System and method for monitoring operation state of plate facing equipment

Abstract

The invention discloses an operation state monitoring system and method for plate facing equipment, and relates to the technical field of intelligent fault prediction, and the method comprises the steps: collecting multi-dimensional data, and carrying out the preprocessing; based on the preprocessed multi-dimensional data, multi-dimensional features are extracted, and a fault risk index is generated; constructing an intelligent fault assessment model according to the fault risk index and the multi-dimensional features, and outputting a predicted fault result and specific fault display; calculating an environment compensation coefficient by combining a predicted fault result, specific fault display, multi-dimensional data and a fault risk index, and dynamically adjusting an equipment operation state; according to the method, the multi-level intelligent fault assessment model is constructed, and the fault risk index and the multi-dimensional features are comprehensively considered, so that the accuracy and robustness of fault prediction are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent fault prediction, and particularly to a system and method for monitoring the operating state of a sheet finishing device. Background Art

[0002] In modern industrial production, monitoring the operating state of a sheet finishing device is crucial for ensuring product quality and production efficiency. Traditional monitoring methods mainly rely on single-sensor data, such as temperature, pressure, etc., and judge the operating state of the device through manual inspection or simple threshold alarms. However, this method has many limitations, such as incomplete data collection, untimely fault warning, high false alarm rate, etc. With the development of sensor technology and data analysis technology, the collection and processing of multi-dimensional data have become possible, which provides a new way for more accurate device state monitoring. In recent years, the application of machine learning and artificial intelligence technologies in fault prediction and diagnosis has become increasingly widespread. By constructing intelligent models, potential faults of devices can be more effectively identified, and the accuracy and reliability of prediction can be improved.

[0003] Although certain progress has been made in existing multi-dimensional data collection and intelligent fault prediction technologies, there are still some deficiencies in practical applications. First, traditional methods are not fine enough in data preprocessing, and noise and outliers are not effectively filtered, resulting in a decrease in the accuracy of subsequent analysis. Second, existing fault assessment models often only consider single or a few features and fail to fully utilize the comprehensive information of multi-dimensional data, resulting in insufficient comprehensiveness and accuracy of fault prediction. In addition, existing technologies lack an effective mechanism for dynamically adjusting the operating state of the device and cannot flexibly adjust parameters according to real-time environmental conditions and device states, thus affecting the operating efficiency and stability of the device. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a method for monitoring the operating state of a sheet finishing device to solve the problem of low accuracy of fault prediction.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, the present invention provides a method for monitoring the operating state of a sheet finishing device, which includes collecting multi-dimensional data and performing preprocessing; extracting multi-dimensional features based on the preprocessed multi-dimensional data to generate a fault risk index; constructing an intelligent fault assessment model according to the fault risk index and multi-dimensional features, and outputting a predicted fault result and a specific fault display; calculating an environmental compensation coefficient by combining the predicted fault result, the specific fault display, the multi-dimensional data, and the fault risk index, and dynamically adjusting the operating state of the device.

[0007] As a preferred solution of the method for monitoring the operating state of the board finishing equipment of the present invention, wherein: the multi-dimensional data includes temperature data, pressure data, vibration data, working speed data, ambient temperature data, ambient humidity data, air quality data, acoustic data, image data, temperature distribution data and historical operation data.

[0008] As a preferred solution of the method for monitoring the operating state of the board finishing equipment of the present invention, wherein: the preprocessing includes noise filtering, data smoothing, outlier removal processing, acoustic signal spectrum analysis and image edge detection of the multi-dimensional data.

[0009] As a preferred solution of the method for monitoring the operating state of the board finishing equipment of the present invention, wherein: the extracting multi-dimensional features to generate a fault risk index, the expression is, Extract temperature data, pressure data, vibration data and working speed data from the preprocessed multi-dimensional data as physical features; Extract the square term of the vibration data from the preprocessed multi-dimensional data as a polynomial feature; Extract the temperature data change rate and historical operation data from the preprocessed multi-dimensional data as time series features; Extract the acoustic signal spectrum, image edge data and temperature distribution data from the preprocessed multi-dimensional data as fault display features; Based on the extraction of physical features, polynomial features and time series features, calculate the fault risk index, the expression is: ; Wherein, is the fault risk index, is the number of feature items, is the change amount of the i-th temperature data, is the time interval of the temperature data change rate, is the square term of the vibration frequency in the i-th vibration data, is the load borne by the influence of the i-th working speed data, is the operating pressure of the i-th pressure data, is the influence function of the historical fault data, and i is the index of the physical feature.

