A supervision and early warning system and method for tobacco leaf baking processing

Abstract

The present invention relates to the field of processing supervision technology, and discloses a supervision and early warning system and method for tobacco leaf baking processing, the system comprising: a monitoring module and an early warning unit, wherein the early warning unit comprises: an acquisition module, an evaluation module, a central control module and an early warning module. The present invention is responsible for real-time monitoring of the ventilation volume and ventilation temperature of each monitoring area through the monitoring module, and transmits these data to the acquisition module. The acquisition module obtains data difference values ​​in combination with historical data, uses the central control module to filter data and identifies data anomalies through a Gaussian mixture model, and then determines the abnormal monitoring area. The evaluation module calculates the early warning score based on the baking data of the abnormal area, and finally the early warning module determines whether to trigger the early warning information. This anomaly detection method based on data difference analysis and Gaussian mixture model can effectively identify potential risks and ensure the stability and safety of the baking process.

Description

Technical Field

[0001] The present invention relates to the technical field of processing supervision, and in particular to a supervision and early warning system and method for tobacco leaf baking processing. Background Art

[0002] In the process of tobacco leaf baking, temperature control accuracy is crucial to the quality of the final tobacco leaves. Tobacco leaf baking is a complex process that is highly dependent on factors such as temperature, humidity and ventilation. Especially in large-scale baking production, there are often differences in temperature and humidity in different areas. During the baking process, factors such as the humidity of the tobacco leaves, ventilation volume, and ventilation temperature will affect the heat distribution, causing the temperature in some areas to be too high or too low, thus affecting the flavor, aroma and quality of the tobacco leaves.

[0003] At present, the temperature control monitoring system in the tobacco baking process mostly relies on a single temperature and humidity sensor or a simple ventilation control method, which cannot monitor the dynamic changes of each baking area in real time and comprehensively. Due to the differences in humidity and ventilation conditions in different areas, the data of a single sensor may not reflect the actual situation of the whole situation, which may easily lead to imbalance of temperature control in some areas. In addition, the existing system lacks differentiated analysis based on historical data and real-time monitoring data, and cannot detect potential abnormal situations in time. It also lacks an effective early warning mechanism and cannot respond quickly when problems arise, thus affecting the accuracy and stability of temperature control in the entire baking process.

[0004] Therefore, it is urgent to invent a supervision and early warning technology for tobacco leaf baking processing to solve the problem that the temperature control monitoring system in the existing tobacco leaf baking process is unable to comprehensively and real-time monitor and analyze the dynamic changes of each baking area, resulting in temperature control imbalance and potential abnormalities cannot be discovered in time, lack of effective early warning mechanism, and thus affecting the temperature control accuracy and stability of the baking process. Summary of the invention

[0005] In view of this, the present invention proposes a supervision and early warning system and method for tobacco leaf baking processing, aiming to solve the problem that the temperature control monitoring system in the existing tobacco leaf baking process is unable to comprehensively and real-time monitor and analyze the dynamic changes of each baking area, resulting in temperature control imbalance and potential abnormalities cannot be discovered in time, lack of an effective early warning mechanism, and thus affecting the temperature control accuracy and stability of the baking process.

[0006] The present invention proposes a tobacco leaf baking process supervision and early warning system, comprising:

[0007] A plurality of monitoring modules are provided, and the plurality of monitoring modules are respectively configured in monitoring areas divided based on preset distances, and the monitoring modules are configured to monitor the ventilation volume, real-time temperature and ventilation temperature of the monitoring areas;

[0008] An early warning unit is electrically connected to each monitoring module, and includes: a collection module, an evaluation module, a central control module and an early warning module;

[0009] The acquisition module is configured to be electrically connected to each of the monitoring modules respectively, and the acquisition module is configured to collect baking data in each of the monitoring areas during baking and historical baking data of each of the monitoring areas, wherein the baking data includes the ventilation volume and ventilation temperature of each of the monitoring areas monitored by each sensor;

[0010] The central control module is electrically connected to the acquisition module, and the central control module is configured to obtain a data difference value based on the baking data in each of the monitoring areas and the baking data in adjacent historical time periods; the central control module is also configured to determine a data difference threshold of the baking data based on the historical baking data of each of the monitoring areas, and the central control module is also configured to filter each of the data difference values ​​based on the data difference values, and obtain a filtered data abnormal value based on a Gaussian mixture model, and determine an abnormal monitoring area based on the data abnormal value;

[0011] The evaluation module is electrically connected to the central control module, and the evaluation module is configured to obtain baking data of the abnormal monitoring area and determine a warning score of the abnormal monitoring area according to the baking data;

[0012] The early warning module is electrically connected to the evaluation module, and the early warning module is configured to determine whether to send early warning information according to the early warning score.

[0013] Furthermore, when the central control module determines the data difference threshold of the baking data according to the historical baking data of each monitoring area, it includes:

[0014] ;

[0015] Among them, Y represents the data difference threshold, Represents the mean difference of historical normal baking data, represents the standard deviation of the historical normal baking data, and k represents the adjustment factor, where:

[0016] ;

[0017] ;

[0018] ;

[0019] Among them, Wt represents the data difference value of historical normal baking data, represents the baking data of the sensor at time t, It represents the baking data of the monitoring area at time t-1, and N represents the historical normal baking data at N times.

