Abnormal operation monitoring method and system of tobacco production equipment, medium and terminal

Abstract

The invention discloses an abnormal operation monitoring method and system for tobacco production equipment, a medium and a terminal, and belongs to the technical field of tobaccos, and the method comprises the steps: constructing a communication Internet of Things model of a plurality of tobacco production equipment; calling a communication direction in the communication internet-of-things model in real time to mark the equipment operation data acquired from each tobacco production equipment to obtain a communication direction label; performing first screening operation on the equipment operation data according to an equipment operation limit value corresponding to the communication connection node to obtain screened equipment operation data; performing second screening operation on the screened equipment operation data and the communication direction label based on a support vector machine classification model after model training is completed to obtain target equipment operation data; and if the target equipment operation data contains the abnormal monitoring object in the equipment operation blacklist, determining that the target tobacco production equipment corresponding to the target equipment operation data operates abnormally. According to the invention, the problem of low monitoring effectiveness of tobacco product equipment is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of tobacco production, and more specifically, to a method, system, medium and terminal for monitoring abnormal operation of tobacco production equipment. Background Art

[0002] With the rapid advancement of information technology, especially the deep integration of big data processing and control technology, industrial enterprises are undergoing a profound digital transformation. In the tobacco industry, this trend is particularly significant because it not only concerns the improvement of production efficiency, but also directly affects multiple key links such as product quality, cost control and safe production. Among them, tobacco production equipment includes sub-equipment that realizes different process functions in the tobacco production process, for example, including transmission equipment, filtering equipment, packaging equipment, marking equipment, etc. In order to safely and effectively produce tobacco products, it is necessary to monitor the above multiple sub-equipment in real time.

[0003] At present, the monitoring of existing tobacco production equipment is usually based on manual inspections combined with the results of the products produced. Then, when the production situation of each tobacco production equipment is monitored based on the product results, the product has already been produced. At this time, it will greatly affect the output rate of the product, and it is impossible to ensure the normal operation of the monitored tobacco production equipment while not reducing the production quality of tobacco products. The goal of this greatly reduces the effectiveness of the monitoring of tobacco product equipment. Summary of the invention

[0004] An object of the present invention is to provide a method, system, medium and terminal for monitoring abnormal operation of tobacco production equipment, so as to solve the problem of low monitoring effectiveness of the aforementioned tobacco production equipment.

[0005] According to one aspect of the present invention, a method for monitoring abnormal operation of tobacco production equipment is provided, comprising the following steps:

[0006] Constructing a communication Internet of Things model for multiple tobacco production equipment, wherein the communication Internet of Things model includes communication directions and communication connection nodes of different tobacco production equipment;

[0007] When the tobacco production equipment is started for production, the communication direction in the communication Internet of Things model is retrieved in real time, and the equipment operation data collected from each of the tobacco production equipment is marked based on the communication direction to obtain a communication direction label;

[0008] performing a first screening operation on the device operation data according to the device operation limit corresponding to the communication connection node to obtain screened device operation data, where different communication connection nodes correspond to different device operation limit values;

[0009] Performing a second screening operation on the filtered device operation data and the communication direction label based on the support vector machine classification model for which model training has been completed, to obtain target device operation data;

[0010] If the target device operation data contains an abnormal monitoring object in the device operation blacklist, it is determined that the target tobacco production equipment corresponding to the target device operation data is operating abnormally.

[0011] Optionally, constructing a communication Internet of Things model for multiple tobacco production equipment includes:

[0012] Receiving communication instructions from each of the tobacco production equipment, and retrieving production process information between different tobacco production equipment;

[0013] Determine the communication direction according to the device identity carried in the communication instruction in combination with the production process information, and determine the communication connection node based on the device information of the tobacco production equipment and the production node in the production process information;

[0014] The communication Internet of Things model is constructed based on the communication connection node and the communication direction, and the communication Internet of Things model is displayed in a control interface.

[0015] Optionally, the real-time calling of the communication direction in the communication Internet of Things model to mark the equipment operation data collected from each of the tobacco production equipment to obtain the communication direction label includes:

[0016] Collecting the equipment operation data generated by the tobacco production equipment, the equipment operation data including operation speed, equipment temperature, operation quantity, and operation time;

[0017] The tag name matching the communication direction is retrieved, and the equipment operation data is marked based on the tag name to obtain the communication direction tag, wherein the tag name is generated based on the production sequence between the tobacco production equipment.

[0018] Optionally, performing a first screening operation on the device operation data according to the device operation limit corresponding to the communication connection node to obtain the screened device operation data includes:

[0019] Loading the equipment operation limit corresponding to each of the communication connection nodes, wherein the equipment operation limit is stored in an operation limit pool, wherein the operation limit pool stores equipment operation limit values ​​updated by different communication connection nodes according to time conditions, equipment conditions, and product conditions;

[0020] If the equipment operation limit value includes a maximum value and a minimum value, performing a first screening operation on the equipment operation data according to the maximum value and the minimum value to obtain screened equipment operation data;

[0021] If the equipment operation limit value includes a maximum value or a minimum value, a first screening operation is performed on the equipment operation data according to the maximum value or the minimum value to obtain screened equipment operation data.

