Edible oil production line control method and system based on Internet of Things

Abstract

The invention discloses an edible oil production line control method and system based on the Internet of Things, and relates to the technical field of edible oil production control, and the system can intelligently monitor the health condition of equipment in real time, carry out dynamic analysis and prediction, and automatically generate maintenance suggestions through the Internet of Things technology and a machine learning algorithm. According to the intelligent diagnosis and predictive maintenance mode, the accuracy of equipment management is improved, measures can be taken in advance before faults occur, and the equipment fault occurrence probability and the production downtime are greatly reduced. According to the system, a large amount of work depending on manual inspection and judgment in a traditional maintenance mode can be reduced through automatic data acquisition, analysis and processing. The frequency and accuracy of manual inspection are difficult to guarantee, however, the system guarantees the continuity and accuracy of equipment health state monitoring through automatic data acquisition and analysis, and equipment fault missing detection caused by human errors is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of edible oil production control, and in particular to an edible oil production line control method and system based on the Internet of Things. Background Art

[0002] Smart manufacturing refers to a production method that uses information technology and automation technology to improve production efficiency, reduce costs and improve product quality, covering multiple aspects such as automated control, data collection, and intelligent decision-making. In the field of smart manufacturing, the application of Internet of Things technology has become an important means to improve production line efficiency, reduce energy consumption, and improve quality control. Specifically, in the edible oil production line control system, the Internet of Things connects various equipment, production links, and quality control processes in real time through sensors and networks, realizes data collection, transmission, analysis, and feedback, and thus optimizes each link in the production process.

[0003] At present, the equipment management and maintenance in edible oil production lines mostly rely on manual inspections and regular maintenance plans. However, the traditional maintenance model has many shortcomings. First, there are certain time intervals and blind spots in manual inspections, and it is impossible to grasp the operating status and potential failure risks of the equipment in real time. Secondly, regular maintenance plans may not accurately reflect the true health status of the equipment, which may lead to excessive maintenance or production downtime due to failures. In addition, the lack of data-driven predictive maintenance strategies has led to equipment failures occurring at unexpected times, and problems cannot be discovered and solved in advance, increasing downtime and production losses. The existing maintenance methods cannot dynamically adjust the maintenance plan according to the actual operation of the equipment, nor can they identify potential failures of the equipment in a timely manner, affecting production efficiency and the service life of the equipment.

[0004] The shortcomings of this traditional maintenance method are mainly due to over-reliance on human factors and fixed maintenance plans, which lead to a disconnect between equipment maintenance and actual operating conditions. The status and operating environment of equipment change over time, however, fixed regular maintenance plans cannot accurately reflect this change. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides an edible oil production line control method and system based on the Internet of Things, which solves the problems mentioned in the background technology.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: an edible oil production line control system based on the Internet of Things, including a data acquisition module, a data preprocessing module, a feature extraction and health status analysis module, a fault prediction and maintenance suggestion module, a dynamic maintenance plan adjustment module and an overall data visualization and reporting module; The data acquisition module collects the operation data of the equipment, including temperature T, vibration f and pressure p, through the sensors installed on the production equipment, and fits them into the data operation set W; The data preprocessing module cleans and standardizes the collected data set W to obtain a health data set WK; The feature extraction and health status analysis module extracts features from the health data set WK, including the temperature change rate Tb, the vibration frequency fv and the pressure fluctuation Pb, to form a fault feature set GF, and uses a machine learning model to analyze the relationship between the fault feature set GF and the equipment fault, builds an equipment health assessment model, assesses the equipment health status, and obtains a health assessment result H; The fault prediction and maintenance suggestion module is based on historical fault data and fault feature set GF, and uses regression analysis algorithm to build a fault prediction model to predict the equipment failure risk in the future and obtain the future failure risk Prf(t); The dynamic maintenance plan adjustment module automatically adjusts the maintenance plan according to the health assessment result H and the future failure risk Prf(t); The overall data visualization and reporting module visualizes the equipment health status, equipment failure risk and maintenance plan, including fixed-period health reports, maintenance logs and failure analysis reports.

[0007] Preferably, the data acquisition module includes a data acquisition unit and a data integration unit; The data acquisition unit acquires the operation data of the equipment, including acquiring the temperature T of the equipment through a temperature sensor, acquiring the vibration f during the operation of the equipment through a vibration sensor, and acquiring the pressure P during the operation of the equipment through a pressure sensor; The data integration unit combines the collected operation data to form a data operation set W.

[0008] Preferably, the data preprocessing module includes a data cleaning unit and a data standardization unit; The data cleaning unit cleans the data running set W, including outlier detection, missing value processing and noise removal, to obtain a data cleaning set WC; Outlier detection involves using the three-times standard deviation rule to detect and remove outliers; Missing value handling includes filling missing values ​​using interpolation methods; Noise removal includes applying Kalman filter denoising method to remove noise caused by environmental interference and equipment operation fluctuations; The data standardization unit performs standardization processing on the data cleaning set WC, unifies the sensor data from different sources into the same dimension and scale, and obtains the health data set WK; The health data set WK is obtained by the following formula: ; Where WKd represents the d-th data in the healthy data set WK, WCd represents the d-th data in the data cleaning set WC, μd represents the mean of data d, and σd represents the standard deviation of data d.

