system can more accurately judge the flatness of the paper surface. Different from the traditional method that only relies on the flatness detection data, the present invention uses multi-source data for comprehensive analysis, improves the accuracy and reliability of the detection results, and can more comprehensively analyze the factors that cause flatness changes in the production process.

[0022] During the data transmission process of the present invention, through real-time transmission and processing, the system can timely discover and feedback the flatness problems in production, realize real-time monitoring of the production process, and the communication module sends the judgment results of the control module to the preset server in real time, so that the production management personnel can monitor the changes in the flatness of the paper in real time, make timely adjustments, and ensure the production quality.

[0023] As a preferred embodiment of the above, Figure 2As shown, the detection data is a data string of the surface finish of the paper. During the paper transmission process, the detection is carried out in real time through a fixed sensor matrix. Figure 2 In the paper graphic in the upper left corner, each dot corresponds to a sensor in the sensor matrix, where the rectangular range outside the dots is the detection range of the sensor, and the sensor is a photoelectric sensor or a laser sensor; the sensor emits a light beam or a laser beam, which illuminates the surface of the paper, and measures the surface smoothness of the paper according to the intensity and / or scattering angle of the reflected light; other set data are data strings formatted into a unified data structure, including real-time detection of the movement speed, tension, temperature and humidity of the paper.

[0024] In this preferred embodiment, the multiple data obtained by the sensor matrix detection can be arranged according to the different positions of each sensor to obtain a surface finish data string. Similarly, other set data can also be arranged in sequence to obtain a data string after the data structure is unified.

[0025] In this embodiment, the moving speed of the paper on the production line directly affects the production efficiency and quality. The change in speed may cause uneven distribution of the surface tension of the paper, thereby affecting the surface flatness. Monitoring the moving speed can help identify quality problems caused by speed changes and make timely adjustments. As for the tension problem of the paper, excessive tension may cause the paper to break or deform, and insufficient tension may cause wrinkles or uneven surfaces. Different tension levels will cause changes in the internal stress of the paper, thereby affecting the flatness of the final product. By monitoring the tension, the production parameters can be optimized to ensure the uniformity and consistency of the paper surface. Temperature changes will cause thermal expansion or contraction of the paper material, affecting its size and surface state. The temperature changes of the equipment on the production line, such as dryers and heaters, will also affect the dryness and surface finish of the paper. Monitoring the temperature helps stabilize the production conditions and reduce quality fluctuations. As for the humidity problem, paper is a highly hygroscopic material, and its water content will change with the change of environmental humidity. Humidity changes may cause hygroscopic expansion or drying shrinkage of the paper, affecting the flatness and dimensional stability. Appropriate humidity control can reduce static electricity problems in paper production, avoid paper adhesion and accumulation, and ensure a smooth production process.

[0026] The above parameters jointly affect the flatness and quality of paper production. By monitoring and analyzing these parameters, we can more comprehensively understand and control various variables in production, thereby improving the quality and consistency of the final product. This multi-dimensional data selection and analysis method not only improves the accuracy of detection, but also provides strong support for the optimization and improvement of the production process. The acquisition or collection of the above parameters can be achieved through existing technologies and will not be repeated here.

[0027] In the above embodiment, surface finish, moving speed, tension, temperature and humidity are usually analog signals. As a preferred embodiment of this embodiment, the signal conversion module includes: The data receiving unit is provided with independent input ports respectively to receive data transmitted through the first transmission path and the second transmission path; the analog-to-digital conversion ADC unit processes the received analog signal and converts it into a digital signal; the digital-to-analog conversion DAC unit processes the received digital signal and converts it into an analog electrical signal; the signal modulation unit modulates the analog electrical signal to meet the predetermined signal standard and characteristics.

[0028] Digital signal processing can bring higher accuracy, reliability and flexibility. Digital signals are easy to process, store and transmit, and are easier to integrate with digital control systems. In the process of data transmission, by converting digital signals into electrical signals and modulating them, the signals can be made to meet specific transmission standards and characteristics. For example, the frequency, amplitude or phase of the signal can be changed by modulation to adapt to different transmission media (such as cables, radio waves, optical fibers, etc.) and communication standards, which helps to reduce interference and noise during transmission, improve the transmission quality and reliability of signals, optimize bandwidth utilization, and transmit more data within limited bandwidth. After being converted into electrical signals during implementation, amplifiers can be used to amplify the signals to meet the needs of long-distance transmission.