[0022] Optionally, before the support vector machine classification model based on the completed model training performs a second screening operation on the filtered device operation data and the communication direction label to obtain the target device operation data, the method further includes:

[0023] Determine an adjustment parameter corresponding to a basic penalty parameter based on the number of devices parsed from the communication direction tag, and determine a target penalty parameter based on a sum and difference between the basic penalty parameter and the adjustment parameter;

[0024] Constructing a Gaussian radial basis function classifier and obtaining model training samples, wherein the model training samples include device operation samples with classification labels and communication direction label samples;

[0025] The Gaussian radial basis function classifier is trained based on the model training samples and the target penalty parameters to obtain the trained support vector machine classification model.

[0026] Optionally, the abnormal operation monitoring method of tobacco production equipment further includes:

[0027] If the target device operation data is a data type, the device operation data blacklist is retrieved, and it is queried whether the target device operation data is an abnormal monitoring data object in the device operation data blacklist, wherein the abnormal monitoring data object includes a data format abnormal object, a data content abnormal object, and a data collection time abnormal object.

[0028] Optionally, the abnormal operation monitoring method of tobacco production equipment further includes:

[0029] If the target device operation data is of image type, feature extraction is performed on the target device operation data to obtain operation features and device features, and a device operation feature blacklist is retrieved;

[0030] The operation feature and the device feature are queried to see whether they are abnormal monitoring image objects in the device operation feature blacklist, wherein the abnormal monitoring image objects include image color objects, image area objects, and image data stream objects.

[0031] Another aspect of the present invention provides an abnormal operation monitoring system for tobacco production equipment, including a main control server, several data acquisition devices;

[0032] The main control server is used to construct a communication Internet of Things model for multiple tobacco production equipment, wherein the communication Internet of Things model includes communication directions and communication connection nodes of different tobacco production equipment;

[0033] When the tobacco production equipment is started for production, the communication direction in the communication Internet of Things model is retrieved in real time to mark the equipment operation data collected from each of the tobacco production equipment to obtain a communication direction label;

[0034] performing a first screening operation on the device operation data according to the device operation limit corresponding to the communication connection node to obtain screened device operation data, where different communication connection nodes correspond to different device operation limit values;

[0035] Performing a second screening operation on the filtered device operation data and the communication direction label based on the support vector machine classification model for which model training has been completed, to obtain target device operation data;

[0036] If the target device operation data contains an abnormal monitoring object in the device operation blacklist, determining that the target tobacco production equipment corresponding to the target device operation data is operating abnormally;

[0037] The data acquisition device is used to collect equipment operation data from the tobacco production equipment.

[0038] According to another aspect of the present invention, a storage medium is provided, wherein at least one executable instruction is stored in the storage medium, and the executable instruction enables a processor to execute operations corresponding to the above-mentioned abnormal operation monitoring method of tobacco production equipment.

[0039] According to another aspect of the present invention, there is provided a terminal, comprising: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus;

[0040] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute operations corresponding to the abnormal operation monitoring method of the tobacco production equipment.

[0041] The abnormal operation monitoring method, system, medium and terminal of tobacco production equipment disclosed in the present invention have the following technical effects:

[0042] Compared with the prior art, the present invention constructs a communication Internet of Things model for multiple tobacco production equipment, wherein the communication Internet of Things model includes communication directions and communication connection nodes of different tobacco production equipment; when the tobacco production equipment is started for production, the communication direction in the communication Internet of Things model is retrieved in real time to mark the equipment operation data collected from each of the tobacco production equipment to obtain a communication direction label; a first screening operation is performed on the equipment operation data according to the equipment operation limit corresponding to the communication connection node to obtain screened equipment operation data; a second screening operation is performed on the screened equipment operation data and the communication direction label based on a support vector machine classification model that has completed model training to obtain target equipment operation data; if the target equipment operation data includes an abnormal monitoring object in the equipment operation blacklist, the target tobacco production equipment corresponding to the target equipment operation data is determined to be operating abnormally, thereby realizing real-time abnormal discrimination of abnormal operation in the production process of the tobacco production equipment, greatly increasing the timeliness of abnormal monitoring, reducing the impact of equipment operation abnormality on product quality and output rate, and at the same time ensuring the accuracy of abnormal monitoring and improving the effectiveness of abnormal monitoring of tobacco production equipment.

[0043] Further features and advantages of the present invention will become apparent from the following detailed description of exemplary embodiments of the present invention with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the invention and, together with the description, serve to explain the principles of the invention.

[0045] Figure 1 A schematic flow chart of a method for monitoring abnormal operation of tobacco production equipment provided by an embodiment of the present invention;

[0046] Figure 2 A flowchart of constructing a communication Internet of Things model for multiple tobacco production equipment in an embodiment of the present invention is provided;

[0047] Figure 3 A block diagram showing the composition of an abnormal operation monitoring system for tobacco production equipment provided by an embodiment of the present invention is shown;

[0048] Figure 4 A schematic diagram of the structure of a terminal provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0049] Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that the relative arrangement of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present invention unless otherwise specifically stated.

[0050] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the invention, its application, or uses.

[0051] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered as part of the specification.

[0052] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0053] The present invention proposes an embodiment of a method for monitoring abnormal operation of tobacco production equipment. Specifically, Figure 1 As shown, the method comprises the following steps:

[0054] 101. Build a communication Internet of Things model for multiple tobacco production equipment.

