A multi-sensor-based system and method for measuring the adhesive strength at the sealing edge of paper cups.

Abstract

This invention relates to the technical field of strength measurement, and discloses a multi-sensor-based system and method for measuring the adhesive strength of paper cup sealing edges. The invention first uses force and displacement sensors to measure the adhesive strength of a batch of paper cup sealing samples to obtain initial adhesive strength measurement data. Second, the initial adhesive strength measurement data is filled with missing data and cleaned of duplicate data to obtain processed adhesive strength measurement data. Then, based on the processed adhesive strength measurement data, the adhesive strength of the batch of paper cup sealing samples is calculated. Finally, a backpropagation (BP) neural network is trained to obtain a BP neural network model, which is used to predict the adhesive strength of subsequent batches of paper cup sealing samples. This invention uses multiple sensors to acquire adhesive strength measurement data, and calculates the adhesive strength after data processing, resulting in an accurate and objective method.

Description

Technical Field

[0001] This invention relates to the technical field of strength measurement, specifically to a multi-sensor-based system and method for measuring the adhesive strength of paper cup sealing edges. Background Technology

[0002] Traditional methods for measuring adhesive strength often rely on manual labor. This process is not only labor-intensive and time-consuming, but the results are also often subjective, lacking data processing and leading to errors. When measuring the adhesive strength at the edge of paper cups, multiple sensors are not used, resulting in slow measurement speeds and limited applicability due to the lack of prediction of adhesive strength at the edge of paper cups. Summary of the Invention

[0003] To address the problems in related technologies, this invention provides a multi-sensor-based system and method for measuring the adhesive strength at the sealing edge of paper cups, thereby overcoming the aforementioned technical problems in existing related technologies.

[0004] To solve the above-mentioned technical problems, the present invention is achieved through the following technical solution:

[0005] A method for measuring the adhesive strength at the sealing edge of a paper cup based on multiple sensors includes the following steps:

[0006] S1. Obtain a batch of paper cup sealing samples, maintain the measurement environment using temperature and humidity sensors, and measure the shear strength, tensile strength and peel strength data of the paper cup sealing samples using force and displacement sensors to obtain the initial adhesive strength measurement data matrix.

[0007] S2. Fill in missing data and clean up duplicate data in the initial bond strength measurement data matrix to obtain the processed bond strength measurement data matrix;

[0008] S3. Calculate the bonding strength of the processed bonding strength measurement data, and take the average bonding strength as the bonding strength of this batch of paper cup edge sealing samples to complete the bonding strength measurement of the paper cup edge sealing.

[0009] S4. Obtain a new batch of paper cup sealing samples to get a new adhesive strength measurement data matrix. Combine the new adhesive strength measurement data matrix to train a BP neural network to obtain a BP neural network model. Use this model to predict the adhesive strength of subsequent batches of paper cup sealing samples, thus achieving the prediction of adhesive strength at the paper cup sealing edge.

[0010] Preferably, step S1 includes the following steps:

[0011] S11. Obtain a batch of paper cup sealing samples, place the paper cup sealing samples in a measurement environment, use a temperature sensor and a humidity sensor to measure the temperature and humidity of the measurement environment, and divide the paper cup sealing samples into paper cup sealing stretching samples, paper cup sealing shearing samples and paper cup sealing peeling samples.

[0012] S12. Apply stretching, shearing and peeling operations to the paper cup edge-stretching sample, paper cup edge-shearing sample and paper cup edge-peeling sample until the paper cup edge falls off, and use force sensor and displacement sensor to record the displacement distance under different force states, respectively, to obtain paper cup edge-stretching strength data set, paper cup edge-shearing strength data set and paper cup edge-peeling strength data set.

[0013] S13. Combine the paper cup edge sealing tensile strength data set, paper cup edge sealing shear strength data set, and paper cup edge sealing peel strength data set to obtain the initial adhesive strength measurement data matrix.