[0029] As a preferred embodiment of the above, Figure 3 As shown, the control module includes: An input interface unit receives a first electrical signal and a second electrical signal; a curve drawing unit draws a first change curve and a second change curve respectively according to changes of the first electrical signal and the second electrical signal over time; and a judgment logic unit executes a set logic judgment algorithm on the first change curve and the second change curve to judge the surface flatness of the paper.

[0030] The above curve can help identify the changing trends and patterns of the surface smoothness of paper, thereby providing more detailed analysis results. This graphical representation makes the identification and resolution of problems more intuitive. After the change curve is generated, these curves are analyzed using the set logical judgment algorithm. This method can make more refined judgments and analyses of the data, reduce misjudgments caused by fluctuations in analog signals, and improve the accuracy of judging the surface smoothness of paper.

[0031] In an actual production environment, the two data strings are relatively stable under normal circumstances, which means that these signals will not fluctuate dramatically within a certain period of time. Therefore, the relative relationship between the two curves is often stable. When special circumstances occur, the change in the surface flatness of the paper can be determined through a logical judgment algorithm.

[0032] As a preferred embodiment of the above, Figure 4 As shown, the logic judgment algorithm includes: S1: Acquire the values ​​of the first change curve and the second change curve at the same time point to form a data pair; S2: Input the data pair into the neural network model to identify abnormal situations; S3: Determine the surface flatness of the paper according to the recognition result.

[0033] The most prominent technical advantages of the above embodiment are as follows: the setting data such as the paper moving speed, tension, temperature and humidity are integrated into a common dimension, and their comprehensive influence is represented by a curve. This method does not focus on the independent effect of a single setting data, but analyzes the overall influence of its combination on the finish, which is particularly effective for complex production environments. When dealing with a large number of variables, data pair analysis helps to reduce redundant information and noise in the analysis, making the model easier to understand and interpret. After simplifying the setting data into one dimension, the dimensionality of the data can be reduced, and the calculation complexity and data management difficulty can be reduced.

[0034] Through two-dimensional data pairs, the patterns and relationships between the two curves can be quickly identified. This simplified analysis method can significantly improve the efficiency of problem detection and resolution, and can reduce computing resource requirements and data storage space. It is especially important in real-time analysis and large-scale data processing scenarios. By focusing on the most important variables, analysis resources can be allocated more efficiently, improving the response speed and flexibility of the entire system. Data pair analysis can help isolate and reduce the impact of irrelevant factors and improve the robustness and reliability of the model.

[0035] In specific implementation, the following method can be adopted: the input layer of the neural network model receives the two numerical values ​​of the data pair respectively, the hidden layer performs feature extraction and pattern recognition, and the output layer uses the sigmoid activation function for binary classification. After model training and evaluation, it can be used in work scenarios.

[0036] As a preferred embodiment of the above embodiment, it further includes an abnormality correction module, which corrects other setting data before the data transmission module transmits the data, including: The abnormal data identification unit continuously predicts other set data through the second neural network model to obtain the current prediction value, compares the current other set data with the current prediction value, and determines whether the current other set data is abnormal data; the data correction unit corrects the abnormal data according to the set rules.

[0037] Correcting abnormal data through the above-mentioned abnormal revision module can effectively eliminate errors caused by sensor drift, noise interference or data transmission errors, ensure that the data transmitted to the control system is accurate and reliable, and directly improve the system's control ability over the production process; by timely correcting abnormal data, the production process can be maintained under stable operating conditions, reduce fluctuations and uncertainties, prevent chain reactions caused by abnormal data, and reduce the risk of equipment failure or production interruption.

