A sensor fault detection method, quality testing, model construction method, and device and medium

Abstract

The application provides a sensor fault detection method, quality testing, a model construction method, equipment and a medium. The detection method comprises the following steps: acquiring a cigarette packet yield value of a packaging unit within a preset time; inputting the cigarette packet yield value into a trained sensor quality detection model for analysis; and judging whether a target sensor of the packaging unit is faulty based on the analysis result. The sensor damage caused by manual disassembly in the physical inspection process is avoided, and unnecessary costs are generated. The sensor quality detection model is constructed to detect the sensor fault, which improves the detection accuracy and reduces the artificialization. In addition, when the sensor fails, an alarm information is sent to notify the relevant staff, which can realize timely and rapid repair and reduce the possibility of defective cigarette packets flowing into the market.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, and in particular to a sensor fault detection method, quality testing, model building method, equipment, and medium. Background Technology

[0002] As consumers become increasingly quality-conscious, they not only focus on the taste and smoking experience of tobacco products but also pay attention to the aesthetics and quality of the packaging. This places higher demands on the manufacturing capabilities and quality control of cigarette manufacturers. Therefore, tobacco packaging machines, as crucial equipment in cigarette packaging, are often equipped with quality inspection sensors to detect the quality of cigarette packaging at different processing stations. Over time and with continuous use, wear and tear on parts, aging of electronic components, or damage to wiring can cause sensor components to wear out or age due to prolonged use. Simultaneously, sensors may be affected by external factors such as temperature, humidity, chemicals, or physical impacts in the working environment, or suffer mechanical damage such as vibration, impact, or improper installation, all of which can damage sensor components or connections, leading to performance degradation or even failure.

[0003] Therefore, cigarette manufacturers regularly inspect the splicing joint quality detection sensors manually. However, in terms of operation, since the sensors are precision instruments, manual disassembly and assembly can easily damage them, resulting in unnecessary costs. In terms of detection efficiency, it is sometimes found that maintenance personnel can easily miss or misdetect the sensors. Furthermore, the process is time-consuming, has a small sample size, and is inconvenient to operate, which increases the risk of defective cigarette packs entering the market.

[0004] With the continuous advancement of digitalization in the tobacco industry, intelligent quality testing and diagnostic methods have become an urgent issue for the industry to address. Summary of the Invention

[0005] In view of the shortcomings of the prior art described above, the purpose of this disclosure is to provide an efficient and intelligent method, device and medium for sensor fault detection, quality testing and model building, and to solve the problems in the related technologies.

[0006] The first aspect of this disclosure provides a sensor fault detection method applied to sensor status detection of a packaging machine unit, comprising: acquiring the cigarette pack output value of the packaging machine unit within a preset time period; inputting the cigarette pack output value into a trained sensor quality detection model for analysis to obtain a predicted value of the cigarette pack output within the preset time period; and determining whether the target sensor of the packaging machine unit is faulty based on the predicted value of the cigarette pack output.

[0007] In an embodiment of the first aspect, obtaining the cigarette pack output value of the packaging unit within a preset time period includes: recording the cigarette pack output of the packaging unit at two adjacent moments when the target sensor is triggered to remove defective cigarette packs; and calculating the difference between the two adjacent cigarette pack outputs, wherein the difference is the cigarette pack output value within the preset time period.

[0008] In an embodiment of the first aspect, triggering the target sensor to reject defective cigarette packs includes: determining that the cigarette pack is a defective cigarette pack when the thickness of the aluminum foil exceeds a preset threshold, wherein the thickness of the aluminum foil is obtained by the target sensor.

[0009] In an embodiment of the first aspect, determining whether the target sensor of the packaging unit is faulty based on the predicted value of the cigarette pack output includes: calculating the approximate confidence interval of the predicted value; determining whether the cigarette pack output value falls within the approximate confidence interval of the predicted value; if the cigarette pack output value exceeds the approximate confidence interval of the predicted value, then determining that the target sensor of the packaging unit is faulty and issuing a fault alarm.

[0010] In an embodiment of the first aspect, the method further includes: setting a sensor monitoring alarm log for the packaging unit, wherein the information recorded in the monitoring alarm log includes: the time when the target sensor malfunction is determined and the time when the target sensor malfunction is resolved; wherein the time when the target sensor malfunction is resolved is the time when the cigarette pack output value of the packaging unit first falls into the approximate confidence interval after the target sensor malfunction is determined; and based on the target sensor monitoring alarm log, deriving trend statistics of the target sensor malfunction.