[0055] In an embodiment of the present invention, tobacco production equipment corresponding to different operating stations in the production process is configured on the tobacco production line, for example, preheating and cutting equipment, tobacco balancing equipment, cigarette forming equipment, filter equipment and sealing equipment, etc. There are mutual constraints and containment of operation data such as operation progress and operation speed between each tobacco production equipment. This containment can be the upstream equipment of the assembly line to the downstream equipment of the assembly line, or the downstream equipment of the assembly line to the upstream equipment of the assembly line. For example, the working speed of the cigarette forming equipment and the filter equipment needs to be kept relatively synchronized. After the sealing operation is confirmed to be completed, the downstream printing equipment can perform the coding operation. Therefore, each tobacco production equipment needs to use communication to exchange operation data and production information between each other. In the communication relationship, other tobacco production equipment that has a communication relationship with the current tobacco production equipment is the communication connection node of the current tobacco production equipment, and the sending direction of the communication data between the current tobacco production equipment and its communication connection node is the communication direction. In order to accurately monitor the entire process of tobacco production, it is necessary to collect the operating data of tobacco production equipment according to the communication direction and communication connection nodes, and then it is necessary to build a communication Internet of Things model that includes the communication directions and communication connection nodes of different tobacco production equipment.

[0056] 102. When the tobacco production equipment is started for production, the communication direction in the communication Internet of Things model is retrieved in real time to mark the equipment operation data collected from each of the tobacco production equipment to obtain a communication direction label.

[0057] In the embodiment of the present invention, the equipment operation data is collected based on the data acquisition devices (such as speed sensors, temperature sensors) and image acquisition devices (such as monitors, CCD cameras) configured in each tobacco production equipment or near the tobacco production equipment. The equipment operation data of different tobacco production equipment is acquired separately based on independent operation data acquisition devices, and there is a lack of correlation between the operation data of each equipment. However, in fact, due to the close relationship between tobacco production equipment, there is also a correlation between the production process of the corresponding equipment operation data. Therefore, it is necessary to associate the equipment operation data of each tobacco production equipment based on the communication direction. Specifically, the corresponding communication direction is matched from the communication Internet of Things model based on the tobacco production equipment corresponding to the operation data, and the corresponding equipment operation data is marked according to the communication direction to obtain a communication direction label.

[0058] It should be noted that through the construction of the communication Internet of Things model and the marking of equipment operation data based on the communication direction, it is possible to associate the equipment operation data of each tobacco production equipment based on the generated process flow, providing a more relevant screening basis for the subsequent screening of abnormal equipment operation data.

[0059] 103. Perform a first screening operation on the device operation data according to the device operation limit corresponding to the communication connection node to obtain screened device operation data, where different communication connection nodes correspond to different device operation limit values.

[0060] In the embodiment of the present invention, different communication connection nodes correspond to different tobacco production equipment, and different tobacco production equipment has different equipment operation limits. The equipment operation limit is used to characterize the extreme operation value of the corresponding tobacco production equipment, and may specifically include one or more of the operation speed limit, the equipment temperature limit, the operation quantity limit, and the operation time limit. The above equipment operation limit may be a fixed value, or a dynamic value updated in real time according to the correlation between the equipment and the specific production operation conditions. The equipment operation limit may include a maximum value and a minimum value, or may include one of the maximum value and the minimum value, which is specifically determined according to the operation characteristics of different tobacco production equipment. The embodiment of the present invention does not specifically limit the equipment operation limit. The equipment operation limit is used to extract abnormal equipment operation data from the equipment operation data of the corresponding communication connection node to obtain the filtered equipment operation data.

[0061] It should be noted that the equipment operation data is screened based on the equipment operation limit, so that abnormal equipment operation data can be extracted from the equipment operation control dimension.

[0062] 104. Perform a second screening operation on the filtered device operation data and the communication direction label based on the support vector machine classification model for which model training has been completed to obtain target device operation data.

[0063] In an embodiment of the present invention, the support vector machine classification model that has completed model training is a binary classification model based on equipment operation samples and communication direction label samples. The screened equipment operation data and its corresponding communication direction label are input into the support vector machine classification model that has completed model training for classification processing, so that the screened equipment operation data can be classified. After the classification of the equipment operation data is obtained, the equipment operation data with abnormal classification results is determined as the target equipment operation data. The target equipment operation data is the equipment operation data with abnormalities in both the equipment operation control dimension and the equipment operation control direction dimension.

[0064] It should be noted that since the equipment operation data is real-time fluctuating data, it will also change along with the equipment control adjustment. In addition, the data collection process of the data collection device will also be disturbed, causing fluctuations in the collected equipment operation data. Therefore, simply identifying anomalies in the equipment operation data based on the equipment operation limit is not enough to determine that the corresponding tobacco production equipment is operating abnormally. It is also necessary to further combine the communication direction label of the equipment operation data to achieve accurate extraction of abnormal equipment operation data.

[0065] 105. If the target device operation data contains an abnormal monitoring object in the device operation blacklist, it is determined that the target tobacco production equipment corresponding to the target device operation data is operating abnormally.

[0066] In the embodiment of the present invention, the abnormal monitoring objects in the equipment operation blacklist can be abnormal monitoring objects of data types, such as abnormal data content, abnormal data collection time, etc., or can be abnormal monitoring objects corresponding to image features, such as image color, objects in the image, etc., which are not specifically limited in the embodiment of the present invention. By extracting the abnormalities in the equipment operation data from different dimensions and matching the extracted target equipment operation data based on the abnormal monitoring objects, the abnormal operation status of the tobacco production equipment is accurately determined, and the equipment operation data is the process data of the production. After the abnormality is determined, timely intervention can be made to avoid affecting the production quality, thereby effectively improving the effectiveness of abnormal monitoring.