[0020] Further, when the central control module screens the data difference values ​​according to the data difference values, it includes:

[0021] The relationship between the data difference value and the data difference threshold is used to eliminate small difference data:

[0022] When the data difference value is greater than the data difference threshold, the evaluation module retains the baking data corresponding to the data difference value;

[0023] When the data difference value is less than or equal to the data difference threshold, the evaluation module determines the baking data corresponding to the data difference value as small difference data and removes the data.

[0024] Furthermore, when the central control module obtains the outliers of the filtered data based on the Gaussian mixture model and determines the abnormal monitoring area according to the outliers of the data, it includes:

[0025] Calculate the outlier value of each filtered data based on the Gaussian mixture model:

[0026] ;

[0027] Among them, p(x|Θ) represents the outlier value of data x, K is the number of Gaussian components, π k Expressed as the mixing coefficient of the kth component, It is expressed as the probability density of data x belonging to a certain Gaussian component;

[0028] When the outlier value of the screened data is greater than 0.8, the data is determined to be the abnormal data, and the monitoring area corresponding to the abnormal data is identified.

[0029] Furthermore, when the evaluation module determines the early warning score of the abnormal monitoring area according to the baking data, it includes:

[0030] The central control module is further configured to obtain a mean ventilation volume of each normal monitoring area, and determine a warning score of the abnormal monitoring area according to a ventilation volume difference between the mean ventilation volume and the ventilation volume of the abnormal monitoring area:

[0031] The central control module is configured with a first preset ventilation volume difference and a second preset ventilation volume difference, and the central control module is further configured to determine the early warning score of the abnormal monitoring area according to the relationship between the ventilation volume difference and the first preset ventilation volume difference and the second preset ventilation volume difference:

[0032] When the ventilation volume difference is less than the first preset ventilation volume difference, the central control module determines that the warning score of the abnormal monitoring area is L1;

[0033] When the ventilation volume difference is greater than or equal to the first preset ventilation volume difference, and the ventilation volume difference is less than the second preset ventilation volume difference, the central control module determines that the early warning score of the abnormal monitoring area is L2;

[0034] When the ventilation volume difference is greater than or equal to the second preset ventilation volume difference, the central control module determines that the warning score of the abnormal monitoring area is L3;

[0035] Wherein, the first preset ventilation volume difference is smaller than the second preset ventilation volume difference, and L1<L2<L3.

[0036] Furthermore, when the evaluation module determines that the early warning score of the abnormal monitoring area is Li, i=1, 2, 3, it includes:

[0037] The evaluation module is further configured to obtain the moisture content and baking time of the tobacco leaf batch, and the evaluation module is further configured to determine the baking score according to the moisture content, baking time, and the temperature gradient data and ventilation temperature of adjacent baking periods of the abnormal monitoring area:

[0038] ;

[0039] Wherein, S is the baking score, W std is the preset tobacco leaf moisture content, W act is the actual moisture content of tobacco leaves, T std is the preset zone temperature, T act is the actual area temperature, G i is the temperature gradient data of the adjacent baking periods, G avg is the preset temperature gradient reference value, G max is the preset temperature change rate, V act is the ventilation temperature, V std is the preset ventilation temperature, α, β and γ are weight coefficients, and α, β and γ are not zero;

[0040] The evaluation module is further configured to determine whether to adjust the early warning score Li of the abnormal monitoring area according to the baking score.

[0041] Furthermore, when the evaluation module determines whether to adjust the early warning score Li of the abnormal monitoring area according to the baking score, it includes:

[0042] The evaluation module is further configured to determine whether to adjust the early warning score Li of the abnormal monitoring area according to the relationship between the baking score and a preset baking score pre-configured by the evaluation module:

[0043] When the baking score is lower than or equal to the preset baking score, the evaluation module determines not to adjust the early warning score Li of the abnormal monitoring area;

[0044] When the baking score is higher than the preset baking score, the evaluation module determines an adjustment coefficient according to the relationship between the baking score and the preset baking score, and adjusts the warning score according to the adjustment coefficient.

[0045] Furthermore, when the evaluation module determines the adjustment coefficient according to the relationship between the baking score and the preset baking score, it includes:

[0046] The evaluation module is further configured to obtain a score difference between the baking score and the preset baking score, and determine the adjustment coefficient according to a relationship between the score difference and a first preset score difference and a second preset score difference;

[0047] When the score difference is lower than or equal to the first preset score difference, the evaluation module determines the adjustment coefficient to be V1;

[0048] When the score difference is higher than the first preset score difference, and the score difference is lower than or equal to the second preset score difference, the evaluation module determines that the adjustment coefficient is V2;

[0049] When the score difference is higher than the second preset score difference, the evaluation module determines that the adjustment coefficient is V3;

[0050] The first preset score difference is smaller than the second preset score difference, and 1<V1<V2<V3<1.5.

[0051] Furthermore, when the early warning module determines whether to send early warning information according to the early warning score, it includes:

[0052] The early warning module is further configured to determine whether to send the early warning information according to the relationship between the early warning score and a preset early warning score configured by the early warning module:

[0053] When the warning score is less than the preset warning score, the warning module determines not to send the warning information;

[0054] When the warning score is greater than or equal to the preset warning score, the warning module determines to send the warning information.