[0011] The second aspect of this disclosure discloses a sensor quality testing method, comprising: acquiring the cigarette pack output value of a packaging unit within a preset time period, and determining whether a target sensor of the packaging unit is faulty using the sensor fault detection method described in any embodiment of the first aspect; comparing the actual state of the target sensor with the expected state of the target sensor corresponding to the cigarette pack output value, and determining the test result.

[0012] In a second aspect embodiment, the method further includes: if the test result passes, storing the corresponding cigarette pack production value of the packaging unit within a preset time period.

[0013] This disclosure, in its third aspect, discloses a method for constructing a sensor quality detection model, comprising: acquiring multiple sets of cigarette pack output values ​​calculated based on each sensor from multiple packaging units as a training set, wherein the multiple sets of cigarette pack output values ​​calculated based on each sensor include: recording the cigarette pack output of the target packaging unit at two adjacent times when the packaging units remove defective cigarette packs; calculating the difference between the two adjacent cigarette pack outputs; labeling the cigarette pack output values ​​calculated based on each sensor in the training set; training an initial sensor quality detection model based on each sensor and its label in the training set; and forming the final sensor quality detection model.

[0014] This disclosure discloses a fourth aspect of an electronic device, the electronic device comprising: a processor and a memory; wherein the memory is used to store a computer program; the processor is used to execute the computer program stored in the memory to cause the electronic device to perform the sensor fault detection method according to any embodiment of the first aspect, or the sensor quality testing method according to any one of the second aspects, or the sensor quality detection model construction method according to the third aspect.

[0015] The fifth aspect of this disclosure discloses a computer-readable storage medium having a computer program stored thereon, which, when executed by an electronic device, implements the sensor fault detection method according to any embodiment of the first aspect, or the sensor quality testing method according to any claim of the second aspect, or the sensor quality testing model construction method according to an embodiment of the third aspect.

[0016] As described above, in this embodiment, the production volume of cigarette packs within a preset time period is analyzed using a sensor quality detection model to determine whether the sensor is faulty. This avoids sensor damage caused by manual disassembly during physical inspection, thus preventing unnecessary costs. The use of a sensor quality detection model improves detection accuracy and reduces manual intervention. Furthermore, issuing an alarm to relevant personnel when a sensor malfunctions enables timely and rapid repairs, reducing the likelihood of defective cigarette packs entering the market. Attached Figure Description

[0017] Figure 1 A flowchart illustrating a sensor fault detection method according to an embodiment of this disclosure is shown.

[0018] Figure 2 A schematic flowchart of a sensor quality testing method according to an embodiment of this disclosure is shown.

[0019] Figure 3 A flowchart illustrating a method for constructing a sensor quality detection model according to an embodiment of this disclosure is shown.

[0020] Figure 4A schematic diagram of the circuit structure of an electronic device according to an embodiment of the present disclosure is shown. Detailed Implementation

[0021] The following specific examples illustrate the implementation of this disclosure. Those skilled in the art can easily understand other advantages and effects of this disclosure from the information disclosed herein. This disclosure can also be implemented or applied through other different specific embodiments, and various details in this disclosure can be modified or changed according to different viewpoints and application modules without departing from the spirit of this disclosure. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of this disclosure can be combined with each other.

[0022] In this disclosure, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic represented in connection with that embodiment or example is included in at least one embodiment or example of this disclosure. Furthermore, the specific features, structures, materials, or characteristics represented may be combined in any suitable manner in any one or a group of embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples represented in this disclosure, as well as the features of those different embodiments or examples.

[0023] Furthermore, the terms "first" and "second" are used for illustrative purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the representation of this disclosure, "a set" means two or more, unless otherwise explicitly specified.

[0024] For the purpose of clarity, devices unrelated to the description are omitted, and the same or similar components throughout the specification are given the same reference numerals.

[0025] The technical terms used herein are for reference only to specific embodiments and are not intended to limit the scope of this disclosure. The singular form used herein includes the plural form unless the statement explicitly indicates otherwise. The word "comprising" as used in this specification means to specify a particular characteristic, region, integer, step, operation, element, and / or component, and does not exclude the presence or addition of other characteristics, regions, integers, steps, operations, elements, and / or components.