[0010] As a preferred solution of the method for monitoring the operating state of the board finishing equipment of the present invention, wherein: the constructing an intelligent fault evaluation model, the specific steps are, Define the input as the fault risk index and the fault display feature, and the output as the fault result and the specific fault display; Define the first layer of the intelligent fault assessment model. Input the fault risk index and fault display characteristics, introduce the stacking method to train the logistic regression model, decision tree model and random forest model and combine them, and output the direct prediction result through the logistic regression model in the combination; Define the second layer of the intelligent fault assessment model. Input the direct prediction result and fault display characteristics, and calculate the final prediction result through the decision tree model and random forest model in the combination. The expression is: ; Among them, is the output result of the final intelligent fault assessment model, is the number of base models, is the fault display characteristic, is the th direct prediction result of the base model, is the non-linear transformation function, is the th importance measure of the base model, is the index of the base model, is the complexity measure function, is the regularization term, is the th confidence of the base model, is the correction term, is the input value of the correction term; Define the third layer of the intelligent fault assessment model. Input the final prediction result and fault display characteristics, and output the predicted fault result and specific fault display through the logistic regression model, decision tree model and random forest model in the combination , and the expression is: ; Among them, is the category of the fault display, is the input feature, is the given input feature under which the probability value belonging to the category is.

[0011] As a preferred solution of the method for monitoring the operation state of the plate finishing equipment described in the present invention, among them: The calculation environment compensation coefficient dynamically adjusts the operation state of the equipment. The specific steps are: Dynamically adjust the equipment parameters by using the dynamic environment compensation algorithm. The expression is: ; Among them, is the environment compensation coefficient, is the external environment temperature, is the internal temperature of the equipment, External environmental humidity is the current load of the device is the environmental index is the device index; Set a threshold value, and dynamically adjust the operating state of the device according to and the specific fault display.

[0012] As a preferred solution of the method for monitoring the operating state of the sheet finishing device described in the present invention, wherein: the set threshold value dynamically adjusts the operating state of the device according to and the specific fault display. The specific steps are as follows When > A, the specific fault display is a serious fault, the environmental conditions have a high impact on the device, and the device is adjusted; When B < ≤ A, the specific fault display is a medium fault, the environmental conditions have a medium impact on the device, and the device is moderately adjusted; When ≤ B, the specific fault display is a minor fault, the environmental conditions have a small impact on the device, and the device maintains its current working state.

[0013] In a second aspect, the present invention provides a system for monitoring the operating state of a sheet finishing device, including an acquisition module, a fault risk module, an intelligent evaluation module, and an adjustment module; the acquisition module is used to acquire multi-dimensional data and perform preprocessing; the fault risk module is used to extract multi-dimensional features based on the preprocessed multi-dimensional data and generate a fault risk index; the intelligent evaluation module is used to construct an intelligent fault evaluation model according to the fault risk index and multi-dimensional features, and output a predicted fault result and a specific fault display; the adjustment module is used to calculate an environmental compensation coefficient by combining the predicted fault result, the specific fault display, the multi-dimensional data, and the fault risk index, and dynamically adjust the operating state of the device.

[0014] In a third aspect, the present invention provides a computer device, including a memory and a processor, and the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the method for monitoring the operating state of the sheet finishing device described in the first aspect of the present invention is implemented.

[0015] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, wherein: when the computer program is executed by the processor, any step of the method for monitoring the operating state of the sheet finishing device described in the first aspect of the present invention is implemented.