[0009] Preferably, the feature extraction and health status analysis module includes a feature extraction unit and a health status assessment and fault prediction unit; The feature extraction unit extracts features from the healthy data set WK, including the temperature change rate Tb, the vibration frequency fv and the pressure fluctuation Pb, to form a fault feature set GF; The temperature change rate Tb is obtained by the following formula: ; In the formula, T(t) represents the temperature at time t, T(t-1) represents the temperature at time t-1, and Δt represents the time interval; The vibration frequency fv is obtained by the following formula: ; In the formula, represents the Fourier transform of the vibration signal; f(t) represents the vibration at time t, and argmax represents the maximum value of the parameter function of the function; The pressure fluctuation Pb is obtained by the following formula: ; Where N represents the total number of data points, P(ti) represents the pressure data of the i-th data point at time t; μP represents the mean value of pressure; The health status assessment and fault prediction unit analyzes the fault feature set GF according to the machine learning model, constructs an equipment health assessment model, and assesses the health status of the equipment; The equipment health assessment model will be trained based on the feature set GF of the equipment under different operating conditions; The training formula of the equipment health assessment model is as follows: ; In the formula, h(GF) represents the predicted health status of the device, and Model represents the machine learning model; it is used to predict the health status of the device from the feature set GF; Based on the trained equipment health assessment model, predict the equipment and obtain the equipment health assessment result H; The health assessment result H of the device is obtained by the following formula: ; In the formula, H(t) represents the health status of the device at time t, h(GF(t)) represents the health status of the device predicted at time t based on the fault feature set GF, and Evaluate represents the evaluation function. Specifically, it represents the evaluation result, 0 represents health and 1 represents fault.

[0010] Preferably, the fault prediction and maintenance suggestion module includes a fault prediction model building unit and a maintenance suggestion generating unit; The fault prediction model building unit trains a regression analysis model by using historical fault data and a fault feature set GF, and builds a fault prediction model by the regression analysis model to predict the fault occurrence probability Prf of the future fault risk of the equipment; The failure probability Prf is obtained by the following formula: ; In the formula, represents the intercept term of the fault prediction model, They represent the fault prediction model parameters of temperature change rate Tb, vibration frequency fv and pressure fluctuation Pb respectively, and Q represents the error term.

[0011] Preferably, the maintenance suggestion generating unit calculates the future failure risk Prf(t) of the equipment according to the acquired failure occurrence probability Prf, and generates repair and maintenance suggestions for the equipment; Through the trained regression model, the future failure risk Prf(t) is calculated based on the failure probability Prf and the failure feature set GF; The future failure risk Prf(t) is obtained by the following formula: ; In the formula, represents the adjustment coefficient, Tb(t) represents the temperature change rate at time t, fv represents the vibration frequency at time t, and Pb represents the pressure fluctuation at time t; By comparing the acquired future failure risk Prf(t) with the preset failure risk threshold TPrf, the risk status of the equipment can be determined; The risk status of the device is obtained by matching: When the future failure risk Prf(t) is less than the failure risk threshold TPrf, it means that the equipment has no failure risk; When the future failure risk Prf(t) ≥ the failure risk threshold TPrf, it means that the equipment has a failure risk.

[0012] Preferably, the dynamic maintenance plan adjustment module includes a health summary unit and a maintenance plan adjustment unit; The health summary unit determines the health status of the device based on the comparison between the acquired health assessment result H and the preset health threshold TrH; When the health assessment result H> the health threshold TrH, it means that the health status of the device is normal; When the health assessment result H ≤ the health threshold TrH, it means that the health status of the equipment is abnormal, and the maintenance plan is initiated, including preventive maintenance, repair suggestions and periodic maintenance.

[0013] Preventive maintenance: When the risk of equipment failure is high, it is recommended to perform preventive maintenance such as inspection, lubrication, and replacement of parts in advance; Repair suggestions: When the equipment has failed or is close to failure, it is recommended to repair it immediately; Periodic maintenance: Based on the service life of the equipment, regular maintenance is recommended, such as replacing filters and cleaning the equipment.

[0014] Avoid downtime caused by excessive maintenance or equipment failure based on the operating status of the equipment; if a piece of equipment is in good health, the system can postpone the maintenance plan to reduce maintenance costs; and if an impending failure is predicted, the system will arrange maintenance in advance.

[0015] Preferably, the maintenance plan adjustment unit automatically adjusts the maintenance plan Mty according to the acquired health assessment result H and future failure risk Prf(t); The maintenance plan Mty is obtained by the following formula: ; In the formula, Mpo represents the deferred maintenance plan, which means that the equipment is in good health and has a low risk of failure, so maintenance can be postponed, and Mad represents the advanced maintenance plan, which means that the equipment is in poor health or has a high risk of failure, so maintenance and repairs need to be carried out in advance.

[0016] Preferably, the overall data visualization and reporting module includes a data visualization unit and a report generation and scheduling unit; The data visualization unit displays the equipment health status, failure risk and maintenance plan in the form of intuitive diagrams; The report generation and scheduling unit regularly generates health reports, maintenance logs and fault analysis reports based on real-time equipment data and equipment health status; Among them, the health report includes a summary of the health status of the equipment, temperature change rate Tb, vibration frequency fv and pressure fluctuation Pb, providing the overall health level of the equipment; The maintenance log includes the specific operation, time, maintenance content and execution personnel information of each maintenance; Fault analysis reports include generating equipment fault analysis reports based on historical fault data and current predictions.

[0017] The edible oil production line control method based on the Internet of Things includes the following steps: Step 1: The data acquisition module collects the operating data of the equipment by installing sensors on the production equipment, including temperature T, vibration f and pressure p, and fits them into the data operation set W; Step 2: The data preprocessing module cleans and standardizes the collected data set W to obtain the health data set WK; Step 3: The feature extraction and health status analysis module extracts features from the health data set WK, including the temperature change rate Tb, vibration frequency fv, and pressure fluctuation Pb, to form a fault feature set GF, and uses a machine learning model to analyze the relationship between the fault feature set GF and the equipment fault, build an equipment health assessment model, assess the equipment health status, and obtain the health assessment result H; Step 4: The fault prediction and maintenance suggestion module builds a fault prediction model based on historical fault data and fault feature set GF by using regression analysis algorithm to predict the equipment failure risk in the future and obtain the future failure risk Prf(t); Step 5: The dynamic maintenance plan adjustment module automatically adjusts the maintenance plan according to the health assessment result H and the future failure risk Prf(t); Step 6: The overall data visualization and reporting module visualizes the equipment health status, equipment failure risk and maintenance plan, including fixed-cycle health reports, maintenance logs and failure analysis reports.