[0014] Preferably, step S2 includes the following steps:

[0015] S21. Use the KNN algorithm to fill in the missing data in the sample data of the initial bond strength measurement data matrix to obtain the processed bond strength measurement data matrix;

[0016] S22. Use the improved SNM algorithm to perform repeated data cleaning on the sample data in the processed adhesive strength measurement data matrix to obtain the processed adhesive strength measurement data matrix.

[0017] Preferably, step S21 includes the following steps:

[0018] S211. Select sample data from the initial bond strength measurement data matrix to obtain a first tensile strength data set, a first shear strength data set, and a first peel strength data set; calculate the Euclidean distance from the missing data in the first tensile strength data set to the other sample data to obtain a first Euclidean distance set; calculate the Euclidean distance from the missing data in the first shear strength data set to the other sample data to obtain a second Euclidean distance set; calculate the Euclidean distance from the missing data in the first peel strength data set to the other sample data to obtain a third Euclidean distance set.

[0019] S212. Set the number of nearest neighbors k, find the k smallest Euclidean distances in the first Euclidean distance set, the second Euclidean distance set, and the third Euclidean distance set respectively, and add them to the first nearest neighbor set, the second nearest neighbor set, and the third nearest neighbor set respectively;

[0020] The missing data in the first tensile strength data set is filled using the first tensile strength data corresponding to the most frequent Euclidean distance in the first nearest neighbor set; the missing data in the first shear strength data set is filled using the first shear strength data corresponding to the most frequent Euclidean distance in the second nearest neighbor set; the missing data in the first peel strength data set is filled using the first peel strength data corresponding to the most frequent Euclidean distance in the third nearest neighbor set; all missing data in the initial bond strength measurement data matrix are filled sequentially to obtain the processed bond strength measurement data matrix.

[0021] Preferably, step S22 includes the following steps:

[0022] S221. Select sample data from the processed bond strength measurement data matrix to obtain a data set to be cleaned. Calculate the attribute values ​​of the data to be cleaned in the data set to be cleaned and sort the attribute values ​​to obtain an attribute value set. Map the attributes in the attribute value set to a vector space to obtain a feature vector set corresponding to the data set to be cleaned. Calculate the cosine similarity of all data to be cleaned in the data set to be cleaned based on the feature vectors in the feature vector set.

[0023] S222. Set a sliding window, place the sliding window in the dataset to be cleaned, and set a similarity threshold. By comparing the cosine similarity of the data to be cleaned in the sliding window with the similarity threshold, the duplicate data in the sliding window is cleaned until all the duplicate data in the processed adhesive strength measurement data matrix is ​​cleaned, and the processed adhesive strength measurement data matrix is ​​obtained.

[0024] Preferably, step S3 includes the following steps:

[0025] S31. Based on the adhesive strength measurement data in the processed adhesive strength measurement data matrix, set the adhesive area at the paper cup sealing edge as s, and calculate the adhesive strength according to the adhesive strength calculation formula.

[0026] S32. Calculate the bonding strength of all bonding strength measurement data in the processed bonding strength measurement data matrix in sequence, and calculate the average value of the bonding strength of all bonding strength measurement data, which is recorded as the average value of bonding strength. Use the average value of bonding strength as the bonding strength of the paper cup sealing sample in this batch to complete the bonding strength measurement of the paper cup sealing.

[0027] Preferably, step S4 includes the following steps:

[0028] S41. Obtain a new batch of paper cup sealing samples. Use force sensors and displacement sensors to test the new batch of paper cup sealing samples respectively, record the displacement distance under different force states, and after missing data filling and duplicate data cleaning, obtain a new adhesive strength measurement data matrix. Train the BP neural network to obtain the BP neural network model.

[0029] S42. Obtain the bonding length and width of the paper cup sealing edge of this batch of paper cup sealing samples, and combine it with the bonding strength of the paper cup sealing edge of this batch of paper cup sealing samples to obtain a bonding strength prediction set. Input the bonding strength prediction set into the BP neural network model, and output the prediction results of the bonding strength of the paper cup sealing edge of subsequent batches of paper cup sealing samples to complete the prediction of the bonding strength of the paper cup sealing edge.