[0016] The beneficial effects of the present invention are as follows: By constructing a multi-level intelligent fault assessment model and comprehensively considering the fault risk index and multi-dimensional features, the present invention improves the accuracy and robustness of fault prediction. The first layer of the model outputs a preliminary prediction result through a combination of logistic regression, decision tree, and random forest. The second layer further combines the decision tree and random forest models to calculate the final prediction result. The third layer determines the specific fault type based on the final prediction result and the fault display characteristics. This multi-level model structure can not only more accurately predict the fault results of the equipment but also specifically classify the fault types, providing a basis for the maintenance and adjustment of the equipment, reducing false alarms and missed alarms, and improving the operation efficiency and safety of the equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0018] Figure 1 It is a flowchart of the method for monitoring the operation status of the board finishing equipment in Embodiment 1.

[0019] Figure 2 It is a block diagram of the system for monitoring the operation status of the board finishing equipment in Embodiment 1. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following will provide a detailed description of the specific embodiments of the present invention with reference to the accompanying drawings of the specification.

[0021] Many specific details are set forth in the following description to facilitate a full understanding of the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0022] Secondly, the so-called "one embodiment" or "embodiment" refers to a specific feature, structure, or characteristic that can be included in at least one implementation manner of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment that excludes other embodiments.

[0023] Embodiment 1, referring to Figure 1 and Figure 2 , is the first embodiment of the present invention. This embodiment provides a method for monitoring the operation status of a board finishing equipment, including the following steps: S1. Collect multi-dimensional data and perform preprocessing.

[0024] S1.1. Collect multi-dimensional data. Specifically, Furthermore, internal sensors collect temperature data, pressure data, vibration data, working speed data, acoustic data, and temperature distribution data; External sensors collect ambient temperature data, ambient humidity data, air quality data, and image data; Extract historical operation data from the device database, including fault information, maintenance records, operation habits, etc.; Internal sensors are: thermocouple sensors, pressure sensors, vibration sensors, optical encoders, acoustic sensors, and infrared thermal imagers; Collect temperature data through thermocouple sensors. Thermocouple sensors are used to detect the temperature of various key components of the device (such as heating elements, rollers, motors, etc.); Thermocouple sensors can work in high-temperature environments, while RTD sensors provide accurate temperature measurements and are widely used especially in the lower to medium temperature ranges; Pressure sensors collect pressure data. Pressure sensors are used to measure the pressure changes inside the device and are usually installed in pneumatic devices, hydraulic systems, or other components where pressure changes occur; Pressure sensors can help monitor whether the device is operating within a safe pressure range and avoid failures caused by overpressure; Vibration sensors collect vibration data. Vibration sensors are used to monitor the vibration conditions of key components in the device (such as motors, bearings, etc.); By monitoring vibration data, it is possible to detect whether the device has mechanical failures or component wear. Abnormal vibrations are usually early signs of device failures; Optical encoders collect working speed data. Optical encoders are used to measure the rotational or linear motion speed of the device; Optical encoders are widely used to detect the rotational speed of motors, while Hall effect sensors are commonly used to measure the speed of motor rotors or other rotating components; Acoustic sensors collect acoustic data and monitor the sound signals during device operation; Acoustic sensors can detect abnormal sounds inside the device, such as abnormal noises, etc. These sounds may be early signs of device failures; Infrared thermal imagers collect temperature distribution data and monitor the temperature distribution on the surface of the device; Infrared thermal imagers can provide a temperature distribution map of the device surface to help detect local overheating areas, which may be potential locations of failures; External sensors are: temperature sensors, humidity sensors, dust sensors, and cameras; Temperature sensors collect ambient temperature data. Temperature sensors are used to measure the temperature of the environment where the device is located, helping the system understand the impact of the external environment on the device. Ambient temperature sensors are usually installed on the device shell or in the operating environment to monitor temperature changes in the workshop; The humidity sensor collects environmental humidity data. The humidity sensor is used to measure the humidity of the environment where the device is located. Humidity can affect the drying speed of the coating, the adhesion of materials, and the overall operating efficiency of the device; the humidity sensor is generally installed in the workshop or around the device to continuously monitor the environmental humidity; The dust sensor collects air quality data. The dust sensor is used to monitor the air quality in the working environment of the device, especially the dust concentration. Dust or pollutants in the air may affect the operating efficiency of the device and even cause failures; the air quality sensor can detect particulate matter, VOC (volatile organic compounds), etc. in the air; The camera collects image data and captures images during the operation of the device; the image data can be used to detect abnormalities on the surface of the device, such as cracks, wear, etc., and can also be used to monitor the working state of the device.