[0018] The present invention provides an edible oil production line control method and system based on the Internet of Things, which has the following beneficial effects: (1) When the system is running, the system can monitor the health status of the equipment in real time and intelligently, conduct dynamic analysis and prediction, and automatically generate maintenance suggestions through the Internet of Things technology and machine learning algorithms. This intelligent diagnosis and predictive maintenance mode not only improves the accuracy of equipment management, but also takes measures in advance before a failure occurs, greatly reducing the probability of equipment failure and production downtime. The system of the present invention can automatically collect, analyze and process data, reducing the large amount of work that relies on manual inspections and judgments in the traditional maintenance mode. The frequency and accuracy of manual inspections are difficult to guarantee, but the system ensures the continuity and accuracy of equipment health status monitoring through automated data collection and analysis, avoiding missed equipment failures due to human errors.

[0019] The system's intelligent analysis and visual reporting functions can help managers understand the status of equipment in real time, so as to arrange production plans and resource allocation more effectively. Through predictive maintenance, the downtime of the production line and the loss of production capacity caused by failures are reduced, while the resource utilization of the equipment is improved and the production process is optimized. The system can generate periodic health reports, fault analysis reports and maintenance logs to provide reliable data support for decision makers. The automatic generation and visual display of these reports make equipment management more transparent, help managers make more informed decisions, and improve the flexibility and responsiveness of production line operations.

[0020] (2) The health status assessment model based on machine learning can learn from different equipment operating states and make predictions based on historical data and current features. This method is more scientific and reliable than the traditional empirical method, and can provide dynamic health assessments and predict impending failures. Through this intelligent prediction, fault intervention can be carried out in advance, reducing production downtime and maintenance costs. By dynamically adjusting the maintenance plan, the system can avoid problems caused by excessive maintenance or neglect of maintenance in traditional maintenance methods. The dynamic maintenance plan can be adjusted according to the actual operating status of the equipment, avoiding unnecessary maintenance work, while ensuring that maintenance can be carried out in advance when the equipment is at risk of failure. This not only improves the operating efficiency of the equipment, but also reduces the equipment maintenance cost and improves the availability of the production line.

[0021] (3) By training the regression analysis model, the historical fault data and the fault feature set GF are fully utilized to accurately predict the future failure risk of the equipment. Compared with the traditional experience-based maintenance method, this method based on data analysis and regression modeling can provide more scientific and accurate prediction results, reduce the probability of misjudgment, and thus achieve more effective fault management. By predicting the future failure risk of the equipment, the system can generate repair and maintenance recommendations in advance before the failure occurs. Unlike traditional scheduled maintenance or post-failure maintenance, this dynamic maintenance method based on actual data and prediction results is more flexible and accurate, and can reduce maintenance costs and unnecessary maintenance frequency while ensuring the normal operation of the equipment.

[0022] By comparing the future failure risk Prf(t) with the preset failure risk threshold TPrf, the risk status of the equipment is automatically determined and corresponding maintenance decisions are made. This automated risk assessment and maintenance decision-making process reduces the errors and uncertainties of human judgment, and can formulate equipment maintenance plans more timely and accurately, avoiding human negligence or delays.

[0023] (4) By comparing the health assessment results with the health threshold, the health status of the equipment can be intelligently judged, avoiding the human error of equipment health assessment in the traditional maintenance mode. This makes maintenance decisions more scientific and accurate, and avoids equipment failures caused by excessive maintenance or negligence. By combining equipment health assessment and fault prediction, the system can flexibly adjust the maintenance plan, avoiding the disadvantages of fixed-time periodic maintenance and ensuring that the equipment receives the necessary maintenance at the appropriate time. This dynamic adjustment can effectively reduce unnecessary maintenance operations, reduce operating costs, and reduce production interruptions caused by equipment failures.

[0024] The dynamically adjusted maintenance plan can perform preventive maintenance and repairs in a timely manner based on real-time evaluation results and fault prediction results, significantly reducing equipment downtime caused by failures or excessive maintenance. The production line can run more efficiently, improving overall production efficiency. The data visualization unit displays key information such as the health status of the equipment, failure risks, and maintenance plans in a graphical manner, allowing managers to quickly understand the health status and potential problems of the equipment, so as to make more efficient and accurate decisions. Managers no longer need to rely on complex data analysis, but can see the key indicators of the equipment at a glance. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a schematic diagram of the flow chart of the control system of an edible oil production line based on the Internet of Things of the present invention; Figure 2 The figure is a schematic diagram of the steps of the edible oil production line control method based on the Internet of Things of the present invention. DETAILED DESCRIPTION

[0026] The technical solutions in the embodiments of the present invention will be clearly and completely described 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. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0027] Example 1 The present invention provides an edible oil production line control system based on the Internet of Things. Figure 1 , including data acquisition module, data preprocessing module, feature extraction and health status analysis module, fault prediction and maintenance suggestion module, dynamic maintenance plan adjustment module and overall data visualization and reporting module; The data acquisition module collects the operation data of the equipment, including temperature T, vibration f and pressure p, through the sensors installed on the production equipment, and fits them into the data operation set W; The data preprocessing module cleans and standardizes the collected data set W to obtain a health data set WK; The feature extraction and health status analysis module extracts features from the health data set WK, including the temperature change rate Tb, the vibration frequency fv and the pressure fluctuation Pb, to form a fault feature set GF, and uses a machine learning model to analyze the relationship between the fault feature set GF and the equipment fault, builds an equipment health assessment model, assesses the equipment health status, and obtains a health assessment result H; The fault prediction and maintenance suggestion module is based on historical fault data and fault feature set GF, and uses regression analysis algorithm to build a fault prediction model to predict the equipment failure risk in the future and obtain the future failure risk Prf(t); The dynamic maintenance plan adjustment module automatically adjusts the maintenance plan according to the health assessment result H and the future failure risk Prf(t); The overall data visualization and reporting module visualizes the equipment health status, equipment failure risk and maintenance plan, including fixed-period health reports, maintenance logs and failure analysis reports.