[0055] Compared with the prior art, the beneficial effect of the present invention is that by setting multiple monitoring modules and dividing the baking area according to a preset distance, independent monitoring of each area in the tobacco leaf baking process can be achieved. This regional monitoring method can effectively reflect the dynamic changes of key parameters such as ventilation volume, temperature and humidity in different areas during the baking process. The independent data collection of each monitoring area ensures that the baking state of each area can be fully and accurately grasped, thereby avoiding the limitation of relying on only a single sensor in the traditional system, making the monitoring of temperature control accuracy more detailed and comprehensive. In addition, through the cooperation of the acquisition module and the central control module, not only can the baking data of each monitoring area be collected in real time, but also the historical data can be compared and analyzed with the real-time data. By calculating the data difference value, the central control module can identify the change trend of temperature and humidity in each area and determine whether there is an abnormal situation. This dynamic monitoring method based on the difference value can timely capture the slight changes that may occur in the temperature control process, which is more sensitive and accurate than the traditional temperature control monitoring method, and effectively avoids the baking quality problems caused by temperature imbalance or excessive moisture accumulation. An outlier screening algorithm based on a Gaussian mixture model can automatically screen out abnormal data from a large amount of monitoring data to ensure that only the most representative abnormal data is processed. The outliers after data screening can more accurately reflect the baking areas where there may be problems, and the abnormal areas are scored through the evaluation module. The early warning module determines whether it is necessary to issue an early warning message based on the scoring results. This intelligent early warning mechanism can respond to potential risks in the baking process in a timely manner and effectively improve the reliability and response speed of supervision. Finally, by combining monitoring data, difference analysis and anomaly detection, warnings can be issued in advance before temperature control imbalance or other potential problems occur. Baking process personnel can adjust relevant parameters in a timely manner based on the early warning information, avoiding tobacco quality problems caused by temperature fluctuations or excessive moisture, thereby ensuring the consistency and high quality of the final product.

[0056] On the other hand, the present application also provides a regulatory early warning method for tobacco leaf baking processing, comprising:

[0057] Monitoring areas are divided based on preset distances, and monitoring modules for monitoring ventilation volume and ventilation temperature are configured in each monitoring area;

[0058] Collecting baking data in each monitoring area during baking and historical baking data in each monitoring area; obtaining data difference values ​​according to the baking data in each monitoring area and the baking data in adjacent historical time periods;

[0059] Determine a data difference threshold of the baking data according to the historical baking data of each monitoring area, and filter each data difference value according to the data difference value;

[0060] Obtaining outliers of the filtered data based on a Gaussian mixture model, and determining an abnormal monitoring area according to the outliers of the data;

[0061] The baking data of the abnormal monitoring area is obtained, a warning score of the abnormal monitoring area is determined according to the baking data, and whether to send a warning message is determined according to the warning score.

[0062] It can be understood that the monitoring and early warning system and method for tobacco leaf baking processing in the above-mentioned embodiments of the present invention have the same beneficial effects and will not be described in detail. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0064] Figure 1 A functional block diagram of a tobacco leaf baking process supervision and early warning system provided by an embodiment of the present invention;

[0065] Figure 2 A flowchart of a supervision and early warning method for tobacco leaf baking processing provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0066] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, in the absence of conflict, the embodiments of the present invention and the features described in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0067] like Figure 1 As shown, in some embodiments of the present application, this embodiment provides a tobacco leaf baking process supervision and early warning system, including: a monitoring module and an early warning unit.

[0068] Specifically, a plurality of monitoring modules are provided and the plurality of monitoring modules are respectively configured in monitoring areas divided based on preset distances, and the monitoring modules are configured to monitor the ventilation volume, real-time temperature and ventilation temperature of the monitoring areas.

[0069] The early warning unit is electrically connected to each monitoring module, and the early warning unit includes: a collection module, an evaluation module, a central control module and an early warning module. The collection module is configured to be electrically connected to each monitoring module, and the collection module is configured to collect baking data in each monitoring area during baking and historical baking data of each monitoring area, wherein the baking data includes the ventilation volume and ventilation temperature of each monitoring area monitored by each sensor. The central control module is electrically connected to the collection module, and the central control module is configured to obtain data difference values ​​according to the baking data in each monitoring area and the baking data of adjacent historical periods. The central control module is also configured to determine the data difference threshold of the baking data according to the historical baking data of each monitoring area, and the central control module is also configured to filter each data difference value according to the data difference value, and obtain the filtered data abnormal value based on the Gaussian mixture model, and determine the abnormal monitoring area according to the data abnormal value. The evaluation module is electrically connected to the central control module, and the evaluation module is configured to obtain the baking data of the abnormal monitoring area, and determine the early warning score of the abnormal monitoring area according to the baking data. The early warning module is electrically connected to the evaluation module, and the early warning module is configured to determine whether to send an early warning message according to the early warning score.