[0067] In an embodiment of the present invention, for further explanation and limitation, as Figure 2 As shown, step 101 constructs a communication Internet of Things model for multiple tobacco production equipment, including:

[0068] 201. Receive communication instructions from each of the tobacco production equipment, and retrieve production process information between different tobacco production equipment.

[0069] 202. Determine the communication direction according to the device identity carried in the communication instruction in combination with the generated process flow information, and determine the communication connection node based on the device information of the tobacco production equipment and the production node in the production process flow information.

[0070] 203. Construct the communication Internet of Things model based on the communication connection node and the communication direction, and output the communication Internet of Things model in a control interface.

[0071] In an embodiment of the present invention, a communication instruction is an instruction sent by a tobacco production device to other tobacco production devices associated with it in the production process, which is used to explain its current operating status to other tobacco production devices so that other tobacco production devices can control the production operation according to its communication instruction. Among them, the sending object of the communication instruction is preferably an adjacent tobacco production device of the current tobacco production device, such as, in the production process, the tobacco production device of the previous production node or the tobacco production device of the next production node of the production node where the current tobacco production device is located. In addition, the sending object of the communication instruction can also be a tobacco production device that is not adjacent to the current tobacco production device. For example, when product accumulation occurs in the packaging device, the packaging device can send communication instructions to multiple non-adjacent upstream tobacco production devices to alleviate the situation of product accumulation. The communication instruction carries the identity of the tobacco production device receiving the communication instruction (such as device code, device number, etc.). The direction of the tobacco production device currently sending the communication instruction and the device receiving the communication instruction in the production process information is the communication direction, which can be expressed as A device → B device, B device → C device, wherein B device is the communication connection node of A device, and C device is the communication connection node of B device.

[0072] In one embodiment of the present invention, for further explanation and limitation, the step of real-time calling the communication direction in the communication Internet of Things model to mark the equipment operation data collected from each of the tobacco production equipment to obtain the communication direction label includes:

[0073] Collecting the equipment operation data generated by the tobacco production equipment;

[0074] A tag name matching the communication direction is retrieved, and the device operation data is marked based on the tag name to obtain the communication direction tag.

[0075] In the embodiment of the present invention, since the communication direction in the communication Internet of Things model is constructed based on the communication instructions received in real time, it will be updated according to the different communication instructions received. Therefore, in order to improve the accuracy of the equipment operation data marking, it is necessary to call the communication direction in the communication Internet of Things model in real time. And call the tag name that matches the communication direction. The tag name is generated based on the production order between tobacco production equipment. For example, the production order of tobacco production equipment A in the communication direction to tobacco production equipment C is 1, then the tag name is AC-1; the production order of tobacco production equipment C in the communication direction is 3, then the tag name is AC-3. And the equipment operation data is marked based on this tag name. Among them, the equipment operation data includes one or more of the operating speed, equipment temperature, number of operations, and operating time.

[0076] In one embodiment of the present invention, for further explanation and limitation, the step performs a first screening operation on the device operation data according to the device operation limit corresponding to the communication connection node, and the screened device operation data includes:

[0077] Loading the equipment operation limit corresponding to each of the communication connection nodes;

[0078] If the equipment operation limit value includes a maximum value and a minimum value, performing a first screening operation on the equipment operation data according to the maximum value and the minimum value to obtain screened equipment operation data;

[0079] If the equipment operation limit value includes a maximum value or a minimum value, a first screening operation is performed on the equipment operation data according to the maximum value or the minimum value to obtain screened equipment operation data.

[0080] In the embodiment of the present invention, the equipment operation limit is stored in the operation limit pool, and the operation limit pool stores the equipment operation limit updated by different communication connection nodes according to time conditions, equipment conditions, and product conditions, that is, the equipment operation limit of each tobacco production equipment is updated in real time according to one or more of the time conditions, equipment conditions, and product conditions. Among them, the time condition can be determined based on the operation time in the equipment operation data of the associated equipment or global equipment of the communication connection node; the equipment condition can be determined based on the operation speed, equipment temperature, and operation quantity in the equipment operation data of the associated equipment or global equipment of the communication connection node; the product condition can be determined based on the material number of the tobacco product currently produced and the production quality of the tobacco product; of course, the time condition, equipment condition, and product condition can also be determined in combination with other tobacco production conditions, which are not specifically limited in the embodiment of the present invention.

[0081] It should be noted that, when the equipment operation limit includes both maximum and minimum values, the filtered equipment operation data is abnormal equipment operation data that does not satisfy the equipment operation interval formed by the maximum and minimum values. When the equipment operation limit includes only maximum values, the filtered equipment operation data is abnormal equipment operation data that is greater than the maximum value; when the equipment operation limit includes only minimum values, the filtered equipment operation data is abnormal equipment operation data that is less than the minimum value. Filtering equipment operation data based on equipment operation limits that are updated in real time according to current production conditions can improve the matching degree between the limits and the current equipment operation data, thereby improving the accuracy of screening.

[0082] In one embodiment of the present invention, for further explanation and limitation, the step performs a second screening operation on the filtered device operation data and the communication direction label based on the support vector machine classification model for which model training has been completed, and before obtaining the target device operation data, the method further includes:

[0083] Determine an adjustment parameter corresponding to a basic penalty parameter based on the number of devices parsed from the communication direction tag, and determine a target penalty parameter based on a sum and difference between the basic penalty parameter and the adjustment parameter;

[0084] Construct a Gaussian radial basis function classifier and obtain model training samples;

[0085] The Gaussian radial basis function classifier is trained based on the model training samples and the target penalty parameters to obtain the trained support vector machine classification model.