[0028] In this embodiment, based on the Internet of Things technology, the system can fully understand the health status of the equipment by real-time monitoring of the equipment operating status including temperature T, vibration f and pressure p. The data acquisition module obtains the operating data of various types of equipment in real time through sensors. Combined with machine learning analysis, the system can automatically identify the trend of equipment health changes and timely warn of potential failure risks. This intelligent equipment health management mode greatly improves the accuracy and real-time performance of equipment monitoring.

[0029] The fault prediction and maintenance suggestion module predicts the future failure risk of the equipment through machine learning algorithms based on historical fault data and current equipment operation status. Compared with the traditional regular maintenance mode, this system can realize predictive maintenance, identify the potential failure risk of the equipment in advance, avoid failures during the production process, and reduce downtime and economic losses caused by sudden failures. Predictive maintenance significantly improves the stability and production efficiency of the production line. The dynamic maintenance plan adjustment module automatically adjusts the maintenance plan according to the health assessment results and failure risk prediction of the equipment to avoid unnecessary excessive maintenance. The system reasonably arranges maintenance time according to the actual health status of the equipment and fault prediction, avoids the cost waste caused by excessive maintenance, and reduces the downtime loss caused by equipment failure. In this way, not only the equipment utilization efficiency is improved, but also the operating cost of the production line is reduced.

[0030] Through continuous health monitoring and intelligent diagnosis, the system can intervene in time according to the actual condition of the equipment and extend the service life of the equipment. It avoids equipment damage caused by excessive use or long-term lack of maintenance. The system's health assessment model comprehensively analyzes equipment operation data, which can help maintenance personnel better understand the true health status of the equipment, reasonably arrange maintenance time, and improve the long-term reliability of the equipment. Through real-time health status monitoring, fault prediction and optimized maintenance plans, the system can ensure that the equipment always maintains a good operating state during the production process and reduce the downtime of the equipment due to failures. Ultimately, it can extend the service life of the equipment, improve the reliability of the equipment, and ensure the stability and efficiency of the edible oil production line in long-term operation.

[0031] Example 2 This embodiment is explained in Example 1, please refer to Figure 1 ,Specifically: the data acquisition module includes a data acquisition unit and a data integration unit; The data acquisition unit acquires the operation data of the equipment, including acquiring the temperature T of the equipment through a temperature sensor, acquiring the vibration f during the operation of the equipment through a vibration sensor, and acquiring the pressure P during the operation of the equipment through a pressure sensor; The data integration unit combines the collected operation data to form a data operation set W.

[0032] The data preprocessing module includes a data cleaning unit and a data standardization unit; The data cleaning unit cleans the data running set W, including outlier detection, missing value processing and noise removal, to obtain a data cleaning set WC; Outlier detection involves using the three-times standard deviation rule to detect and remove outliers; Missing value handling includes filling missing values ​​using interpolation methods; Noise removal includes applying Kalman filter denoising method to remove noise caused by environmental interference and equipment operation fluctuations; The data standardization unit performs standardization processing on the data cleaning set WC, unifies the sensor data from different sources into the same dimension and scale, and obtains the health data set WK; The health data set WK is obtained by the following formula: ; Where WKd represents the d-th data in the healthy data set WK, WCd represents the d-th data in the data cleaning set WC, μd represents the mean of data d, and σd represents the standard deviation of data d.

[0033] The feature extraction and health status analysis module includes a feature extraction unit and a health status assessment and fault prediction unit; The feature extraction unit extracts features from the healthy data set WK, including the temperature change rate Tb, the vibration frequency fv and the pressure fluctuation Pb, to form a fault feature set GF; The temperature change rate Tb is obtained by the following formula: ; In the formula, T(t) represents the temperature at time t, T(t-1) represents the temperature at time t-1, and Δt represents the time interval; The vibration frequency fv is obtained by the following formula: ; In the formula, represents the Fourier transform of the vibration signal; f(t) represents the vibration at time t, and argmax represents the maximum value of the parameter function of the function; The pressure fluctuation Pb is obtained by the following formula: ; Where N represents the total number of data points, P(ti) represents the pressure data of the i-th data point at time t; μP represents the mean value of pressure; The health status assessment and fault prediction unit analyzes the fault feature set GF according to the machine learning model, constructs an equipment health assessment model, and assesses the health status of the equipment; The equipment health assessment model will be trained based on the feature set GF of the equipment under different operating conditions; The training formula of the equipment health assessment model is as follows: ; In the formula, h(GF) represents the predicted equipment health status, and Model represents the machine learning model; Based on the trained equipment health assessment model, predict the equipment and obtain the equipment health assessment result H; The health assessment result H of the device is obtained by the following formula: ; Where H(t) represents the health status of the device at time t, h(GF(t)) represents the health status of the device predicted at time t based on the fault feature set GF, and Evaluate represents the evaluation function.