[0025] S1.1. Pretreatment, specifically, Furthermore, perform noise filtering, data smoothing, and outlier removal on the data to generate clean data and ensure the accuracy and continuity of the data; Remove high-frequency noise through a FIR low-pass filter and retain a smoother signal; Effectively remove sharp noise points by replacing each point in the data with the median of the data in its neighborhood; Process the device dynamic parameter noise through a Kalman filter; Perform windowing on the data using simple moving average and weighted moving average, calculate the average value of the data within the moving window, and smooth short-term fluctuations; Assign decay weights to the data, pay attention to recent data, and gradually fade out past data, which is suitable for time series data; Perform convolution processing through a Gaussian kernel function to reduce noise and local drastic changes; Use the 3σ rule. For data with a normal distribution, by calculating the mean and standard deviation, remove outliers outside the range of μ±3σ; Detect outliers through the interquartile range IQR of the data and remove data points below the lower limit (Q1 - 1.5IQR) or above the upper limit (Q3 + 1.5IQR); Use clustering algorithms such as k-means or DBSCAN to remove abnormal data by identifying outliers in the data; Use a time series model (such as ARIMA) to predict the normal range and remove outliers that deviate significantly from the predicted values; Use the fast Fourier transform (FFT) to convert the time-domain signal into a frequency-domain signal, analyze the frequency components of the frequency-domain signal, identify abnormal signals within a specific frequency range, extract key features in the spectrum, such as the main frequency, spectral peak, etc., for subsequent fault diagnosis; Convert the color image to a grayscale image to simplify subsequent processing. Through steps such as Gaussian filtering, gradient calculation, non-maximum suppression, and double-threshold detection, extract the edges in the image, and extract key features in the edge detection results, such as edge length and edge direction, for subsequent fault diagnosis.

[0026] It should be noted that by collecting multi-dimensional data, including internal sensor and external sensor data, as well as historical operation data, the operation status of the device is comprehensively monitored; internal sensors (such as thermocouple sensors, pressure sensors, vibration sensors, optical encoders, acoustic sensors, and infrared thermal imagers) can accurately detect key parameters inside the device to ensure that the device operates in a safe and efficient state. External sensors (such as temperature sensors, humidity sensors, dust sensors, and cameras) monitor the external environmental conditions where the device is located, helping the system understand the impact of the external environment on the device and ensuring that the device adapts to different working environments. Through preprocessing steps, including noise filtering, data smoothing, and outlier removal, clean data is generated to ensure the accuracy and continuity of the data, improving the reliability of fault diagnosis.

[0027] S2. Based on the preprocessed multi-dimensional data, extract multi-dimensional features and generate a fault risk index.

[0028] Furthermore, extract temperature data, pressure data, vibration data, and working speed data from the preprocessed multi-dimensional data as physical features; The temperature data reflects the internal temperature and environmental temperature of the device and is used to analyze the thermal state of the device; The pressure data reflects the pressure borne by the device during operation; The vibration data reflects the mechanical vibration frequency of the device during operation; The working speed data reflects the operating efficiency of the device and approximately represents the load condition of the device; The environmental data reflects the external environmental temperature, humidity, air quality, etc. of the device and is an important background factor affecting the operation of the device; Extract the square term of the vibration data from the preprocessed multi-dimensional data as a polynomial feature; The square term of the vibration data is: ; Extract the temperature data change rate and historical operation data from the preprocessed multi-dimensional data as time series features; Extract the acoustic signal spectrum, image edge data, and temperature distribution data from the preprocessed multi-dimensional data as fault display features; Extract acoustic signal spectrum features, such as band energy and peak frequency, from the acoustic data; Extract the edge, texture, and color histogram features of the image from the image data; Extract the temperature distributions of different parts from the temperature distribution data, such as the hot spot area and the temperature gradient; Based on the extraction of physical features, polynomial features, and time series features, calculate the fault risk index, and the expression is: ; Where, is the fault risk index, is the number of feature items, is the change amount of the i-th temperature data, that is, the difference between the temperature data at adjacent time points, is the time interval of the temperature data transformation rate, is the square term of the vibration frequency in the i-th vibration data, and the vibration frequency is related to the mechanical stress, is the load borne by the influence of the i-th working speed data, is the operating pressure of the i-th pressure data, is the influence function of the historical fault data, and i is the index of the physical feature.