[0038] During the implementation process, the automated correction mechanism reduces the need for manual monitoring and intervention, reduces labor costs and human errors, and reduces quality problems and production interruptions. The correction mechanism can save time and production costs. Of course, during the implementation process, the corrected data often comes from sudden situations, which are often unexpected, without obvious precursors, and difficult to predict in advance through conventional monitoring methods. Sudden events usually change rapidly in a short period of time, which may cause production parameters to deviate rapidly from normal values. In the present application, when the above-mentioned sudden situation can be resolved by itself in a sufficiently short time, the correction is meaningful, and continuous production is guaranteed while avoiding human intervention. When the sudden situation lasts for a long time, further processing is required. Therefore, as a preferred embodiment of the above embodiment, the abnormal revision module also includes an abnormal alarm unit, which sets an alarm threshold, and outputs an alarm message when the degree and / or frequency of occurrence of abnormal data exceeds the alarm threshold.

[0039] When an emergency situation resolves itself in a sufficiently short time, such as a minor sensor failure or environmental fluctuation, in this case, automated correction can effectively respond to short-term fluctuations and ensure continuous production. When an emergency situation lasts for a long time, the constant alarm unit can identify persistent anomalies and remind operators to take appropriate measures. Setting reasonable alarm thresholds based on production conditions can ensure that the system can promptly identify the degree and frequency of abnormalities.

[0040] For the abnormal revision module, the automated correction mechanism and the abnormal alarm unit jointly construct an intelligent production control system that can effectively respond to emergencies and ensure the continuity and quality of production. This comprehensive approach not only improves the level of production automation.

[0041] As a preference of the above embodiment, the setting rule is: Replace the abnormal data with the mean of several normal data. The mean of normal data can be calculated from historical data or normal data in the current window period. This method can smooth out the anomalies and maintain data stability.

[0042] Alternatively, the abnormal data can be replaced with the current predicted value. The second neural network model can be trained using historical normal data and current production data to predict the expected value of the set data in real time. Continuous prediction means that the model will be continuously updated to adapt to current production conditions. The abnormal data can be directly replaced with the predicted value of the neural network model. This method can quickly adjust the abnormal data to the expected level and has strong adaptability.

[0043] In the above embodiment, the second neural network model can be selected in a specific model form as needed, such as a multi-layer perceptron, which is composed of an input layer, a hidden layer and an output layer. The input features may include various setting parameters related to paper production, such as the moving speed, tension, temperature and humidity of the paper, which are usually input into the model in the form of a multi-dimensional vector. Before input into the neural network, the data may be pre-processed, such as normalized or standardized, to improve the training effect and prediction accuracy of the model; The hidden layer extracts and transforms the input data through a series of neurons and nonlinear activation functions (such as ReLU, sigmoid, tanh). The output layer usually contains one or more neurons, which output prediction results according to task requirements. When each set parameter needs to be predicted separately, an output node can be set for each parameter. If the prediction results of multiple set parameters are output as a whole, a multi-output node structure can be designed.

[0044] Embodiment 2 The paper flatness detection method performs flatness detection during the paper conveying process, such as Figure 5 As shown, including: A1: Acquire the production data of paper, identify data combinations therefrom, and arrange different data combinations in chronological order, wherein each data combination includes the detection data of the flatness of paper and other related setting data within a set time range; A2: transmitting the identified detection data and other setting data through the first transmission path and the second transmission path respectively; A3: Convert the transmission data of the first transmission path and the second transmission path into electrical signals respectively to obtain a first electrical signal and a second electrical signal; A4: Draw the time-varying curves of the first electrical signal and the second electrical signal, and judge the flatness of the paper according to the relative relationship between the two curves; A5: Send the judgment result of the control module to the preset server.

[0045] Embodiment 3 A storage medium stores a paper flatness detection program. When the paper flatness detection program is executed by a processor, the paper flatness detection method described in the second embodiment is implemented.

[0046] The technical effects achieved by Embodiment 2 and Embodiment 3 are as described in Embodiment 1 and will not be described again here.

[0047] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. Paper flatness detection system, which detects the flatness of the paper during its transmission. It is characterized in that include: A detection data identification module, identifying a data combination in the production data of the paper, each of the data combinations includes detection data of the flatness of the paper and other related set data within a set time range, and different data combinations are arranged in chronological order; A data transmission module, transmitting the identified detection data and other setting data through a first transmission path and a second transmission path respectively; A signal conversion module, which converts the transmission data of the first transmission path and the second transmission path into electrical signals respectively, to obtain a first electrical signal and a second electrical signal; A control module, plotting a curve of the first electrical signal and the second electrical signal changing with time, and judging the flatness of the paper according to the relative relationship between the two curves; The communication module sends the judgment result of the control module to a preset server.