[0070] It is understandable that the monitoring modules are arranged at various preset positions in the baking area, and each monitoring module is responsible for the ventilation volume and ventilation temperature detection of a specific monitoring area. By acquiring these data in real time, the environmental changes in each monitoring area can be accurately grasped, providing basic data for subsequent data analysis and early warning. The acquisition module collects the baking data of each area in real time through electrical connection with each monitoring module, and stores these data for comparison with historical data. The collected data not only includes the temperature and humidity information of the current baking state, but also compares these real-time data with the baking data of adjacent historical periods. Through this comparison, the dynamic changes of each monitoring area can be identified, and it can be determined whether temperature control fluctuations or humidity abnormalities have occurred. After data acquisition, the central control module analyzes the monitoring data based on the calculation of the difference value of baking data and historical data. The module obtains the difference value between the baking data of each monitoring area and the data of the adjacent period, and determines whether it exceeds the predetermined difference threshold based on these difference values. By screening the data difference value, the central control module can efficiently find abnormal data in large-scale data to ensure the accuracy and timeliness of system response. Gaussian mixture model (GMM) is the core algorithm in data screening. Through this model, it is possible to identify abnormal data from a large amount of data. The Gaussian mixture model uses probability density functions to model data, divides data into multiple categories, and extracts abnormal data that is significantly different from the normal fluctuation range. Based on this method, it is possible to accurately determine which areas in the baking process have abnormalities, and then determine the abnormal monitoring areas. This intelligent analysis method can effectively avoid the deviation of manual analysis in traditional methods and improve the accuracy of abnormal identification. Finally, the evaluation module analyzes the baking data in the abnormal monitoring area and assigns a warning score to each abnormal area according to the set algorithm. The evaluation score is based on the defined baking data features, such as temperature and humidity differences, ventilation volume and other indicators. These scores help determine which abnormal areas need timely intervention. The warning module determines whether to send a warning message based on the score of the evaluation module. Once the warning score of a certain area exceeds the threshold, an alarm will be issued to prompt the operator to adjust the temperature control parameters or take other measures.

[0071] Specifically, the central control module determines the data difference threshold of the baking data based on the historical baking data of each monitoring area, including:

[0072] .

[0073] Among them, Y represents the data difference threshold, Represents the mean difference of historical normal baking data, represents the standard deviation of the historical normal baking data, and k represents the adjustment factor, where:

[0074] .

[0075] .

[0076] .

[0077] Among them, Wt represents the data difference value of historical normal baking data, represents the baking data of the sensor at time t, It represents the baking data of the monitoring area at time t-1, and N represents the historical normal baking data at N times.

[0078] Specifically, when the central control module screens each data difference value according to the data difference value, it includes: determining the relationship between the data difference value and the data difference threshold value, and eliminating small difference data: when the data difference value is greater than the data difference threshold value, the evaluation module retains the baking data corresponding to the data difference value. When the data difference value is less than or equal to the data difference threshold value, the evaluation module determines the baking data corresponding to the data difference value as small difference data and eliminates it.

[0079] Specifically, the central control module obtains the abnormal value of the filtered data based on the Gaussian mixture model, and determines the abnormal monitoring area according to the data abnormal value, including: calculating the abnormal value of each filtered data based on the Gaussian mixture model:

[0080] .

[0081] Among them, p(x|Θ) represents the outlier value of data x, K is the number of Gaussian components, π k Expressed as the mixing coefficient of the kth component, It is expressed as the probability density of data x belonging to a certain Gaussian component. When the outlier value of the filtered data is greater than 0.8, the data is determined to be abnormal data, and the monitoring area corresponding to the abnormal data is identified.

[0082] It can be understood that the central control module sets a difference threshold Y by introducing an adjustment factor k based on the difference mean and difference standard deviation of historical normal baking data. This threshold is used to identify abnormal fluctuations in baking data in each monitoring area. Specifically, Y is defined by the mean and standard deviation of historical data to ensure that data within the normal fluctuation range is retained, and data that is significantly beyond the range is regarded as abnormal, so that the fluctuation range of baking data can be flexibly adjusted to ensure the stability of the temperature control system. Secondly, when screening data difference values, the central control module determines which data should be retained and which data should be eliminated based on the data difference threshold. When the data difference value exceeds the preset threshold, the evaluation module will retain the data and further analyze it, believing that it may be a normal temperature fluctuation or a reasonable change in the baking process. For data that is less than or equal to the threshold, it is considered that the fluctuation of these data is within the normal range, so it will be eliminated to avoid irrelevant data interfering with the subsequent analysis process. The next operation of the central control module is to calculate the outliers of the screened data based on the Gaussian mixture model. The Gaussian mixture model uses a probability density function to model the data, and by dividing the data into multiple Gaussian components, the probability density of each data point belonging to different components is calculated. This process can help determine the abnormality of data through statistical methods. When the outlier value of the filtered data point is higher than the set threshold (0.8), the system considers the data as abnormal data, and then identifies the monitoring area corresponding to the abnormal data, further ensuring the rapid location of temperature control anomalies. The outlier value calculation based on the Gaussian mixture model has strong robustness and can extract potential abnormal patterns from complex baking data. This method can not only cope with temperature and humidity fluctuations, but also automatically identify which areas of baking data have potential risks through accurate probability analysis, and further optimize the early warning mechanism. Finally, according to the abnormal data identified by the Gaussian mixture model, the abnormal monitoring area can be marked and warned. Through high-probability outliers, the early warning module can judge and send corresponding alarms to remind operators to take timely measures to make adjustments. This process ensures the accuracy and stability of temperature control during baking, avoids uneven temperature or excessive moisture caused by differences between regions, and thus improves the quality and efficiency of tobacco baking processing.