[0026] Although not explicitly defined, all terms, including technical and scientific terms used herein, shall have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains. Terms defined in commonly used dictionaries shall be further interpreted as having a meaning consistent with the relevant technical literature and the message of the present disclosure, and shall not be over-interpreted as having an ideal or overly formulaic meaning unless otherwise defined.

[0027] One of the packaging steps in cigarette pack production is rolling cigarettes with aluminum foil. The aluminum foil itself is spliced ​​manually, meaning used aluminum foil and new foil are glued together with double-sided tape to allow for alternating use of the left and right rolls. This splicing process creates paper joints. Furthermore, aluminum foil itself has joints during production. The thickness of a single sheet of aluminum foil is fixed; any significant variation in thickness is considered a joint. Modern tobacco machinery manufacturers and cigarette factories use sensors on packaging machines to measure the thickness of the aluminum foil in cigarette packs to identify joints, then reject packs with joints at the rejection point. Therefore, if the sensor measuring the aluminum foil thickness malfunctions, quality control during the cigarette rolling process becomes impossible, and cigarette packs containing aluminum foil joints risk entering the market.

[0028] Therefore, fault detection of sensors used to measure the thickness of aluminum foil in cigarette packs is crucial. However, manual periodic inspections of the sensors cannot detect faults in a timely manner; moreover, the sensors are precision components, and manual disassembly during physical inspection can damage them. Therefore, this application proposes an intelligent method capable of real-time monitoring of sensor faults.

[0029] The technical solutions of the present invention will now be described in detail with reference to the accompanying drawings.

[0030] like Figure 1 The diagram shows a flowchart of a sensor fault detection method according to an embodiment of the present disclosure, which includes steps S11-S13.

[0031] Step S11: Obtain the cigarette pack output value of the packaging unit within a preset time.

[0032] In some embodiments, obtaining the cigarette pack output value of the packaging unit within a preset time period includes: recording the cigarette pack output of the packaging unit at two adjacent moments when the target sensor is triggered to remove defective cigarette packs; and calculating the difference between the two adjacent cigarette pack outputs, wherein the difference is the cigarette pack output value within the preset time period.

[0033] In some embodiments, triggering the target sensor to reject defective cigarette packs includes: determining that the cigarette pack is a defective cigarette pack when the thickness of the aluminum foil exceeds a preset threshold, wherein the thickness of the aluminum foil is obtained by the target sensor.

[0034] For example, taking the GDX2 packaging machine as an example, when the quality detection sensor of the splicing joint of the aluminum foil paper triggers the rejection of cigarette packs, the output of the packaging machine corresponding to the rejection time is read and recorded. Under the premise that the quality detection sensor of the splicing joint of the aluminum foil paper is working normally, the difference in output between two adjacent rejection times is calculated.

[0035] Step S12: Input the cigarette pack production value into the trained sensor quality detection model for analysis to obtain the predicted value of the cigarette pack production within the preset time period.

[0036] Step S13: Determine whether the target sensor of the packaging unit is faulty based on the predicted value of the cigarette pack output.

[0037] In some implementations, determining whether the target sensor of the packaging unit is faulty based on the predicted value of the cigarette pack output includes: calculating the approximate confidence interval of the predicted value; determining whether the cigarette pack output value falls within the approximate confidence interval of the predicted value; if the cigarette pack output value exceeds the approximate confidence interval of the predicted value, then determining that the target sensor of the packaging unit is faulty and issuing a fault alarm.

[0038] For example, to verify whether a target sensor is faulty, an anomaly extraction method is used, with a 95% confidence interval as the anomaly selection criterion. The confidence interval is defined by the upper and lower confidence limits of the statistic. For a given set of sample data with mean μ and standard deviation σ, the 100(1-α)% confidence interval of the overall data mean is (μ-Z). α / 2 σ,μ+Z α / 2 σ), where α is the area covered by the non-confidence level within the normal distribution, Z α / 2 This corresponds to the standard score. Specifically, in the sensor quality detection model, μ is the average value of the training set during training; σ is the standard deviation of the training set; and α is the significance level, which is usually set to 0.05, with a corresponding confidence level of (1-α) = 95%. When the sensor status is determined to be normal, the corresponding cigarette pack production is updated in the database, enabling the sensor quality detection model to self-update.