[0034] In this embodiment, this embodiment uses multiple sensors to collect the operation data of the equipment in a comprehensive and real-time manner, which can timely reflect the operation status of the equipment. The data acquisition module is composed of a data acquisition unit and a data integration unit. The temperature sensor, vibration sensor and pressure sensor are used to continuously monitor the operation parameters of the equipment to ensure the accuracy and comprehensiveness of the data. This process provides real and comprehensive equipment operation data, which can help managers of the production line to grasp the working status of the equipment in real time and improve monitoring efficiency. Through data cleaning and standardization processing, the system can eliminate noise, outliers and missing values ​​in the data to ensure the accuracy and reliability of subsequent analysis. Outlier detection uses the triple standard deviation rule, missing value processing uses the interpolation method, and noise removal is performed through Kalman filtering. These processing methods can greatly improve data quality, remove interference factors that may affect the analysis results, and ensure the accuracy of the equipment health status assessment model. The feature extraction and health status analysis module extracts features such as temperature change rate, vibration frequency and pressure fluctuation to form a fault feature set GF, which provides a basis for the evaluation of the health status of the equipment and fault prediction. Through feature extraction, the dynamic operation status of the equipment can be comprehensively evaluated, providing accurate input data for subsequent fault prediction. This analysis not only helps identify the current health status of equipment, but also detects potential failures in advance, avoiding unnecessary equipment downtime and losses during the production process.

[0035] Through the machine learning-based health assessment and fault prediction unit, the system can automatically learn and build a device health assessment model to predict the health status of the device. The model is trained based on the fault feature set GF of the device under different operating conditions, and can more accurately assess the health status and potential risks of the device. Through the trained model, the health status of the device can be evaluated in real time, so as to discover potential faults and risks in advance and reduce the probability of equipment failure.

[0036] By combining the equipment health assessment results with the fault feature set GF, the system can predict future failure risks before equipment problems occur and provide accurate maintenance recommendations. This predictive maintenance method helps adjust production plans and avoid production line shutdowns due to equipment failures. In addition, automatically generated maintenance plans can avoid excessive or missed maintenance, reduce maintenance costs, and ensure that equipment operates in the best condition.

[0037] Example 3 This embodiment is explained in Example 2. Please refer to Figure 1 ,Specifically: the fault prediction and maintenance suggestion module includes a fault prediction model building unit and a ,maintenance suggestion generating unit; The fault prediction model building unit trains a regression analysis model by using historical fault data and a fault feature set GF, and builds a fault prediction model by the regression analysis model to predict the fault occurrence probability Prf of the future fault risk of the equipment; The failure probability Prf is obtained by the following formula: ; In the formula, represents the intercept term of the fault prediction model, They represent the fault prediction model parameters of temperature change rate Tb, vibration frequency fv and pressure fluctuation Pb respectively, and Q represents the error term.

[0038] The maintenance suggestion generating unit calculates the future failure risk Prf(t) of the equipment according to the acquired failure occurrence probability Prf, and generates repair and maintenance suggestions for the equipment; Through the trained regression model, the future failure risk Prf(t) is calculated based on the failure probability Prf and the failure feature set GF; The future failure risk Prf(t) is obtained by the following formula: ; In the formula, represents the adjustment coefficient, Tb(t) represents the temperature change rate at time t, fv represents the vibration frequency at time t, and Pb represents the pressure fluctuation at time t; By comparing the acquired future failure risk Prf(t) with the preset failure risk threshold TPrf, the risk status of the equipment can be determined; The risk status of the device is obtained by matching: When the future failure risk Prf(t) is less than the failure risk threshold TPrf, it means that the equipment has no failure risk; When the future failure risk Prf(t) ≥ the failure risk threshold TPrf, it means that the equipment has a failure risk.

[0039] In this embodiment, by using historical fault data and fault feature set GF to train the regression analysis model, a fault prediction model is successfully constructed, which can accurately predict the probability of future equipment failures. The probability of failure Prf obtained by the regression analysis model provides a reliable quantitative basis for the health status of the equipment, so that potential failure risks can be warned in advance and sudden downtime caused by equipment failures can be reduced. By obtaining the future failure risk Prf (t) and comparing it with the preset failure risk threshold TPref, the system can intelligently generate repair and maintenance recommendations for the equipment. When the future failure risk of the equipment exceeds the threshold, the system will automatically generate an advance repair and maintenance plan to avoid production downtime due to equipment failures and help production line managers better arrange maintenance work.

[0040] The combination of the fault prediction model and the maintenance suggestion generation unit can automatically adjust and optimize the maintenance plan. The system can not only issue fault risk warnings in a timely manner, but also adjust maintenance strategies based on real-time data and prediction results to avoid excessive maintenance or ignoring potential problems of the equipment. This intelligent management method can not only effectively reduce equipment downtime, but also reduce maintenance costs and ensure the continuous and stable operation of the production line. Through accurate fault prediction and dynamic maintenance plan adjustment based on this prediction, the sudden failure and downtime of equipment can be significantly reduced. Timely warnings and maintenance strategies can ensure that equipment is repaired in time when the risk of failure increases, thereby avoiding the long-term impact of equipment failure on production and improving the reliability of the production line and the availability of equipment.

[0041] This embodiment can identify possible equipment failures in advance based on the prediction results, and make timely maintenance arrangements, thus avoiding production downtime caused by equipment failures. Through fault risk warnings and advance maintenance arrangements, this system can effectively reduce the frequency of equipment failures and reduce downtime caused by sudden failures, thereby improving the overall efficiency of the production line. Based on the fault risk prediction results, the system can provide a scientific basis for equipment maintenance, allowing production line managers to allocate maintenance resources more reasonably and avoid excessive and ineffective maintenance. Through predictive maintenance, the system helps managers optimize the use of maintenance resources, improve the overall resource allocation efficiency of the production line, and reduce maintenance costs.