[0029] By introducing polynomial features (square terms of vibration data), time series features (temperature data transformation rate and historical operation data), and the influence function of historical fault data, this formula more comprehensively reflects the operating state and potential fault risks of the equipment, improves the accuracy and reliability of fault prediction, helps to detect and prevent faults in advance, and enhances the operating efficiency and safety of the equipment.

[0030] It should be noted that by extracting physical features, polynomial features, and time series features from the preprocessed multi-dimensional data to generate the fault risk index, a comprehensive evaluation of the equipment operating state is achieved; physical features (temperature, pressure, vibration, working speed data) reflect the working state and load conditions of the equipment, polynomial features (square terms of vibration data) enhance the sensitivity to mechanical stress, and time series features (temperature data transformation rate and historical operation data) consider the dynamic changes and historical fault conditions of the equipment; the fault display features extracted from the acoustic signal spectrum, image edge data, and temperature distribution data provide multi-angle fault detection bases.

[0031] S3. Construct an intelligent fault assessment model based on the fault risk index and multi-dimensional features, and output the predicted fault results and specific fault displays.

[0032] Furthermore, define the input as the fault risk index and fault display features, and the output as the fault results and specific fault displays (multi-classification: such as overheating, wear, mechanical failure, etc.); Define the intelligent fault assessment model as three layers; Define the first layer of the intelligent fault assessment model. Input the fault risk index and fault display features, introduce the stacking method to train the logistic regression model, decision tree model, and random forest model and combine them, and output the direct prediction result through the logistic regression model in the combination; Define the second layer of the intelligent fault assessment model. Input the direct prediction result and fault display features, and calculate the final prediction result through the decision tree model and random forest model in the combination. The expression is: ; where, is the output result of the final intelligent fault assessment model, that is, the probability or category of the predicted fault, is the number of base models, indicating the number of models participating in the combination, are the fault display features, is the th direct prediction output (such as probability value or category) of the base model (logistic regression, decision tree, random forest), is a non-linear transformation function used to perform non-linear processing on the output of the base model to enhance the complexity expression ability, is the th importance measure of the base model, which can be the training score, feature importance, or performance on the validation set of the model, is the index of the base model, is a complexity measure function used to evaluate the complexity of the input features and adjust the evaluation result, is a regularization term used to balance the model complexity and prevent overfitting, obtained by analyzing the feature complexity of the training data, is the th confidence or uncertainty measure of the base model. During the model training stage, evaluate the performance of each model (such as accuracy, F1-score, AUC, etc.) through cross-validation or the validation set, standardize the performance metrics into confidence values, and adjust the weights of each base model in the comprehensive evaluation, is a correction term, based on the adjustment of the overall error distribution and dynamically calculated through the real-time collected environmental data, is the input value of the correction term; Define the third layer of the intelligent fault assessment model. Input the final prediction result and fault display features, and output the predicted fault result and specific fault display through the logistic regression model, decision tree model, and random forest model in the combination , and the expression is: ; where, is the category of the fault display, such as "overheat", "wear", "mechanical failure", is the input feature, is the given input feature under which the probability value of belonging to the category is.

[0033] For example: The setting of the specific fault display threshold is as follows: When > 0.9 and T > 80 ∘ °C, overheat is displayed; When 0.7 < ≤ 0.9 and V > 1000, wear is displayed; When ≤ 0.7 and I > 0.5, mechanical failure is displayed; where T is the temperature of the hot spot area in the temperature distribution feature, V is the band energy in the acoustic feature, and I is the edge intensity in the image feature; The fault risk index = 0.85; Fault display features: The features of the acoustic data are: band energy 1000, peak frequency 5000 Hz; The features of the image data are: edge intensity 0.7, texture uniformity 0.9; The features of the temperature distribution data are: hot spot area temperature 80 °C, temperature gradient 10 °C / m; The output of the first layer model = 0.8; The output of the second layer model = 0.9; The output of the third layer model = 0.9. According to the defined fault display category, = 0.9, T = 80 °C, V = 1000, I = 0.7, so the specific fault display M = overheat; The specific threshold setting is determined according to the actual situation and relevant professionals. The above is only for example; This formula introduces nonlinear transformation, complexity measurement, regularization, and model confidence. The obtained prediction results improve the accuracy and robustness of the intelligent fault assessment model, can better handle complex fault situations, enhance the reliability of fault prediction, reduce false alarms and missed alarms, and optimize the equipment maintenance strategy.