[0039] The technical solutions described above enable real-time monitoring of the sensor status of the packaging unit, thereby ensuring that relevant personnel can promptly repair the sensor after it malfunctions, and improving the accuracy of removing defective cigarette packs containing aluminum foil joints.

[0040] In some implementations, the method further includes: setting a sensor monitoring alarm log for the packaging unit, wherein the information recorded in the monitoring alarm log includes: the time when the target sensor malfunction is determined and the time when the target sensor malfunction is resolved; wherein, the time when the target sensor malfunction is resolved is the time when the cigarette pack output value of the packaging unit first falls into the approximate confidence interval after the target sensor malfunction is determined; and based on the target sensor monitoring alarm log, deriving trend statistics of the target sensor malfunction.

[0041] As a specific example, when the cigarette pack output value exceeds the approximate confidence range of the predicted value, the monitoring alarm log is checked for any alarm records indicating sensor malfunction. If none are found, a new alarm record for the sensor is added, and the alarm record creation time is set to the current time. An alarm notification is then displayed on the HMI interface connected to the sensor of the packaging unit. After maintenance by staff, when the cigarette pack output value of the packaging unit falls back into the approximate confidence range for the first time, the alarm notification is cleared on the HMI interface, and the time of alarm clearing and the corresponding packaging machine output value are updated in the monitoring alarm log.

[0042] In addition, the sensor's monitoring and alarm logs are available for staff to query, enabling the tracing of historical fault data. Furthermore, based on the fault occurrence and resolution times, the number of failures and trends of the target sensor can be statistically analyzed, thereby identifying the causes of sensor failures.

[0043] like Figure 2 As shown, a sensor quality testing method according to an embodiment of the present disclosure is illustrated, including steps S21-S22.

[0044] Step S21: Obtain the cigarette pack production value of the packaging unit within a preset time, and determine whether the target sensor of the packaging unit is faulty by the sensor fault detection method described in any one of the first aspects.

[0045] Specifically, when the target sensor detects that the thickness of the aluminum foil in a cigarette pack exceeds a preset threshold, it indicates that the cigarette pack is packaged with aluminum foil containing spliced ​​joints, and the pack is determined to be defective. At this point, the sensor is triggered to remove the defective pack. The packaging unit displays real-time production data and records the cigarette pack production at that time. Similarly, the real-time production of the packaging unit is recorded when the sensor is triggered to remove a defective pack, and the difference between the two adjacent production values ​​is calculated to obtain the actual output value of the packaging unit during that period. It is understandable that if the sensor malfunctions, it will trigger the removal of cigarette packs too early or too late, affecting the time interval for calculating the cigarette pack output value. With a fixed packaging machine production capacity, the output value is positively correlated with the length of the time interval. Therefore, it is possible to indirectly determine whether the target sensor is malfunctioning by using the difference in packaging machine output values ​​without manual physical inspection of the sensor.

[0046] Step S22: Compare the actual state of the target sensor with the expected state of the target sensor corresponding to the tobacco pack production value to determine the test result.

[0047] Specifically, cigarette packs typically use aluminum foil of the same specifications, so the time intervals for triggering the rejection of defective packs are also roughly the same. Therefore, the sensor quality detection model can be used to predict the output of the packaging unit. When testing the accuracy of the sensor quality detection model, the difference in output between two consecutive triggers of rejecting defective packs is compared with the predicted value. If the output difference falls within the error range of the predicted value, the sensor quality detection model passes the test.

[0048] In some implementations, the sensor quality testing method further includes: storing the corresponding cigarette pack production value of the packaging unit within a preset time when the test result passes, so as to continuously update the model and realize the archiving of production value after the test case execution fails and subsequent problem tracing.

[0049] like Figure 3 As shown, a method for constructing a sensor quality detection model according to an embodiment of the present disclosure is illustrated, including steps S31-S34.

[0050] Step S31: Obtain multiple sets of cigarette pack output values ​​calculated based on each sensor from multiple packaging units as a training set. The multiple sets of cigarette pack output values ​​calculated based on each sensor include: recording the cigarette pack output of the target packaging unit at two adjacent moments when the packaging unit removes defective cigarette packs; and calculating the difference between the two adjacent cigarette pack outputs.