[0083] It can be seen that the central control module calculates the difference mean and standard deviation of historical normal baking data, and combines the adjustment factor k to determine the data difference threshold to ensure that the baking data can fluctuate within a reasonable range under different environmental conditions. This design enables the threshold to be flexibly adjusted when facing external factors such as different humidity and temperature, avoiding missed detection or false alarms due to overly strict threshold settings. Secondly, by screening the data difference values ​​and eliminating small difference data, the impact of interference signals can be effectively reduced. When the data difference value is large, it will be retained and further processed to ensure that changes with practical significance are monitored in a timely manner, while small fluctuations that do not affect the overall baking effect are excluded. This process enhances the noise filtering capability, improves the accuracy of abnormal monitoring, and reduces the interference of irrelevant data on subsequent analysis. When calculating the outlier value of the filtered data, the Gaussian mixture model is used to calculate the probability density of each filtered data point. This method can divide the data into multiple Gaussian components and accurately judge the abnormality of each data point. Through this method, the central control module can estimate whether each data point is abnormal based on the density of the Gaussian component, thereby accurately identifying the problem area in the baking process. This model is more sophisticated than the simple threshold method and can handle more complex and variable baking data. When the outlier value of the Gaussian mixture model is greater than 0.8, the data will be identified as abnormal and the corresponding monitoring area will be determined. This processing ensures high sensitivity to abnormal data and can respond in real time to potential problems in the baking process, such as uneven temperature and excessive moisture, so as to ensure that measures are taken in the shortest time to make adjustments and reduce quality fluctuations caused by baking errors. By introducing historical data difference analysis, Gaussian mixture model outlier value calculation and data screening mechanism, the temperature control situation in the baking process can be monitored more accurately and automatic abnormal warning can be realized. This method greatly improves the adaptability to different environments and working conditions, ensures the temperature control accuracy of the baking process, and improves the quality and production efficiency of tobacco leaf processing.

[0084] Specifically, when the evaluation module determines the early warning score of the abnormal monitoring area according to the baking data, it includes: the central control module is also configured to obtain the average ventilation volume of each normal monitoring area, and determine the early warning score of the abnormal monitoring area according to the ventilation volume difference between the average ventilation volume and the ventilation volume of the abnormal monitoring area: wherein the central control module is configured with a first preset ventilation volume difference and a second preset ventilation volume difference, and the central control module is also configured to determine the early warning score of the abnormal monitoring area according to the relationship between the ventilation volume difference and the first preset ventilation volume difference and the second preset ventilation volume difference: when the ventilation volume difference is less than the first preset ventilation volume difference, the central control module determines that the early warning score of the abnormal monitoring area is L1. When the ventilation volume difference is greater than or equal to the first preset ventilation volume difference, and the ventilation volume difference is less than the second preset ventilation volume difference, the central control module determines that the early warning score of the abnormal monitoring area is L2. When the ventilation volume difference is greater than or equal to the second preset ventilation volume difference, the central control module determines that the early warning score of the abnormal monitoring area is L3. wherein the first preset ventilation volume difference is less than the second preset ventilation volume difference, and L1<L2<L3.

[0085] Specifically, when the evaluation module determines that the early warning score of the abnormal monitoring area is Li, i=1,2,3, the evaluation module includes: the evaluation module is also configured to obtain the moisture content and baking time of the tobacco leaf batch, and the evaluation module is also configured to determine the baking score according to the moisture content, baking time, and the temperature gradient data and ventilation temperature of the adjacent baking periods of the abnormal monitoring area:

[0086] .

[0087] Among them, S is the baking score, W std is the preset tobacco leaf moisture content, W act is the actual moisture content of tobacco leaves, T std is the preset zone temperature, T act is the actual area temperature, G i is the temperature gradient data of adjacent baking periods, G avg is the preset temperature gradient reference value, G max is the preset temperature change rate, V act is the ventilation temperature, V std is the preset ventilation temperature, α, β and γ are weight coefficients, and α, β and γ are not zero. The evaluation module is also configured to determine whether to adjust the early warning score Li of the abnormal monitoring area according to the baking score.

[0088] Specifically, the temperature gradient data is the rate of change of temperature in the monitored area.

[0089] Specifically, when the evaluation module determines whether to adjust the early warning score Li of the abnormal monitoring area according to the baking score, it includes: the evaluation module is also configured to determine whether to adjust the early warning score Li of the abnormal monitoring area according to the relationship between the baking score and the preset baking score pre-configured by the evaluation module: when the baking score is lower than or equal to the preset baking score, the evaluation module determines not to adjust the early warning score Li of the abnormal monitoring area. When the baking score is higher than the preset baking score, the evaluation module determines the adjustment coefficient according to the relationship between the baking score and the preset baking score, and adjusts the early warning score according to the adjustment coefficient.

[0090] Specifically, when the evaluation module determines the adjustment coefficient according to the relationship between the baking score and the preset baking score, it includes: the evaluation module is also configured to obtain the score difference between the baking score and the preset baking score, and determine the adjustment coefficient according to the relationship between the score difference and the first preset score difference and the second preset score difference. When the score difference is lower than or equal to the first preset score difference, the evaluation module determines the adjustment coefficient to be V1. When the score difference is higher than the first preset score difference, and the score difference is lower than or equal to the second preset score difference, the evaluation module determines the adjustment coefficient to be V2. When the score difference is higher than the second preset score difference, the evaluation module determines the adjustment coefficient to be V3. Among them, the first preset score difference is less than the second preset score difference, and 1<V1<V2<V3<1.5.