[0034] The correction term is expressed as: ; where is the weight coefficient of the temperature deviation, is the weight coefficient of the humidity, is the weight coefficient of the air quality, is the external environmental temperature, is the optimal operating temperature of the device (the ideal temperature range set according to the operating characteristics of the device), is the external environmental humidity, is the air quality index (such as PM2.5 or PM10, reflecting the degree of air pollution).

[0035] Introducing the correction term takes into account the external environmental temperature deviation, humidity, and air quality index, reflects the impact of the external environment on the device, and improves the accuracy and robustness of fault prediction; It should be noted that by constructing a three-layer intelligent fault assessment model, integrating the fault risk index and multi-dimensional features, accurate prediction and classification of device faults are achieved; the first-layer model outputs a preliminary prediction result through a combination of logistic regression, decision tree, and random forest; the second-layer model introduces non-linear transformation, complexity measurement, regularization, and model confidence to improve the accuracy and robustness of the prediction; the third-layer model determines the specific fault type according to the final prediction result and fault display characteristics; through the collaborative work of multiple models and real-time environmental data correction, the model can more comprehensively reflect the operating state of the device, detect and classify faults in advance, optimize the maintenance strategy, reduce false alarms and missed alarms, and improve the operating efficiency and safety of the device.

[0036] S4. Combine the predicted fault results, specific fault displays, multi-dimensional data, and fault risk index to calculate the environmental compensation coefficient and dynamically adjust the device operating state.

[0037] Furthermore, based on the predicted fault results, multi-dimensional data, fault risk index, and specific fault display output by the intelligent fault assessment model, use the dynamic environmental compensation algorithm to dynamically adjust the device parameters according to the environmental conditions. The expression is: ; where, is the environmental compensation coefficient, indicating the adjustment amount required for the device operating parameters, is the external environmental temperature, is the internal temperature of the device, external environmental humidity, the current load of the device, is the environmental index, is the device index; Set a threshold, and dynamically adjust the device operating state according to and the specific fault display.

[0038] This formula comprehensively considers external environmental parameters, fault risk indices, equipment loads, and intelligent fault assessment results, achieving dynamic and precise adjustment of the equipment's operating state, effectively reducing the impact of environmental changes on equipment performance, enhancing the equipment's operating efficiency and reliability, and reducing the risk of faults.

[0039] S3.1. Set thresholds and, based on perform dynamic adjustment on the equipment. Specifically, When >A, the specific fault indicates a severe fault, the environmental conditions have a high impact on the equipment, and the equipment is adjusted; When B < ≤A, the specific fault indicates a medium fault, the environmental conditions have a medium impact on the equipment, and the equipment is moderately adjusted; When ≤B, the specific fault indicates a minor fault, the environmental conditions have a small impact on the equipment, and the equipment maintains its current working state.

[0040] For example, when >0.7, the specific fault indicates that M is a severe fault (such as overheating, mechanical failure), the environmental conditions have a significant adverse impact on the equipment operation (such as too high environmental temperature, too low humidity, etc.), and it is necessary to quickly take measures for compensation and adjustment (such as automatically reducing power output and controlling the equipment temperature within a safe range); 0.3 < ≤0.7, the specific fault indicates that M is a medium fault (such as wear, slight overheating), the environmental conditions do not have a particularly severe impact on the equipment operation, but certain actions are still required (such as slightly reducing the equipment power, reducing the power output by 10%, and observing the changes in equipment temperature and vibration) to ensure stable equipment operation; ≤0.3, the specific fault indicates that M is a minor fault (such as normal operation, slight vibration), the environmental conditions have a small impact on the equipment operation, and the equipment can continue to work normally in its current operating state without compensation or only requires minor adjustment (maintaining the current working state of the equipment, or slightly increasing the equipment output power).