[0042] Example 4 This embodiment is explained in Example 3, please refer to Figure 1 ,Specifically: the dynamic maintenance plan adjustment module includes a health summary unit and a maintenance plan adjustment unit; The health summary unit determines the health status of the device based on the comparison between the acquired health assessment result H and the preset health threshold TrH; When the health assessment result H> the health threshold TrH, it means that the health status of the device is normal; When the health assessment result H ≤ the health threshold TrH, it means that the health status of the equipment is abnormal, and the maintenance plan is initiated, including preventive maintenance, repair suggestions and periodic maintenance.

[0043] The maintenance plan adjustment unit automatically adjusts the maintenance plan Mty according to the acquired health assessment result H and future failure risk Prf(t); The maintenance plan Mty is obtained by the following formula: ; Where Mpo represents the deferred maintenance plan, and Mad represents the advanced maintenance plan.

[0044] The overall data visualization and reporting module includes a data visualization unit and a report generation and scheduling unit; The data visualization unit displays the equipment health status, failure risk and maintenance plan in the form of intuitive diagrams; The report generation and scheduling unit regularly generates health reports, maintenance logs and fault analysis reports based on real-time equipment data and equipment health status; Among them, the health report includes a summary of the health status of the equipment, temperature change rate Tb, vibration frequency fv and pressure fluctuation Pb, providing the overall health level of the equipment; The maintenance log includes the specific operation, time, maintenance content and execution personnel information of each maintenance; Fault analysis reports include generating equipment fault analysis reports based on historical fault data and current predictions.

[0045] In this embodiment, the health status of the equipment can be accurately determined by comparing the health assessment result H of the equipment with the preset health threshold TrH by the health summary unit. When the health assessment result of the equipment is higher than the threshold, it indicates that the equipment is in good condition and no excessive intervention is required; if the health assessment result is lower than the threshold, the maintenance plan is started, including preventive maintenance and repair suggestions. This intelligent judgment helps production line managers avoid excessive maintenance, while ensuring that the equipment is repaired in time when the health status declines, ensuring that the equipment is always in the best working condition.

[0046] The maintenance plan adjustment unit automatically adjusts the maintenance plan Mty based on the equipment health assessment result H and future failure risk Prf(t). By combining real-time equipment health assessment and failure prediction, the system can accurately predict the actual maintenance needs of the equipment, thereby flexibly adjusting the maintenance strategy to avoid excessive maintenance in fixed cycles while ensuring that repairs and maintenance can be carried out in a timely manner when potential failures occur in the equipment. This dynamic adjustment optimizes the allocation of maintenance resources, reduces unnecessary maintenance costs, and improves the overall efficiency of the production line and the availability of equipment.

[0047] By dynamically adjusting the maintenance plan, this embodiment can avoid excessive downtime of the production line due to equipment maintenance, and effectively avoid sudden downtime caused by ignoring equipment failures. Combining health assessment and fault prediction, the system can intelligently decide when to perform preventive maintenance or repairs, ensuring that the equipment can be maintained in time when it is most needed, thereby reducing the downtime of the production line and improving production efficiency.

[0048] The overall data visualization and reporting module intuitively displays the equipment health status, failure risk and maintenance plan through the data visualization unit and the report generation and scheduling unit. The visualization of equipment health status, failure risk and maintenance plan can help production line managers quickly understand the overall health status and potential problems of the equipment, make decisions quickly, and avoid decision-making errors caused by delayed or inaccurate information.

[0049] The report generation and scheduling unit regularly generates health reports, maintenance logs, and fault analysis reports, allowing managers to obtain detailed equipment health information and maintenance records in a timely manner. The health report provides managers with the overall health level of the equipment, including temperature change rate Tb, vibration frequency fv, and pressure fluctuation Pb parameters, to help evaluate the operating health of the equipment; the maintenance log records the specific operation, time, content, and execution personnel of each maintenance, which is convenient for future tracking and review; the fault analysis report generates equipment fault analysis based on historical data and current predictions, providing important reference for subsequent equipment optimization.

[0050] Example 5 For the control method of edible oil production line based on Internet of Things, please refer to Figure 2 , specifically: including the following steps: Step 1: The data acquisition module collects the operating data of the equipment by installing sensors on the production equipment, including temperature T, vibration f and pressure p, and fits them into the data operation set W; Step 2: The data preprocessing module cleans and standardizes the collected data set W to obtain the health data set WK; Step 3: The feature extraction and health status analysis module extracts features from the health data set WK, including the temperature change rate Tb, vibration frequency fv, and pressure fluctuation Pb, to form a fault feature set GF, and uses a machine learning model to analyze the relationship between the fault feature set GF and the equipment fault, build an equipment health assessment model, assess the equipment health status, and obtain the health assessment result H; Step 4: The fault prediction and maintenance suggestion module builds a fault prediction model based on historical fault data and fault feature set GF by using regression analysis algorithm to predict the equipment failure risk in the future and obtain the future failure risk Prf(t); Step 5: The dynamic maintenance plan adjustment module automatically adjusts the maintenance plan according to the health assessment result H and the future failure risk Prf(t); Step 6: The overall data visualization and reporting module visualizes the equipment health status, equipment failure risk and maintenance plan, including fixed-cycle health reports, maintenance logs and failure analysis reports.

[0051] In this embodiment, by installing sensors on the production equipment and collecting temperature T, vibration f and pressure p parameters in real time, the data acquisition module can accurately capture the operating status of the equipment. This process not only improves the comprehensiveness of the data, but also provides real-time feedback on the status changes of the equipment, providing accurate basic data for subsequent health assessments, fault predictions and maintenance decisions; the data preprocessing module can effectively eliminate the interference of outliers, noise and missing values ​​by cleaning and standardizing the collected data, thereby ensuring the accuracy and consistency of the data used in subsequent analysis. This processing process enables data from different sources to be converted to the same scale, which is convenient for comprehensive analysis and evaluation, and improves the reliability of the data and the accuracy of the analysis.