[0030] Preferably, training the BP neural network to obtain the BP neural network model includes the following steps:

[0031] Let the mean squared error function of the BP neural network be E, and let it be trained according to the gradient descent direction. The number of neurons in the input layer of the BP neural network is... The number of neurons in the output layer of a BP neural network is p represents a constant and Number of neurons in the output layer of a BP neural network Select the adhesive strength measurement data from the new adhesive strength measurement data matrix, calculate the adhesive strength of a new batch of paper cup sealing samples, and then obtain the adhesive length and width at the paper cup sealing edge of the new batch of paper cup sealing samples. Combine the adhesive strength, adhesive length and width at the paper cup sealing edge to form a sample dataset.

[0032] The sample dataset is divided into a training set and a test set. The training set is input into a backpropagation (BP) neural network for training, iterating continuously until the BP neural network converges, resulting in a trained BP neural network. The test set is then input into the trained BP neural network, with a set accuracy threshold. When the error between the output result and the actual value is less than the accuracy threshold, the BP neural network model is obtained; otherwise, the weights are adjusted until the error between the output result and the actual value is less than the accuracy threshold.

[0033] A system for implementing the above-mentioned multi-sensor-based method for measuring the adhesive strength at the edge of a paper cup specifically includes: an initial adhesive strength measurement data acquisition module, an initial adhesive strength measurement data processing module, a paper cup edge-sealing adhesive strength calculation module, and a paper cup edge-sealing adhesive strength prediction module.

[0034] The initial bond strength measurement data acquisition module is used to measure the initial bond strength of paper cup sealing samples using multiple sensors;

[0035] The initial bond strength measurement data processing module is used to fill in missing data and clean up duplicate data in the initial bond strength measurement data;

[0036] The adhesive strength calculation module at the edge of the paper cup is used to calculate the adhesive strength of the paper cup edge-sealing sample using adhesive strength measurement data;

[0037] The paper cup edge sealing adhesive strength prediction module is used to predict the adhesive strength of subsequent batches of paper cup edge sealing samples using a BP neural network.

[0038] The present invention has the following beneficial effects:

[0039] 1. This invention uses tensile, shear, and peeling forces applied to paper cup sealing samples respectively. This measurement method takes into account the external force under various conditions and uses multiple sensors, which can effectively reduce errors and improve the accuracy of adhesive strength results.

[0040] 2. This invention uses the KNN algorithm to fill in missing data in the initial bond strength measurement data. It indirectly fills in the missing data by comparing Euclidean distances. The algorithm is simple in concept, has a good filling effect, and is not sensitive to data.

[0041] 3. This invention uses an improved SNM algorithm to perform duplicate data cleaning on the initial bond strength measurement data and introduces similarity. By comparing the similarity, it aims to determine whether the data is duplicated. This algorithm is simple to use, has high cleaning efficiency, and greatly shortens the detection speed through similarity.

[0042] 4. This invention obtains a BP neural network model by acquiring a trained BP neural network, and realizes the prediction of the bonding strength of subsequent batches of paper cup sealing samples; the neural network can effectively solve complex mathematical problems, with small prediction error and strong prediction feasibility.

[0043] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0044] To more clearly illustrate the technical solutions of the embodiments of the invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the invention. For those skilled in the art, the drawings can be obtained from these drawings without creative effort.

[0045] Figure 1 This is a schematic diagram illustrating the process of measuring the adhesive strength at the edge of a paper cup using a multi-sensor-based adhesive strength measurement system. Detailed Implementation

[0046] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0047] In the description of this invention, it should be understood that the terms "opening", "upper", "lower", "top", "middle", "inner", etc., which indicate orientation or positional relationship, are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the components or elements referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on the invention.