[0041] The specific threshold settings are determined according to the actual situation and relevant professionals. The above is only for illustrative purposes; It should be noted that by calculating the environmental compensation coefficient and dynamically adjusting the device operation state according to its value, the comprehensive quantification and adaptive management of environmental factors and device state are realized; the external environmental temperature, internal device temperature, environmental humidity, failure risk index, device load and intelligent failure assessment results are comprehensively considered to ensure the stable operation of the device under different environmental conditions; by setting thresholds and taking corresponding adjustment measures according to different levels of environmental impact and failure display, the impact of environmental changes on device performance is effectively reduced, the operation efficiency and reliability of the device are improved, the risk of failure occurrence is reduced, and the safe and efficient operation of the device is ensured.

[0042] This embodiment also provides a monitoring system for the operation state of a board finishing device, including: a collection module, a failure risk module, an intelligent evaluation module and an adjustment module; the collection module is used to collect multi-dimensional data and perform preprocessing; the failure risk module is used to extract multi-dimensional features based on the preprocessed multi-dimensional data and generate a failure risk index; the intelligent evaluation module is used to construct an intelligent failure evaluation model according to the failure risk index and multi-dimensional features and output a predicted failure result and a specific failure display; the adjustment module is used to calculate the environmental compensation coefficient by combining the predicted failure result, the specific failure display, the multi-dimensional data and the failure risk index and dynamically adjust the device operation state.

[0043] This embodiment also provides a computer device applicable to the situation of the monitoring method for the operation state of a board finishing device, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the monitoring method for the operation state of a board finishing device as proposed in the above embodiment.

[0044] This computer device can be a terminal, and this computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of this computer device is used to provide computing and control capabilities. The memory of this computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of this computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be realized through WIFI, a carrier network, NFC (Near Field Communication) or other technologies. The display screen of this computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of this computer device can be a touch layer covered on the display screen, or a button, a trackball or a touchpad set on the shell of the computer device, or an external keyboard, a touchpad or a mouse, etc.

[0045] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the method for monitoring the operation state of the sheet finishing equipment as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, abbreviated as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, abbreviated as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, abbreviated as EPROM), programmable read-only memory (Programmable Red-Only Memory, abbreviated as PROM), read-only memory (Read-Only Memory, abbreviated as ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0046] In summary, by constructing a multi-level intelligent fault assessment model and comprehensively considering the fault risk index and multi-dimensional features, the present invention improves the accuracy and robustness of fault prediction; the first layer of the model outputs a preliminary prediction result through a combination of logistic regression, decision tree and random forest; the second layer further combines the decision tree and random forest models to calculate the final prediction result; the third layer determines the specific fault type according to the final prediction result and the fault display characteristics; this multi-level model structure can not only more accurately predict the fault result of the equipment, but also specifically classify the fault type, providing a basis for the maintenance and adjustment of the equipment, reducing false alarms and missed alarms, and improving the operation efficiency and safety of the equipment.

[0047] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A method for monitoring the operating state of a sheet finishing device, characterized in that: including collecting multi-dimensional data and performing preprocessing extracting multi-dimensional features from the preprocessed multi-dimensional data to generate a fault risk index constructing an intelligent fault assessment model based on the fault risk index and multi-dimensional features, and outputting a predicted fault result and a specific fault display calculating an environmental compensation coefficient by combining the predicted fault result, the specific fault display, the multi-dimensional data and the fault risk index, and dynamically adjusting the operating state of the device 2. The method for monitoring the operating state of the sheet finishing equipment according to claim 1, wherein: The multi-dimensional data includes temperature data, pressure data, vibration data, working speed data, ambient temperature data, ambient humidity data, air quality data, acoustic data, image data, temperature distribution data and historical operation data 3. The method for monitoring the operating state of the sheet finishing equipment according to claim 2, wherein: The preprocessing includes noise filtering, data smoothing, outlier removal processing, acoustic signal spectrum analysis and image edge detection for the multi-dimensional data 4. The method for monitoring the operation state of the board finishing equipment according to claim 3, characterized in that: The extracting of multi-dimensional features to generate a fault risk index, the expression is extracting temperature data, pressure data, vibration data and working speed data from the preprocessed multi-dimensional data as physical features extracting the square term of the vibration data from the preprocessed multi-dimensional data as a polynomial feature extracting the temperature data change rate and historical operation data from the preprocessed multi-dimensional data as time series features extracting the acoustic signal spectrum, image edge data and temperature distribution data from the preprocessed multi-dimensional data as fault display features calculating the fault risk index based on the extraction of physical features, polynomial features and time series features, the expression is ； Among them, is the failure risk index, is the number of characteristic items, is the change amount of the i-th temperature data, is the time interval of the temperature data change rate, is the square term of the vibration frequency in the i-th vibration data, is the load borne by the influence of the i-th working speed data, is the operating pressure of the i-th pressure data, is the influence function of historical failure data, and i is the index of physical characteristics.