[0052] The fault prediction and maintenance suggestion module combines historical fault data and feature set GF, builds a fault prediction model through regression analysis algorithm, and can accurately predict the future failure risk of equipment. By obtaining the future failure risk Prf (t), the system can identify potential problems of equipment in advance and make maintenance suggestions for possible failures, ensuring that the production line will not be shut down due to equipment failure at critical moments, reducing production downtime.

[0053] The dynamic maintenance plan adjustment module can automatically adjust the maintenance plan in real time according to the health assessment result H of the equipment and the future failure risk Prf (t). When the equipment is in good health, the system can delay the maintenance plan to reduce maintenance costs; when it is predicted that the equipment is about to fail, the system will arrange maintenance and repair in advance to ensure that the operation of the equipment is not affected. This flexible maintenance plan reduces unnecessary maintenance costs while avoiding downtime caused by excessive maintenance and failures. The data acquisition module ensures comprehensive and timely monitoring of the equipment's operating status through high-precision sensors and real-time data acquisition. The acquisition of real-time data provides an accurate basis for subsequent fault prediction and health assessment, making equipment management more accurate and preventing potential problems from being missed.

[0054] The fault prediction and maintenance suggestion module can arrange maintenance plans in advance according to the predicted fault risks, avoiding production line downtime due to equipment failure. In addition, the fault prediction and maintenance suggestion module can also avoid excessive maintenance and reduce unnecessary maintenance costs. Through accurate fault prediction and intelligent maintenance arrangements, the downtime of the production line can be minimized and production efficiency can be improved. The dynamic maintenance plan adjustment module flexibly adjusts the maintenance plan according to changes in equipment health status and fault risks, so that equipment maintenance not only meets actual needs, but also avoids waste of resources to the greatest extent. The flexible adjustment mechanism optimizes maintenance costs, while ensuring that equipment can be maintained in a timely manner at the right time, reducing the operating risks of the production line.

[0055] This embodiment provides a complete intelligent process from data collection to fault prediction, maintenance recommendations, health assessment, and maintenance plan adjustment. The system can dynamically adjust the maintenance strategy based on the health status and fault prediction data of the equipment to ensure that the equipment is always in good operating condition. This full-process intelligent management greatly improves production efficiency, reduces manual intervention, and reduces management complexity.

[0056] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. The edible oil production line control system based on the Internet of Things is characterized by: It includes data acquisition module, data preprocessing module, feature extraction and health status analysis module, fault prediction and maintenance suggestion module, dynamic maintenance plan adjustment module and overall data visualization and reporting module; The data acquisition module collects the operation data of the equipment, including temperature T, vibration f and pressure p, through sensors installed on the production equipment, and fits them into a data operation set W; The data preprocessing module cleans and standardizes the collected data set W to obtain a health data set WK; The feature extraction and health status analysis module extracts features from the health data set WK, including the temperature change rate Tb, the vibration frequency fv and the pressure fluctuation Pb, to form a fault feature set GF, and uses a machine learning model to analyze the relationship between the fault feature set GF and the equipment fault, builds an equipment health assessment model, assesses the equipment health status, and obtains a health assessment result H; The fault prediction and maintenance suggestion module is based on historical fault data and fault feature set GF, and uses regression analysis algorithm to build a fault prediction model to predict the equipment failure risk in the future and obtain the future failure risk Prf(t); The dynamic maintenance plan adjustment module automatically adjusts the maintenance plan according to the health assessment result H and the future failure risk Prf(t); The overall data visualization and reporting module visualizes the equipment health status, equipment failure risk and maintenance plan, including fixed-period health reports, maintenance logs and failure analysis reports.

2. The edible oil production line control system based on the Internet of Things according to claim 1 is characterized in that: The data acquisition module includes a data acquisition unit and a data integration unit; The data acquisition unit acquires the operation data of the equipment, including acquiring the temperature T of the equipment through a temperature sensor, acquiring the vibration f during the operation of the equipment through a vibration sensor, and acquiring the pressure P during the operation of the equipment through a pressure sensor; The data integration unit combines the collected operation data to form a data operation set W.

3. The edible oil production line control system based on the Internet of Things according to claim 2 is characterized in that: The data preprocessing module includes a data cleaning unit and a data standardization unit; The data cleaning unit cleans the data running set W, including outlier detection, missing value processing and noise removal, to obtain a data cleaning set WC; Outlier detection involves using the three-times standard deviation rule to detect and remove outliers; Missing value handling includes filling missing values ​​using interpolation methods; Noise removal includes applying Kalman filter denoising method to remove noise caused by environmental interference and equipment operation fluctuations; The data standardization unit performs standardization processing on the data cleaning set WC, unifies the sensor data from different sources into the same dimension and scale, and obtains the health data set WK; The health data set WK is obtained by the following formula: ； Where WKd represents the d-th data in the healthy data set WK, WCd represents the d-th data in the data cleaning set WC, μd represents the mean of data d, and σd represents the standard deviation of data d.