[0091] It is understandable that the central control module determines the early warning score of the abnormal monitoring area by calculating the mean ventilation volume of each normal monitoring area. This process dynamically divides different early warning levels (L1, L2, L3) by comparing the relationship between the ventilation volume difference and the two preset ventilation volume differences. This mechanism quantifies the ventilation volume difference into levels in a simple and effective way, which helps to achieve rapid early warning, effectively reduces overly complex analysis operations, and makes the system more efficient and easy to understand. Secondly, when the ventilation volume difference is less than the first preset ventilation volume difference, the evaluation module sets its early warning score as L1; when the ventilation volume difference is between the two preset differences, the score is L2; ​​and when the difference is greater than or equal to the second preset ventilation volume difference, the score is L3. This scoring system is clear and intuitive, and can judge the degree of change in ventilation conditions based on actual monitoring data and quickly feedback to the early warning system. This multi-level early warning system provides a basis for control personnel to respond step by step, ensuring that corresponding measures can be taken according to the degree of urgency when ventilation conditions are abnormal. When further evaluating the abnormal monitoring area, the evaluation module will calculate the baking score based on multiple factors such as the moisture content of the tobacco leaves, baking time, and temperature gradient. Specifically, the evaluation module compares the actual moisture content of tobacco leaves with the preset standard moisture content, combined with temperature data and ventilation temperature differences to comprehensively analyze the quality of the baking process. This comprehensive scoring mechanism takes into account multiple factors, ensures the comprehensiveness and accuracy of the scoring, and avoids the errors that may be caused by a single data source. If the baking score is higher than the preset value, the evaluation module will further calculate the adjustment coefficient and adjust the early warning score of the abnormal monitoring area accordingly. This process is to refine the response mechanism of the abnormal area so that the early warning score can be finely adjusted according to the actual baking quality. The design of the score adjustment differentiates the scores into different levels based on the relationship between the preset score differences, so as to more accurately reflect the possible quality fluctuations in the actual baking process.

[0092] It can be seen that the abnormal monitoring area can be divided into three warning levels of L1, L2 and L3 by calculating the ventilation volume difference of the abnormal monitoring area through the central control module and comparing it with the two preset ventilation volume differences. This grading method can intuitively reflect the different degrees of ventilation abnormalities, which is convenient for operators to quickly take corresponding countermeasures. By reasonably defining the difference range, the changes in ventilation conditions can be accurately reflected and provide a basis for subsequent decision-making. Secondly, the evaluation module not only introduces multiple factors such as the moisture content of tobacco leaves, baking time, temperature gradient and ventilation temperature based on the ventilation volume, further improving the evaluation of the baking process. In this way, the baking quality can be comprehensively judged from multiple angles to avoid misjudgment caused by a single indicator. Through this multi-dimensional comprehensive evaluation, potential problems in the baking process can be identified more accurately. In addition, the evaluation module can dynamically adjust the warning score of the abnormal monitoring area based on the difference between the baking score and the preset score. When the baking score exceeds the preset standard, the warning score will be adjusted according to the adjustment coefficient. This flexible adjustment mechanism ensures that the needs of changes in the baking process can be responded to in real time, avoiding operational errors caused by overly rigid preset standards. Finally, the evaluation module determines the adjustment coefficient and adjusts the warning score based on the relationship between the score difference and the preset score difference. This approach not only improves the accuracy of the score adjustment, but also enables accurate score adjustments in different situations, avoiding excessive or insufficient adjustments. This refined score difference processing method helps ensure that the warning information under different baking conditions is more in line with actual needs.

[0093] Specifically, when the early warning module determines whether to send the early warning information according to the early warning score, it includes: the early warning module is also configured to determine whether to send the early warning information according to the relationship between the early warning score and the preset early warning score configured by the early warning module: when the early warning score is less than the preset early warning score, the early warning module determines not to send the early warning information. When the early warning score is greater than or equal to the preset early warning score, the early warning module determines to send the early warning information.

[0094] It is understandable that the early warning module determines whether it is necessary to trigger the early warning information by comparing the early warning score calculated in real time with the preset early warning score. This comparison method establishes a threshold mechanism. When the early warning score reaches or exceeds the preset standard, the early warning information will be actively issued to ensure timely response to potential risks or abnormal situations. Secondly, by making judgments based on the relationship between the early warning score and the preset score, this method can dynamically adjust the response behavior. When the early warning score is low, it is judged that no intervention is required, so the early warning information is not sent; when the score is high, the early warning is automatically triggered. This mechanism avoids the frequent sending of irrelevant information, effectively saves resources and improves the operating efficiency of the system. Finally, through the preset scoring criteria, the early warning module realizes flexible early warning trigger control. When the actual early warning score exceeds the preset range, the early warning process will be started according to this standard. This scoring-based control method enables the early warning module to adapt to different working environments and actual needs, ensure the accuracy and timeliness of early warning information, and further optimize the reliability of the monitoring system.