[0048] Example 1

[0049] Please see Figure 1 This embodiment discloses a method for measuring the adhesive strength at the sealing edge of a paper cup based on multiple sensors, specifically including the following:

[0050] S1. Obtain a batch of paper cup sealing samples, maintain the measurement environment using temperature and humidity sensors, and measure the shear strength, tensile strength and peel strength data of the paper cup sealing samples using force and displacement sensors to obtain the initial adhesive strength measurement data matrix.

[0051] S1 includes the following steps:

[0052] S11. Obtain a batch of paper cup sealing samples. Place the paper cup sealing samples in a measurement environment. Use a temperature sensor and a humidity sensor to measure the temperature and humidity of the measurement environment. Set the upper and lower limits of the optimal temperature threshold and the upper and lower limits of the optimal humidity threshold. Keep the temperature of the current measurement environment greater than or equal to the lower limit of the optimal temperature threshold and less than or equal to the upper limit of the optimal temperature threshold. Keep the humidity of the current measurement environment greater than or equal to the lower limit of the optimal humidity threshold and less than or equal to the upper limit of the optimal humidity threshold. Divide the paper cup sealing samples into paper cup sealing stretching samples, paper cup sealing shearing samples, and paper cup sealing peeling samples. Select any paper cup from the paper cup sealing stretching samples and record it as the first paper cup. Select any paper cup from the paper cup sealing shearing samples and record it as the second paper cup. Select any paper cup from the paper cup sealing peeling samples and record it as the third paper cup.

[0053] S12. Fix the first, second, and third paper cups on the platform, and apply stretching, shearing, and peeling techniques until the paper cup seals detach. By stretching the seal of the first paper cup, use force sensors and displacement sensors to record the displacement distance under different force conditions to obtain a set of paper cup seal tensile strength data. ,in Represents the force sensor reading recorded for the j-th time, where This represents the displacement sensor reading recorded at the j-th time. A shearing force is applied to the sealing edge of the second paper cup, and force and displacement sensors are used to record the displacement distance under different force conditions, thus obtaining a set of paper cup sealing edge shear strength data. ,in Represents the force sensor reading recorded for the j-th time, where This represents the displacement sensor reading recorded at the j-th time; under peeling conditions, at the edge of the third paper cup, the displacement distance is recorded using both force and displacement sensors under different force states, thus obtaining the paper cup edge peeling strength data set. ,in Represents the force sensor reading recorded for the j-th time, where This represents the displacement sensor reading recorded for the j-th time;

[0054] S13. After measuring all the paper cup edge-sealing tensile samples, paper cup edge-sealing shear samples, and paper cup edge-sealing peel samples, the paper cup edge-sealing tensile strength data matrix is ​​obtained. Paper cup edge sealing shear strength data matrix Paper cup edge seal peel strength data matrix as follows:

[0055] ; ;

[0056] ;

[0057] in, This represents the force sensor reading recorded for the i-th paper cup and the j-th time in the paper cup edge-sealing tension sample. This represents the displacement sensor reading recorded for the i-th paper cup and the j-th time in the paper cup edge sealing stretching sample. This indicates the first sample of paper cup edge sealing cutting. The force sensor reading for the j-th paper cup is recorded. This indicates the first sample of paper cup edge sealing cutting. The displacement sensor readings for the j-th paper cup; Indicating the first sample of paper cup edge sealing peeling The force sensor reading for the j-th paper cup is recorded. Indicating the first sample of paper cup edge sealing peeling The displacement sensor readings for the j-th paper cup;

[0058] By combining the paper cup edge sealing tensile strength data matrix, the paper cup edge sealing shear strength data matrix, and the paper cup edge sealing peel strength data matrix, an initial adhesive strength measurement data matrix is ​​obtained.

[0059] By dividing the paper cup sealing sample into three parts and applying tensile, shear, and peeling forces to the sample respectively, this measurement method considers the external force under various conditions, which can reduce errors. Force and displacement sensors are used to record the displacement distance under different force conditions to obtain the initial adhesive strength measurement data, thereby improving the accuracy of the adhesive strength results.