4. The edible oil production line control system based on the Internet of Things according to claim 1 is characterized in that: The feature extraction and health status analysis module includes a feature extraction unit and a health status assessment and fault prediction unit; The feature extraction unit extracts features from the healthy data set WK, including the temperature change rate Tb, the vibration frequency fv and the pressure fluctuation Pb, to form a fault feature set GF; The temperature change rate Tb is obtained by the following formula: ； In the formula, T(t) represents the temperature at time t, T(t-1) represents the temperature at time t-1, and Δt represents the time interval; The vibration frequency fv is obtained by the following formula: ； In the formula, represents the Fourier transform of the vibration signal; f(t) represents the vibration at time t, and argmax represents the maximum value of the parameter function of the function; The pressure fluctuation Pb is obtained by the following formula: ； Where N represents the total number of data points, P(ti) represents the pressure data of the i-th data point at time t; μP represents the mean value of pressure; The health status assessment and fault prediction unit analyzes the fault feature set GF according to the machine learning model, constructs an equipment health assessment model, and assesses the health status of the equipment; The equipment health assessment model will be trained based on the feature set GF of the equipment under different operating conditions; The training formula of the equipment health assessment model is as follows: ； In the formula, h(GF) represents the predicted equipment health status, and Model represents the machine learning model; Based on the trained equipment health assessment model, predict the equipment and obtain the equipment health assessment result H; The health assessment result H of the device is obtained by the following formula: ； Where H(t) represents the health status of the device at time t, h(GF(t)) represents the health status of the device predicted at time t based on the fault feature set GF, and Evaluate represents the evaluation function.

5. The edible oil production line control system based on the Internet of Things according to claim 4 is characterized in that: The fault prediction and maintenance suggestion module includes a fault prediction model building unit and a maintenance suggestion generating unit; The fault prediction model building unit trains a regression analysis model by using historical fault data and a fault feature set GF, and builds a fault prediction model by the regression analysis model to predict the fault occurrence probability Prf of the future fault risk of the equipment; The failure probability Prf is obtained by the following formula: ； In the formula, represents the intercept term of the fault prediction model, They represent the fault prediction model parameters of temperature change rate Tb, vibration frequency fv and pressure fluctuation Pb respectively, and Q represents the error term.

6. The edible oil production line control system based on the Internet of Things according to claim 5 is characterized in that: The maintenance suggestion generating unit calculates the future failure risk Prf(t) of the equipment according to the acquired failure occurrence probability Prf, and generates repair and maintenance suggestions for the equipment; Through the trained regression model, the future failure risk Prf(t) is calculated based on the failure probability Prf and the failure feature set GF; The future failure risk Prf(t) is obtained by the following formula: ； In the formula, represents the adjustment coefficient, Tb(t) represents the temperature change rate at time t, fv represents the vibration frequency at time t, and Pb represents the pressure fluctuation at time t; By comparing the acquired future failure risk Prf(t) with the preset failure risk threshold TPrf, the risk status of the equipment can be determined; The risk status of the device is obtained by matching: When the future failure risk Prf(t) is less than the failure risk threshold TPrf, it means that the equipment has no failure risk; When the future failure risk Prf(t) ≥ the failure risk threshold TPrf, it means that the equipment has a failure risk.

7. The edible oil production line control system based on the Internet of Things according to claim 6 is characterized in that: The dynamic maintenance plan adjustment module includes a health summary unit and a maintenance plan adjustment unit; The health summary unit determines the health status of the device based on the comparison between the acquired health assessment result H and the preset health threshold TrH; When the health assessment result H> the health threshold TrH, it means that the health status of the device is normal; When the health assessment result H ≤ the health threshold TrH, it means that the health status of the equipment is abnormal, and the maintenance plan is initiated, including preventive maintenance, repair suggestions and periodic maintenance.

8. The edible oil production line control system based on the Internet of Things according to claim 7 is characterized in that: The maintenance plan adjustment unit automatically adjusts the maintenance plan Mty according to the acquired health assessment result H and future failure risk Prf(t); The maintenance plan Mty is obtained by the following formula: ； Where Mpo represents the deferred maintenance plan, and Mad represents the advanced maintenance plan.

9. The edible oil production line control system based on the Internet of Things according to claim 1, characterized in that: The overall data visualization and reporting module includes a data visualization unit and a report generation and scheduling unit; The data visualization unit displays the equipment health status, failure risk and maintenance plan in the form of intuitive diagrams; The report generation and scheduling unit regularly generates health reports, maintenance logs and fault analysis reports based on real-time equipment data and equipment health status; Among them, the health report includes a summary of the health status of the equipment, temperature change rate Tb, vibration frequency fv and pressure fluctuation Pb, providing the overall health level of the equipment; The maintenance log includes the specific operation, time, maintenance content and execution personnel information of each maintenance; Fault analysis reports include generating equipment fault analysis reports based on historical fault data and current predictions.

10. The edible oil production line control method based on the Internet of Things is applied to the edible oil production line control system based on the Internet of Things according to any one of claims 1 to 9, characterized in that: The following steps are involved: Step 1: The data acquisition module collects the operating data of the equipment, including temperature T, vibration f and pressure p, by installing sensors on the production equipment, and fits them into the data operation set W; Step 2: The data preprocessing module cleans and standardizes the collected data set W to obtain the health data set WK; Step 3: The feature extraction and health status analysis module extracts features from the health data set WK, including the temperature change rate Tb, vibration frequency fv, and pressure fluctuation Pb, to form a fault feature set GF, and uses a machine learning model to analyze the relationship between the fault feature set GF and the equipment fault, build an equipment health assessment model, assess the equipment health status, and obtain the health assessment result H; Step 4: The fault prediction and maintenance suggestion module builds a fault prediction model based on historical fault data and fault feature set GF by using regression analysis algorithm to predict the equipment failure risk in the future and obtain the future failure risk Prf(t); Step 5: The dynamic maintenance plan adjustment module automatically adjusts the maintenance plan according to the health assessment result H and the future failure risk Prf(t); Step 6: The overall data visualization and reporting module visualizes the equipment health status, equipment failure risk and maintenance plan, including fixed-cycle health reports, maintenance logs and failure analysis reports.

Images