[0095] In the above embodiment, by setting multiple monitoring modules and dividing the baking area according to a preset distance, independent monitoring of each area during the tobacco leaf baking process can be achieved. This regional monitoring method can effectively reflect the dynamic changes of key parameters such as ventilation volume, temperature and humidity in different areas during the baking process. The independent data collection of each monitoring area ensures that the baking state of each area can be fully and accurately grasped, thereby avoiding the limitation of relying on only a single sensor in the traditional system, making the monitoring of temperature control accuracy more detailed and comprehensive. In addition, through the cooperation of the acquisition module and the central control module, not only can the baking data of each monitoring area be collected in real time, but also the historical data can be compared and analyzed with the real-time data. By calculating the data difference value, the central control module can identify the change trend of temperature and humidity in each area and determine whether there is an abnormal situation. This dynamic monitoring method based on the difference value can capture the slight changes that may occur in the temperature control process in time, which is more sensitive and accurate than the traditional temperature control monitoring method, and effectively avoids the baking quality problems caused by temperature imbalance or excessive accumulation of moisture. An outlier screening algorithm based on a Gaussian mixture model can automatically screen out abnormal data from a large amount of monitoring data through this algorithm to ensure that only the most representative abnormal data is processed. The outliers after data screening can more accurately reflect the baking areas where problems may exist, and the evaluation module can score the abnormal areas. The early warning module determines whether it is necessary to issue an early warning message based on the scoring results. This intelligent early warning mechanism can respond to potential risks in the baking process in a timely manner, effectively improving the reliability and response speed of supervision. Finally, by combining monitoring data, difference analysis and anomaly detection, warnings can be issued in advance before temperature control imbalance or other potential problems occur. Baking process personnel can adjust relevant parameters in a timely manner based on the early warning information, avoiding tobacco quality problems caused by temperature fluctuations or excessive moisture, thereby ensuring the consistency and high quality of the final product.

[0096] In another preferred embodiment based on the above embodiment, Figure 2 As shown, this embodiment provides a supervision and early warning method for tobacco leaf baking processing, comprising:

[0097] Step S100: dividing monitoring areas based on preset distances, and configuring a monitoring module for monitoring ventilation volume and ventilation temperature in each monitoring area.

[0098] Step S200: Collect the baking data in each monitoring area during baking and the historical baking data in each monitoring area. Obtain the data difference value according to the baking data in each monitoring area and the baking data in adjacent historical time periods.

[0099] Step S300: determining a data difference threshold of the baking data according to the historical baking data of each monitoring area, and screening each data difference value according to the data difference value.

[0100] Step S400: Obtain outliers of the filtered data based on a Gaussian mixture model, and determine an abnormal monitoring area according to the outliers of the data.

[0101] Step S500: Obtain baking data of the abnormal monitoring area, determine the warning score of the abnormal monitoring area according to the baking data, and determine whether to send warning information according to the warning score.

[0102] It can be understood that the monitoring and early warning system and method for tobacco leaf baking processing in the above-mentioned embodiments of the present invention have the same beneficial effects and will not be described in detail.

[0103] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0104] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0105] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A tobacco leaf baking process supervision and early warning system, characterized in that: include: A plurality of monitoring modules are provided, and the plurality of monitoring modules are respectively configured in monitoring areas divided based on preset distances, and the monitoring modules are configured to monitor the ventilation volume, real-time temperature and ventilation temperature of the monitoring areas; The early warning unit is electrically connected to each monitoring module, and the early warning unit includes: a collection module, an evaluation module, a central control module and an early warning module; The acquisition module is configured to be electrically connected to each of the monitoring modules respectively, and the acquisition module is configured to collect baking data in each of the monitoring areas during baking and historical baking data of each of the monitoring areas, wherein the baking data includes the ventilation volume and ventilation temperature of each of the monitoring areas monitored by each sensor; The central control module is electrically connected to the acquisition module, and the central control module is configured to obtain a data difference value based on the baking data in each of the monitoring areas and the baking data in adjacent historical time periods; the central control module is also configured to determine a data difference threshold of the baking data based on the historical baking data of each of the monitoring areas, and the central control module is also configured to filter each of the data difference values ​​based on the data difference values, and obtain a filtered data abnormal value based on a Gaussian mixture model, and determine an abnormal monitoring area based on the data abnormal value; The evaluation module is electrically connected to the central control module, and the evaluation module is configured to obtain baking data of the abnormal monitoring area and determine a warning score of the abnormal monitoring area according to the baking data; The early warning module is electrically connected to the evaluation module, and the early warning module is configured to determine whether to send early warning information according to the early warning score; When the central control module determines the data difference threshold of the baking data according to the historical baking data of each monitoring area, it includes: ; Among them, Y represents the data difference threshold, Represents the mean difference of historical normal baking data, represents the standard deviation of the historical normal baking data, and k represents the adjustment factor, where: ; ; ; Where Wt represents the data difference value of historical normal baking data, represents the baking data of the sensor at time t, It represents the baking data of the monitoring area at time t-1, and N represents the historical normal baking data at N times.

2. The tobacco leaf baking process monitoring and early warning system according to claim 1, characterized in that: When the central control module screens the data difference values ​​according to the data difference values, it includes: The relationship between the data difference value and the data difference threshold is used to eliminate small difference data: When the data difference value is greater than the data difference threshold, the evaluation module retains the baking data corresponding to the data difference value; When the data difference value is less than or equal to the data difference threshold, the evaluation module determines the baking data corresponding to the data difference value as small difference data and removes the data.

3. The tobacco leaf baking process monitoring and early warning system according to claim 2, characterized in that: When the central control module obtains the outliers of the filtered data based on the Gaussian mixture model and determines the abnormal monitoring area according to the outliers of the data, it includes: Calculate the outlier value of each filtered data based on the Gaussian mixture model: ; Among them, p(x|Θ) represents the outlier value of data x, K is the number of Gaussian components, π k Expressed as the mixing coefficient of the kth component, It is expressed as the probability density of data x belonging to a certain Gaussian component; When the outlier value of the filtered data is greater than 0.8, the data is determined to be abnormal data, and the monitoring area corresponding to the abnormal data is identified.