2. The paper flatness detection system according to claim 1, It is characterized in that The detection data is a data string of the surface finish of the paper, which is detected in real time by a fixed sensor matrix during the paper transmission process, and the sensor is a photoelectric sensor or a laser sensor; The sensor emits a light beam or a laser beam to illuminate the surface of the paper, and measures the surface finish of the paper according to the intensity and / or scattering angle of the reflected light; The other setting data is a data string formatted into a unified data structure, including the moving speed, tension, temperature and humidity of the paper detected in real time.

3. The paper flatness detection system according to claim 2, It is characterized in that The signal conversion module comprises: a data receiving unit, each provided with an independent input port, for receiving data transmitted through the first transmission path and the second transmission path; The analog-to-digital conversion (ADC) unit processes the received analog signal and converts it into a digital signal; The digital-to-analog conversion DAC unit processes the received digital signal and converts it into an analog electrical signal; The signal modulation unit modulates the analog electrical signal to meet predetermined signal standards and characteristics.

4. The paper flatness detection system according to claim 3, It is characterized in that The control module comprises: An input interface unit, receiving the first electrical signal and the second electrical signal; a curve drawing unit for drawing a first variation curve and a second variation curve respectively according to the variation of the first electrical signal and the second electrical signal over time; The judgment logic unit executes a set logic judgment algorithm on the first change curve and the second change curve to judge the surface flatness of the paper.

5. The paper flatness detection system according to claim 4, It is characterized in that The logic judgment algorithm includes: Acquire values ​​of the first change curve and the second change curve at the same time point to form a data pair; Inputting the data pair into a first neural network model to identify abnormal situations; The surface flatness of the paper is determined based on the recognition result.

6. The paper flatness detection system according to claim 1, It is characterized in that It also includes an abnormal revision module, which corrects the other setting data before the data transmission module transmits the data, including: an abnormal data identification unit, which continuously predicts the other setting data through a second neural network model to obtain a current prediction value, compares the current other setting data with the current prediction value, and determines whether the current other setting data is abnormal data; The data correction unit corrects the abnormal data according to a set rule.

7. The paper flatness detection system according to claim 6, It is characterized in that The anomaly revision module also includes The abnormal alarm unit sets an alarm threshold, and outputs an alarm message when the degree and / or frequency of occurrence of the abnormal data exceeds the alarm threshold.

8. The paper flatness detection system according to claim 6, It is characterized in that The setting rules are: Replacing the abnormal data with the mean of a number of normal data; Alternatively, the abnormal data is replaced by the current predicted value.

9. Paper flatness detection method: perform flatness detection during the paper conveying process. It is characterized in that include: Acquire paper production data, identify data combinations therefrom, and arrange different data combinations in chronological order, wherein each data combination includes detection data on paper flatness and other related setting data within a set time range; Transmitting the identified detection data and other setting data through a first transmission path and a second transmission path respectively; Respectively converting the transmission data of the first transmission path and the second transmission path into electrical signals to obtain a first electrical signal and a second electrical signal; Plotting a curve of the first electrical signal and the second electrical signal changing with time, and judging the flatness of the paper according to the relative relationship between the two curves; The determination result of the control module is sent to a preset server.

10. A storage medium, It is characterized in that The storage medium stores a paper flatness detection program, and when the paper flatness detection program is executed by the processor, the paper flatness detection method according to claim 9 is implemented.

Citation Information

Patent Citations

  • Sensor system and method for determining paper sheet quality parameters

    CA2846816A1

  • Paper product detection management system

    CN111999483A

  • Corrugated board production monitoring method and system, storage medium and intelligent terminal

    CN114838698A

  • Data anomaly diagnosis method based on neural network and electronic equipment

    CN116702005A

  • Quality management system for integrated decoration plate finished product processing

    CN118037126A