[0086] In the embodiment of the present invention, a Gaussian radial basis function classifier is constructed by using a Gaussian function as the kernel function of a support vector machine classification model. A model training sample is obtained, which includes a device operation sample and a communication direction label sample with a classification label. The classification label can be a normal label and an abnormal label, that is, the device operation sample and the communication direction label sample include both positive samples and negative samples.

[0087] Since the equipment operation data is not a linearly separable data set, a radial basis function-Gaussian kernel function is used to map the equipment operation data set to a high-dimensional space to achieve linear separability of the data. However, even if the Gaussian kernel function is introduced, the data in the obtained high-dimensional space cannot be completely linearly separable. Therefore, during the model training process, a penalty parameter is introduced to correct the degree of non-satisfaction of the linear separability constraint to obtain more accurate training results. Specifically, the Gaussian kernel function is set to K (x, z), the penalty parameter is set to C, and the classification decision function of the Gaussian radial basis function classifier is: (1); (2); The training samples are set as T = {(x1, y1); (x2, y2), ..., (xN, yN)}, where xi∈Rn, yi∈{+1, -1}, i=1, 2, ...N; is the constraint parameter; .

[0088] It should be noted that the penalty parameter in the embodiment of the present invention is a penalty parameter modified based on the number of devices in the communication direction tag. The initial penalty parameter is set to , the number of devices is , then the calculation formula of the target penalty parameter C is: (3); or (4); wherein, the application of formula (3) is based on the formula (4) according to the specific number of devices. For example, if the number of devices is less than the first preset threshold, then formula (3) is selected; if the number of devices is greater than the second preset threshold, then formula (4) is selected. wherein, the first preset threshold is less than or equal to the second preset threshold, and the specific threshold can be customized according to the actual application requirements, and the embodiment of the present invention does not make specific limitations. Since the penalty parameter is used to correct the degree to which the device operation data does not satisfy the linear distribution, and the degree to which the device operation data corresponding to different numbers of devices do not satisfy the linear distribution is different, therefore, by determining the adjustment parameter based on the number of devices, and adjusting the basic penalty parameter according to the adjustment parameter, the penalty parameter is dynamically adjusted with the number of samples, thereby improving the matching degree between the support vector machine classification model and the device operation data, thereby improving the classification accuracy.

[0089] In an embodiment of the present invention, for further explanation and limitation, the method further includes:

[0090] If the target device operation data is a data type, the device operation data blacklist is retrieved, and it is queried whether the target device operation data is an abnormal monitoring data object in the device operation data blacklist.

[0091] In the embodiment of the present invention, the abnormal monitoring data object includes a data format abnormal object, a data content abnormal object, and a data collection time abnormal object. Among them, the data format abnormal object is used to characterize the abnormal situation of the data format, for example, the time description format corresponding to the equipment operation data is abnormal, the order of year, month, day and hour, minute and second is disordered; the data precision is abnormal, the two decimal places of the data are missing; the data unit is wrong, etc. The data content abnormal object is used to characterize the situation that the content of the data is inaccurate, for example, the normal operating speed of the tobacco transmission equipment is in the range of 5-10, but the operating speed is 100, and there is an obvious mismatch in magnitude; the current time is 2023, but the time is displayed as 2018, and there is an obvious time content abnormality. The data collection time abnormal object is used to characterize the situation of abnormal data collection frequency, for example, the time interval between two data collections is set to 10s, but the time interval between two data collections is 50s, or 5 groups of data are collected every minute, but there are only 2 groups of data per minute. The abnormal monitoring data objects in the data blacklist can be constructed based on historical abnormal situations. In situations where the same anomalies are prone to recurring during equipment operation or data collection equipment, data blacklists can quickly capture data anomalies and make it easier to locate the causes and solutions based on historical processing experience, thereby improving operational efficiency while improving the accuracy of anomaly monitoring.

[0092] In an embodiment of the present invention, for further explanation and limitation, the method further includes:

[0093] If the target device operation data is of image type, feature extraction is performed on the target device operation data to obtain operation features and device features, and a device operation feature blacklist is retrieved;

[0094] Query whether the operation feature and the device feature are abnormal monitoring image objects in the device operation feature blacklist.

[0095] In the embodiment of the present invention, when the target device operation data is of image type, it is necessary to extract features from the image, and then perform feature matching between the extracted features and the features in the device operation feature blacklist. Among them, the feature extraction method can be a method of jointly using a target detection algorithm and a convolutional neural network, and the extracted features can be shape features, spatial relationship features, etc.; the feature matching process can be similarity matching, or it can be matched based on a local feature matching algorithm. For the feature extraction method and the feature matching method, the corresponding model and algorithm can be selected according to the actual application requirements, and the embodiment of the present invention does not make specific restrictions. Operation features and device features can be extracted from the image. The operation features are the features of tobacco products (such as whether the tobacco product has a logo pattern, whether the product packaging is intact, etc.), and the device features are the features of the device itself (such as whether the transmission device is damaged, whether there are parts missing, etc.). Abnormal monitoring image objects include image color objects, image area objects, and image data stream objects. Among them, the image color object can be both an operation feature and a device feature. For example, the logo color of the tobacco product is set to red, and vermicelli or white is an abnormal object. Another example is the color of the assembly line equipment, which was originally green, but becomes black due to wear. The black color in the green is an abnormal object. The image area object is used to characterize the abnormal content in the image area. For example, according to the setting, the number of tobacco that can be captured in the image area is 10, but there are 9 or 11 tobacco products in the image area, which is an abnormal situation. Under normal circumstances, the cigarette holder and the cigarette are connected as a standard whole, but the cigarette is twisted or the cigarette holder is detached, which is an abnormal situation. The image data stream object is used to characterize the abnormal image acquisition frequency and data missing abnormalities, for example, the acquisition frequency does not meet the set frequency, the data frame is missing, etc.