5. The method for monitoring the operating state of the sheet finishing equipment according to claim 4, characterized in that: The constructing of the intelligent fault assessment model, the specific steps are defining the input as the fault risk index and the fault display features, and the output as the fault result and the specific fault display defining the first layer of the intelligent fault assessment model, inputting the fault risk index and the fault display features, introducing the stacking method to train the logistic regression model, decision tree model and random forest model and combining them, and outputting a direct prediction result through the logistic regression model in the combination defining the second layer of the intelligent fault assessment model, inputting the direct prediction result and the fault display features, and calculating the final prediction result through the decision tree model and random forest model in the combination, the expression is ； Among them, is the output result of the final intelligent fault assessment model, is the number of base models, is the fault display feature, is the direct prediction result of the th base model, is the importance measure of the th base model, is the index of the base model, is the complexity measure function, is the confidence of the th base model, is the correction term, and is the input value of the correction term; Define the third layer of the intelligent fault assessment model, input the final prediction result and the fault display features, and output the predicted fault result and the specific fault display through the logistic regression model, decision tree model, and random forest model in the combination , and the expression is: ； Among them, is the category of fault display, is the input feature, is the given input feature under which the probability value belonging to the category is.

6. The method for monitoring the operating state of the sheet finishing equipment according to claim 5, wherein: The calculating of the environmental compensation coefficient and dynamically adjusting the operating state of the device, the specific steps are dynamically adjusting the device parameters by using a dynamic environmental compensation algorithm, the expression is ； Among them, is the environmental compensation coefficient, is the external environmental temperature, is the internal temperature of the device, is the external environmental humidity, is the current load of the device, is the environmental index, is the device index; Set a threshold value and dynamically adjust the operating state of the device according to and the specific fault display.

7. The method for monitoring the operating state of the sheet finishing equipment according to claim 6, characterized in that: The set threshold is dynamically adjusted according to and the specific fault display for the operating state of the device. The specific steps are as follows: When >A, the specific fault indicates a serious fault, the environmental conditions have a high impact on the equipment, and the equipment is adjusted; When B < ≤ A, the specific fault indicates a medium fault, the environmental conditions have a medium impact on the equipment, and the equipment is moderately adjusted; When ≤ B, the specific fault indicates a minor fault, the environmental conditions have little impact on the device, and the device maintains its current working state.

8. A monitoring system for the operating state of a sheet finishing device, based on the method for monitoring the operating state of a sheet finishing device according to any one of claims 1 to 7, characterized in that: including a collection module, a fault risk module, an intelligent evaluation module and an adjustment module The collection module is used for collecting multi-dimensional data and performing preprocessing The fault risk module is used for extracting multi-dimensional features from the preprocessed multi-dimensional data to generate a fault risk index The intelligent evaluation module is used for constructing an intelligent fault assessment model based on the fault risk index and multi-dimensional features, and outputting a predicted fault result and a specific fault display The adjustment module is used for calculating an environmental compensation coefficient by combining the predicted fault result, the specific fault display, the multi-dimensional data and the fault risk index, and dynamically adjusting the operating state of the device 9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the method for monitoring the operating state of the sheet finishing device according to any one of claims 1 to 7 10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, it implements the steps of the method for monitoring the operating state of the sheet finishing equipment according to any one of claims 1 to 7.

Images