4. The tobacco leaf baking process monitoring and early warning system according to claim 1, characterized in that: When the evaluation module determines the early warning score of the abnormal monitoring area according to the baking data, it includes: The central control module is further configured to obtain a mean ventilation volume of each normal monitoring area, and determine a warning score of the abnormal monitoring area according to a ventilation volume difference between the mean ventilation volume and the ventilation volume of the abnormal monitoring area: The central control module is configured with a first preset ventilation volume difference and a second preset ventilation volume difference, and the central control module is further configured to determine the early warning score of the abnormal monitoring area according to the relationship between the ventilation volume difference and the first preset ventilation volume difference and the second preset ventilation volume difference: When the ventilation volume difference is less than the first preset ventilation volume difference, the central control module determines that the warning score of the abnormal monitoring area is L1; When the ventilation volume difference is greater than or equal to the first preset ventilation volume difference, and the ventilation volume difference is less than the second preset ventilation volume difference, the central control module determines that the early warning score of the abnormal monitoring area is L2; When the ventilation volume difference is greater than or equal to the second preset ventilation volume difference, the central control module determines that the warning score of the abnormal monitoring area is L3; Wherein, the first preset ventilation volume difference is smaller than the second preset ventilation volume difference, and L1<L2<L3.

5. The tobacco leaf baking process monitoring and early warning system according to claim 4, characterized in that: When the evaluation module determines that the early warning score of the abnormal monitoring area is Li, i=1, 2, 3, it includes: The evaluation module is further configured to obtain the moisture content and baking time of the tobacco leaf batch, and the evaluation module is further configured to determine the baking score according to the moisture content, baking time, and the temperature gradient data and ventilation temperature of adjacent baking periods of the abnormal monitoring area; Wherein, S is the baking score, W std is the preset tobacco leaf moisture content, W act is the actual moisture content of tobacco leaves, T std is the preset zone temperature, T act is the actual area temperature, G i is the temperature gradient data of the adjacent baking periods, G avg is the preset temperature gradient reference value, G max is the preset temperature change rate, V act is the ventilation temperature, V std is the preset ventilation temperature, α, β and γ are weight coefficients, and α, β and γ are not zero; The evaluation module is further configured to determine whether to adjust the early warning score Li of the abnormal monitoring area according to the baking score.

6. The tobacco leaf baking process monitoring and early warning system according to claim 5, characterized in that: When the evaluation module determines whether to adjust the early warning score Li of the abnormal monitoring area according to the baking score, it includes: The evaluation module is further configured to determine whether to adjust the early warning score Li of the abnormal monitoring area according to the relationship between the baking score and a preset baking score pre-configured by the evaluation module: When the baking score is lower than or equal to the preset baking score, the evaluation module determines not to adjust the early warning score Li of the abnormal monitoring area; When the baking score is higher than the preset baking score, the evaluation module determines an adjustment coefficient according to the relationship between the baking score and the preset baking score, and adjusts the warning score according to the adjustment coefficient.

7. The tobacco leaf baking process monitoring and early warning system according to claim 6, characterized in that: When the evaluation module determines the adjustment coefficient according to the relationship between the baking score and the preset baking score, it includes: The evaluation module is further configured to obtain a score difference between the baking score and the preset baking score, and determine the adjustment coefficient according to a relationship between the score difference and a first preset score difference and a second preset score difference; When the score difference is lower than or equal to the first preset score difference, the evaluation module determines the adjustment coefficient to be V1; When the score difference is higher than the first preset score difference, and the score difference is lower than or equal to the second preset score difference, the evaluation module determines that the adjustment coefficient is V2; When the score difference is higher than the second preset score difference, the evaluation module determines that the adjustment coefficient is V3; The first preset score difference is smaller than the second preset score difference, and 1<V1<V2<V3<1.

5.

8. The tobacco leaf baking process monitoring and early warning system according to claim 1, characterized in that: When the early warning module determines whether to send early warning information according to the early warning score, it includes: The early warning module is further configured to determine whether to send the early warning information according to the relationship between the early warning score and a preset early warning score configured by the early warning module: When the warning score is less than the preset warning score, the warning module determines not to send the warning information; When the warning score is greater than or equal to the preset warning score, the warning module determines to send the warning information.

9. A method for monitoring and early warning of tobacco leaf baking processing, applicable to a monitoring and early warning system for tobacco leaf baking processing as claimed in any one of claims 1 to 8, characterized in that: include: Monitoring areas are divided based on preset distances, and monitoring modules for monitoring ventilation volume and ventilation temperature are configured in each monitoring area; Collecting baking data in each monitoring area during baking and historical baking data in each monitoring area; obtaining data difference values ​​according to the baking data in each monitoring area and the baking data in adjacent historical time periods; Determine a data difference threshold of the baking data according to the historical baking data of each monitoring area, and filter each data difference value according to the data difference value; Obtaining outliers of the filtered data based on a Gaussian mixture model, and determining an abnormal monitoring area according to the outliers of the data; The baking data of the abnormal monitoring area is obtained, a warning score of the abnormal monitoring area is determined according to the baking data, and whether to send a warning message is determined according to the warning score.

Images