[0096] It should be noted that the equipment on the assembly line is configured with different colors, and the tobacco products also have printed marks of corresponding colors. Based on the color, the wear of the equipment and the abnormality of the product production quality can be accurately identified. The image area object can capture the tobacco products in the image area, monitor the quantity and shape of the products, and realize the monitoring of product quality from the production process. The image data stream object is for monitoring the data acquisition equipment to realize the monitoring of the data acquisition status of the equipment operation data.

[0097] The present invention provides a method for monitoring abnormal operation of tobacco production equipment. Compared with the prior art, the embodiment of the present invention constructs a communication Internet of Things model of multiple tobacco production equipment, wherein the communication Internet of Things model contains communication directions and communication connection nodes of different tobacco production equipment; when the tobacco production equipment is started for production, the communication direction in the communication Internet of Things model is retrieved in real time to mark the equipment operation data collected from each of the tobacco production equipment to obtain a communication direction label; a first screening operation is performed on the equipment operation data according to the equipment operation limit corresponding to the communication connection node to obtain screened equipment operation data, and different communication connection nodes correspond to different equipment operation limits; a second screening operation is performed on the screened equipment operation data and the communication direction label based on a support vector machine classification model for which model training has been completed to obtain target equipment operation data; if the target equipment operation data contains an abnormal monitoring object in an equipment operation blacklist, it is determined that the target tobacco production equipment corresponding to the target equipment operation data is in abnormal operation, thereby realizing real-time abnormal discrimination of abnormal operation of the production process of the tobacco production equipment, greatly increasing the timeliness of abnormal monitoring, reducing the influence of equipment operation abnormality on product quality and output rate, and at the same time ensuring the accuracy of abnormal monitoring, thereby improving the effectiveness of abnormal monitoring of tobacco production equipment.

[0098] Furthermore, as a response to the above Figure 1 In order to realize the method shown in the figure, the embodiment of the present invention also provides an abnormal operation monitoring system for tobacco production equipment, such as Figure 3 As shown, the system includes: a main control server, several data acquisition devices;

[0099] The main control server 31 is used to construct a communication Internet of Things model for multiple tobacco production equipment, wherein the communication Internet of Things model includes communication directions and communication connection nodes of different tobacco production equipment;

[0100] When the tobacco production equipment is started for production, the communication direction in the communication Internet of Things model is retrieved in real time to mark the equipment operation data collected from each of the tobacco production equipment to obtain a communication direction label;

[0101] performing a first screening operation on the device operation data according to the device operation limit corresponding to the communication connection node to obtain screened device operation data, where different communication connection nodes correspond to different device operation limit values;

[0102] Performing a second screening operation on the filtered device operation data and the communication direction label based on the support vector machine classification model for which model training has been completed, to obtain target device operation data;

[0103] If the target device operation data contains an abnormal monitoring object in the device operation blacklist, determining that the target tobacco production equipment corresponding to the target device operation data is operating abnormally;

[0104] The data collection device 32 is used to collect equipment operation data from the tobacco production equipment.

[0105] Furthermore, the main control server is also used to receive communication instructions from each of the tobacco production equipment, and retrieve production process information between different tobacco production equipment;

[0106] Determine the communication direction according to the device identity carried in the communication instruction in combination with the generated process flow information, and determine the communication connection node based on the device information of the tobacco production equipment and the production node in the production process flow information;

[0107] The communication Internet of Things model is constructed based on the communication connection node and the communication direction, and the communication Internet of Things model is output in a control interface.

[0108] Furthermore, the main control server is also used to collect the equipment operation data generated by the tobacco production equipment, and the equipment operation data includes operation speed, equipment temperature, operation quantity, and operation time;

[0109] The tag name matching the communication direction is retrieved, and the equipment operation data is marked based on the tag name to obtain the communication direction tag, wherein the tag name is generated based on the production sequence between the tobacco production equipment.

[0110] Furthermore, the main control server is further used to load the equipment operation limit corresponding to each of the communication connection nodes, the equipment operation limit is stored in an operation limit pool, and the operation limit pool stores equipment operation limit updated by different communication connection nodes according to time conditions, equipment conditions, and product conditions;

[0111] If the equipment operation limit value includes a maximum value and a minimum value, performing a first screening operation on the equipment operation data according to the maximum value and the minimum value to obtain screened equipment operation data;

[0112] If the equipment operation limit value includes a maximum value or a minimum value, a first screening operation is performed on the equipment operation data according to the maximum value or the minimum value to obtain screened equipment operation data.