[0051] Specifically, in the cigarette pack production process, multiple packaging units operate simultaneously, and each unit is equipped with multiple sensors to detect the packaging quality of cigarettes at different processing stations. Therefore, due to differences in the calculation mechanisms of the sensors in each packaging unit, the measured output values ​​of the cigarette packs will vary slightly. Thus, it is necessary to build an independent sensor quality detection model for each sensor, and correspondingly, each sensor quality detection model has its own training set.

[0052] Step S32: Label the tobacco package output value calculated based on each sensor in the training set.

[0053] Specifically, during the model building process, the real-time output data of each packaging machine in the training set is labeled and stored and categorized according to the different sensors on different packaging machine groups.

[0054] Step S33: Train the initial sensor quality detection model based on each sensor and its label in the training set.

[0055] Specifically, the training set for each sensor quality detection model is obtained, and the model is trained until it meets the preset accuracy requirements. In some embodiments, the generalization mean square error Mse is used as the calculation standard. When Mse is less than a threshold, the sensor quality detection model meets the accuracy requirements and has high accuracy.

[0056] Step S34: Form the final sensor quality detection model.

[0057] Through the above embodiments, the classification of different sensor types in the packaging unit is marked, which enables the prediction of cigarette pack output value through each different sensor and the determination of whether the sensor is faulty, thereby improving the accuracy of sensor fault identification.

[0058] like Figure 4 The diagram shown illustrates the structure of an electronic device according to an embodiment of this disclosure.

[0059] The electronic device can execute computer program instructions to perform tasks such as Figure 1 or Figure 2 or Figure 3 The method described herein. Exemplarily, the electronic device may be a distributed computing node system, server group / server, desktop computer, laptop computer, etc., for running, such as... Figure 1 The detection method described herein determines whether the target sensor is faulty based on the results. Alternatively, the electronic device may be an onboard controller, or a cloud-based server / server group, distributed computing node system, etc., that communicates remotely with a local terminal, and executes... Figure 2The test method for the sensor quality detection model is to determine whether the sensor quality detection model has completed training based on the difference between the predicted output value and the actual operating output value.

[0060] The electronic device 40 includes a bus 41, a processor 42, and a memory 43. The processor 42 and the memory 43 can communicate via the bus 41. The memory 43 can store program instructions. The processor 42 implements the method steps in the previous embodiments by running the program instructions stored in the memory 43, such as... Figure 1 or Figure 2 or Figure 3 The method in the middle.

[0061] Bus 41 can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, although only one thick line is used in the diagram, this does not indicate that there is only one bus or one type of bus.

[0062] In some embodiments, processor 42 may be implemented as a central processing unit (CPU), microprocessor unit (MCU), system on chip (System on Chip), or field-programmable array (FPGA). Memory 43 may include volatile memory for temporary data storage during program execution, such as random access memory (RAM).

[0063] The memory 43 may also include non-volatile memory for data storage, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid-state disk (SSD).

[0064] In some embodiments, the electronic device 40 may further include a communicator 44. The communicator 44 is used for communication with the outside world. In specific examples, the communicator 44 may include one or more wired and / or wireless communication circuit modules. For example, the communicator 44 may include one or more of the following: a wired network card, a USB module, a serial interface module, etc. The wireless communication protocols followed by the wireless communication module include, for example, Nearfield Communication (NFC) technology, Infrared (IR) technology, Global System for Mobile Communications (GSM), General Packet Radio Service (GPRS), Code Division Multiple Access (CDMA), Wideband Code Division Multiple Access (WCDMA), Time-Division Code Division Multiple Access (TD-SCDMA), Long Term Evolution (LTE), Bluetooth (BT), Global Navigation Satellite System (GNSS), etc.

[0065] This disclosure also provides a computer-readable storage medium, characterized in that it stores program instructions, which are executed, for example... Figure 1 The aforementioned sensor fault detection method, or Figure 2 The aforementioned sensor fault testing method, or Figure 3 The method for constructing the sensor quality detection model.

[0066] That is, the method steps in the above embodiments are implemented as software or computer code that can be stored in a recording medium (such as CD ROM, RAM, floppy disk, hard disk or magneto-optical disk), or implemented as computer code that is originally stored in a remote recording medium or a non-transitory machine-readable medium and will be stored in a local recording medium after being downloaded via a network, so that the method represented herein can be stored in such software processing on a recording medium using a general-purpose computer, a special processor or programmable or special hardware (such as ASIC or FPGA).