[0113] Furthermore, the main control server is further used to determine an adjustment parameter corresponding to a basic penalty parameter based on the number of devices parsed from the communication direction tag, and determine a target penalty parameter based on a sum and difference between the basic penalty parameter and the adjustment parameter;

[0114] Constructing a Gaussian radial basis function classifier and obtaining model training samples, wherein the model training samples include device operation samples with classification labels and communication direction label samples;

[0115] The Gaussian radial basis function classifier is trained based on the model training samples and the target penalty parameters to obtain the trained support vector machine classification model.

[0116] Furthermore, the main control server is also used to retrieve the device operation data blacklist if the target device operation data is a data type, and to query whether the target device operation data is an abnormal monitoring data object in the device operation data blacklist, wherein the abnormal monitoring data object includes a data format abnormal object, a data content abnormal object, and a data collection time abnormal object.

[0117] Furthermore, the main control server is further configured to extract features from the target device operation data if the target device operation data is of an image type, obtain operation features and device features, and retrieve a device operation feature blacklist;

[0118] The operation feature and the device feature are queried to see whether they are abnormal monitoring image objects in the device operation feature blacklist, wherein the abnormal monitoring image objects include image color objects, image area objects, and image data stream objects.

[0119] The invention provides an abnormal operation monitoring system for tobacco production equipment. Compared with the prior art, the embodiment of the invention constructs a communication Internet of Things model of multiple tobacco production equipment, wherein the communication Internet of Things model contains communication directions and communication connection nodes of different tobacco production equipment; when the tobacco production equipment is started for production, the communication direction in the communication Internet of Things model is retrieved in real time to mark the equipment operation data collected from each of the tobacco production equipment to obtain a communication direction label; a first screening operation is performed on the equipment operation data according to the equipment operation limit corresponding to the communication connection node to obtain screened equipment operation data, and different communication connection nodes correspond to different equipment operation limits; a second screening operation is performed on the screened equipment operation data and the communication direction label based on a support vector machine classification model for which model training has been completed to obtain target equipment operation data; if the target equipment operation data contains an abnormal monitoring object in an equipment operation blacklist, it is determined that the target tobacco production equipment corresponding to the target equipment operation data is abnormally operating, thereby realizing real-time abnormal discrimination of abnormal operation of the production process of the tobacco production equipment, greatly increasing the timeliness of abnormal monitoring, reducing the influence of equipment operation abnormality on product quality and output rate, and at the same time ensuring the accuracy of abnormal monitoring, thereby improving the effectiveness of abnormal monitoring of tobacco production equipment.

[0120] According to one embodiment of the present invention, a storage medium is provided, wherein the storage medium stores at least one executable instruction, and the computer executable instruction can execute the abnormal operation monitoring method of tobacco production equipment in any of the above method embodiments.

[0121] Figure 4 A schematic diagram of the structure of a terminal provided according to an embodiment of the present invention is shown. The specific embodiment of the present invention does not limit the specific implementation of the terminal.

[0122] like Figure 4 As shown, the terminal may include: a processor (processor) 402 , a communication interface (Communications Interface) 404 , a memory (memory) 406 , and a communication bus 408 .

[0123] The processor 402 , the communication interface 404 , and the memory 406 communicate with each other via a communication bus 408 .

[0124] The communication interface 404 is used to communicate with other devices such as clients or other servers.

[0125] The processor 402 is used to execute the program 410, and specifically can execute the relevant steps in the above-mentioned embodiment of the abnormal operation monitoring method of tobacco production equipment.

[0126] Specifically, the program 410 may include program codes, which include computer operation instructions.

[0127] The processor 402 may be a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiment of the present invention. The one or more processors included in the terminal may be processors of the same type, such as one or more CPUs; or processors of different types, such as one or more CPUs and one or more ASICs.

[0128] The memory 406 is used to store the program 410. The memory 406 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0129] The program 410 may be specifically configured to enable the processor 402 to perform the following operations:

[0130] Constructing a communication Internet of Things model for multiple tobacco production equipment, wherein the communication Internet of Things model includes communication directions and communication connection nodes of different tobacco production equipment;

[0131] When the tobacco production equipment is started for production, the communication direction in the communication Internet of Things model is retrieved in real time to mark the equipment operation data collected from each of the tobacco production equipment to obtain a communication direction label;

[0132] performing a first screening operation on the device operation data according to the device operation limit corresponding to the communication connection node to obtain screened device operation data, where different communication connection nodes correspond to different device operation limit values;

[0133] Performing a second screening operation on the filtered device operation data and the communication direction label based on the support vector machine classification model for which model training has been completed, to obtain target device operation data;

[0134] If the target device operation data contains an abnormal monitoring object in the device operation blacklist, it is determined that the target tobacco production equipment corresponding to the target device operation data is operating abnormally.

[0135] The above describes in detail the structure, features and effects of the present invention based on the embodiments shown in the drawings, but the above is only a preferred embodiment of the present invention. It should be noted that the technical features involved in the above embodiments and their preferred methods can be reasonably combined and matched into a variety of equivalent schemes by those skilled in the art without departing from or changing the design ideas and technical effects of the present invention; therefore, the present invention is not limited to the scope of implementation shown in the drawings, and all changes made in accordance with the concept of the present invention, or modifications to equivalent embodiments with equivalent changes, which still do not exceed the spirit covered by the specification and drawings, should be within the protection scope of the present invention.