[0067] In summary, this disclosure provides a sensor fault detection method, quality testing, model building method, equipment, and medium. By analyzing the cigarette pack production value within a preset time period using a sensor quality detection model, it determines whether the sensor is faulty, avoiding sensor damage caused by manual disassembly during physical inspection and the resulting unnecessary costs. Using a sensor quality detection model improves detection accuracy and reduces manual intervention. Furthermore, issuing alarm messages to relevant personnel when a sensor malfunctions enables timely and rapid repairs, reducing the possibility of defective cigarette packs entering the market.

[0068] The above embodiments are merely illustrative of the principles and effects of this disclosure and are not intended to limit this disclosure. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of this disclosure. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in this disclosure should still be covered by the claims of this disclosure.

Claims

1. A sensor fault detection method, applied to sensor status detection in a packaging machine unit, characterized in that, include: Obtaining the cigarette pack output value of the packaging unit within a preset time period includes: recording the cigarette pack output of the packaging unit at two consecutive moments when the target sensor is triggered to remove defective cigarette packs; calculating the difference between the two consecutive cigarette pack outputs, where the difference is the cigarette pack output value within the preset time period; triggering the target sensor to remove defective cigarette packs includes: determining the cigarette pack as defective when the thickness of the aluminum foil exceeds a preset threshold, wherein the thickness of the aluminum foil is obtained by the target sensor; constructing an independent sensor quality detection model for each sensor, and correspondingly, each sensor quality detection model has its own training set; The cigarette pack production value is input into a trained sensor quality detection model for analysis to obtain a predicted value of the cigarette pack production within the preset time period; Determining whether the target sensor of the packaging unit is faulty based on the predicted value of the cigarette pack output includes: calculating the approximate confidence interval of the predicted value; determining whether the cigarette pack output value falls within the approximate confidence interval of the predicted value; if the cigarette pack output value exceeds the approximate confidence interval of the predicted value, then determining that the target sensor of the packaging unit is faulty and issuing a fault alarm; when the sensor status is determined to be normal, updating the corresponding cigarette pack output to the database, so that the sensor quality detection model can be self-updated.

2. The sensor fault detection method as described in claim 1, characterized in that, Also includes: The packaging unit is configured to set a sensor monitoring alarm log, and the information recorded in the monitoring alarm log includes: the time when the target sensor was determined to be faulty and the time when the target sensor fault was resolved; The time when the target sensor fault is resolved is the time when the cigarette pack output value of the packaging unit first falls into the approximate confidence interval after the target sensor fault is determined. Based on the target sensor monitoring alarm logs, the trend statistics of the target sensor failures are obtained.

3. A sensor quality testing method, characterized in that, include: The output value of cigarette packs of the packaging unit within a preset time is obtained, and the target sensor of the packaging unit is determined to be faulty by the sensor fault detection method described in claim 1 or 2. The test result is determined by comparing the actual state of the target sensor with the expected state of the target sensor corresponding to the tobacco pack production value. If the test results are satisfactory, the corresponding cigarette pack production value of the packaging unit within a preset time period is stored.

4. A method for constructing a sensor quality detection model, characterized in that, include: Multiple sets of cigarette pack output values ​​calculated based on each sensor from multiple packaging units are obtained as a training set. The multiple sets of cigarette pack output values ​​calculated based on each sensor include: recording the cigarette pack output of the target packaging unit at two adjacent moments when the packaging unit removes defective cigarette packs; calculating the difference between the two adjacent cigarette pack outputs; and constructing an independent sensor quality detection model for each sensor, with each sensor quality detection model having its own training set. The output value of cigarette packs calculated based on each sensor in the training set is labeled and stored and classified according to the different sensors on different packaging units; The initial sensor quality detection model is trained based on each sensor and its annotation in the training set. The final sensor quality detection model is formed until the model meets the preset accuracy requirements. The generalized mean square error Mse is used as the calculation standard. When Mse is less than the threshold, the sensor quality detection model meets the accuracy requirements.

5. An electronic device, characterized in that, The electronic device includes: Processor and memory; The memory is used to store computer programs; The processor is used to execute the computer program stored in the memory, so that the electronic device performs the sensor fault detection method according to claim 1 or 2, or the sensor quality testing method according to claim 3, or the sensor quality detection model construction method according to claim 4.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by an electronic device, the program implements the sensor fault detection method of claim 1 or 2, the sensor quality testing method of claim 3, or the sensor quality testing model construction method of claim 4.

Images