Claims

1. A method for monitoring abnormal operation of tobacco production equipment, characterized in that: The following steps are involved: Constructing a communication Internet of Things model for multiple tobacco production equipment, wherein the communication Internet of Things model includes communication directions and communication connection nodes of different tobacco production equipment; When the tobacco production equipment is started for production, the communication direction in the communication Internet of Things model is retrieved in real time, and the equipment operation data collected from each of the tobacco production equipment is marked based on the communication direction to obtain a communication direction label; performing a first screening operation on the device operation data according to the device operation limit corresponding to the communication connection node to obtain screened device operation data, where different communication connection nodes correspond to different device operation limit values; Performing a second screening operation on the filtered device operation data and the communication direction label based on the support vector machine classification model for which model training has been completed, to obtain target device operation data; If the target device operation data contains an abnormal monitoring object in the device operation blacklist, it is determined that the target tobacco production equipment corresponding to the target device operation data is operating abnormally.

2. The abnormal operation monitoring method of tobacco production equipment according to claim 1, characterized in that: The communication Internet of Things model for building multiple tobacco production equipment includes: Receiving communication instructions from each of the tobacco production equipment, and retrieving production process information between different tobacco production equipment; Determine the communication direction according to the device identity carried in the communication instruction in combination with the production process information, and determine the communication connection node based on the device information of the tobacco production equipment and the production node in the production process information; The communication Internet of Things model is constructed based on the communication connection node and the communication direction, and the communication Internet of Things model is displayed in a control interface.

3. The abnormal operation monitoring method of tobacco production equipment according to claim 2, characterized in that: The real-time retrieval of the communication direction in the communication Internet of Things model to mark the equipment operation data collected from each of the tobacco production equipment to obtain the communication direction label includes: Collecting the equipment operation data generated by the tobacco production equipment, the equipment operation data including operation speed, equipment temperature, operation quantity, and operation time; The tag name matching the communication direction is retrieved, and the equipment operation data is marked based on the tag name to obtain the communication direction tag, wherein the tag name is generated based on the production sequence between the tobacco production equipment.

4. The abnormal operation monitoring method of tobacco production equipment according to claim 1, characterized in that: The first filtering operation is performed on the device operation data according to the device operation limit corresponding to the communication connection node, and the filtered device operation data is obtained, including: Loading the equipment operation limit corresponding to each of the communication connection nodes, wherein the equipment operation limit is stored in an operation limit pool, wherein the operation limit pool stores equipment operation limit values ​​updated by different communication connection nodes according to time conditions, equipment conditions, and product conditions; If the equipment operation limit value includes a maximum value and a minimum value, performing a first screening operation on the equipment operation data according to the maximum value and the minimum value to obtain screened equipment operation data; If the equipment operation limit value includes a maximum value or a minimum value, a first screening operation is performed on the equipment operation data according to the maximum value or the minimum value to obtain screened equipment operation data.

5. The abnormal operation monitoring method of tobacco production equipment according to claim 1, characterized in that: Before the support vector machine classification model based on the completed model training performs a second screening operation on the filtered device operation data and the communication direction label to obtain the target device operation data, the method further includes: Determine an adjustment parameter corresponding to a basic penalty parameter based on the number of devices parsed from the communication direction tag, and determine a target penalty parameter based on a sum and difference between the basic penalty parameter and the adjustment parameter; Constructing a Gaussian radial basis function classifier and obtaining model training samples, wherein the model training samples include device operation samples with classification labels and communication direction label samples; The Gaussian radial basis function classifier is trained based on the model training samples and the target penalty parameters to obtain the trained support vector machine classification model.

6. The abnormal operation monitoring method of tobacco production equipment according to claim 1, characterized in that: Also includes: If the target device operation data is a data type, the device operation data blacklist is retrieved, and it is queried whether the target device operation data is an abnormal monitoring data object in the device operation data blacklist, wherein the abnormal monitoring data object includes a data format abnormal object, a data content abnormal object, and a data collection time abnormal object.

7. The abnormal operation monitoring method of tobacco production equipment according to claim 1, characterized in that: Also includes: If the target device operation data is of image type, feature extraction is performed on the target device operation data to obtain operation features and device features, and a device operation feature blacklist is retrieved; The operation feature and the device feature are queried to see whether they are abnormal monitoring image objects in the device operation feature blacklist, wherein the abnormal monitoring image objects include image color objects, image area objects, and image data stream objects.

8. An abnormal operation monitoring system for tobacco production equipment, characterized in that: Including main control server and several data acquisition devices; The main control server is used to construct a communication Internet of Things model for multiple tobacco production equipment, wherein the communication Internet of Things model includes communication directions and communication connection nodes of different tobacco production equipment; When the tobacco production equipment is started for production, the communication direction in the communication Internet of Things model is retrieved in real time to mark the equipment operation data collected from each of the tobacco production equipment to obtain a communication direction label; performing a first screening operation on the device operation data according to the device operation limit corresponding to the communication connection node to obtain screened device operation data, where different communication connection nodes correspond to different device operation limit values; Performing a second screening operation on the filtered device operation data and the communication direction label based on the support vector machine classification model for which model training has been completed, to obtain target device operation data; If the target device operation data contains an abnormal monitoring object in the device operation blacklist, determining that the target tobacco production equipment corresponding to the target device operation data is operating abnormally; The data acquisition device is used to collect equipment operation data from the tobacco production equipment.

9. A storage medium storing at least one executable instruction, wherein the executable instruction enables a processor to execute operations corresponding to the abnormal operation monitoring method for tobacco production equipment according to any one of claims 1 to 7.

10. A terminal, comprising: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute an operation corresponding to the abnormal operation monitoring method of tobacco production equipment according to any one of claims 1-7.

Images