[0060] S2. Fill in missing data and clean up duplicate data in the initial bond strength measurement data matrix to obtain the processed bond strength measurement data matrix;

[0061] S2 includes the following steps:

[0062] S21. Use the KNN algorithm to fill in missing data in the sample data of the initial bond strength measurement data matrix to obtain the processed bond strength measurement data matrix. The specific steps are as follows:

[0063] S211. Select sample data from the initial adhesive strength measurement data matrix. The sample data are one row of sample data from the paper cup edge sealing tensile strength data matrix, the paper cup edge sealing shear strength data matrix, and the paper cup edge sealing peel strength data matrix, to obtain the first tensile strength data set, the first shear strength data set, and the first peel strength data set.

[0064] Missing data is found in the first tensile strength data set, the first shear strength data set, and the first peel strength data set. The Euclidean distance from the missing data in the first tensile strength data set to the other sample data is calculated to obtain the first Euclidean distance. The Euclidean distance from the missing data in the first shear strength data set to the other sample data is calculated to obtain the second Euclidean distance. The Euclidean distance from the missing data in the first peel strength data set to the other sample data is calculated to obtain the third Euclidean distance. The calculation formulas are as follows:

[0065] , , ;

[0066] in, Indicates the first Euclidean distance. Indicates the first tensile strength data set One missing data point. Indicates the first tensile strength data set One data point; Indicates the second Euclidean distance. Represents the first shear strength data set. One missing data point. Represents the first shear strength data set. One data point; Indicates the third Euclidean distance. Indicates the first peel strength data set. One missing data point. Indicates the first peel strength data set. One data point;

[0067] S212. Obtain the first Euclidean distance set, the second Euclidean distance set, and the third Euclidean distance set respectively. Set the nearest neighbor number k. Sort the first Euclidean distances in the first Euclidean distance set in ascending order and find the k smallest Euclidean distances. Add the k smallest Euclidean distances to the first nearest neighbor set. Sort the second Euclidean distances in the second Euclidean distance set in ascending order and find the k smallest Euclidean distances. Add the k smallest Euclidean distances to the second nearest neighbor set. Sort the third Euclidean distances in the third Euclidean distance set in ascending order and find the k smallest Euclidean distances. Add the k smallest Euclidean distances to the third nearest neighbor set.

[0068] The missing data in the first tensile strength data set is filled using the first tensile strength data corresponding to the most frequent Euclidean distance in the first nearest neighbor set; the missing data in the first shear strength data set is filled using the first shear strength data corresponding to the most frequent Euclidean distance in the second nearest neighbor set; the missing data in the first peel strength data set is filled using the first peel strength data corresponding to the most frequent Euclidean distance in the third nearest neighbor set; all missing data in the initial bond strength measurement data matrix are filled sequentially to obtain the processed bond strength measurement data matrix;

[0069] S22. Use the improved SNM algorithm to perform duplicate data cleaning on the sample data in the processed bond strength measurement data matrix to obtain the processed bond strength measurement data matrix. The specific steps are as follows:

[0070] S221. Extract any row of sample data from the processed bond strength measurement data matrix to obtain a data set to be cleaned. Calculate the attribute values ​​of the data to be cleaned in the data set to be cleaned and sort the attribute values ​​to obtain an attribute value set. Sort the data to be cleaned in the data set to be cleaned according to the sorted attribute set to obtain a sorted data set to be cleaned.

[0071] Mapping the attributes in the attribute value set to a vector space yields the feature vector set corresponding to the sorted data set to be cleaned. Let the h-th dimension of the g-th feature vector in the feature vector set be denoted as... , the th in the feature vector set The h-th dimension of the eigenvectors is denoted as... ,

[0072] If the dimension of the feature vectors in the feature vector set is l, then the cosine similarity... The calculation formula is as follows:

[0073] ;

[0074] Calculate the cosine similarity of all data to be cleaned in the sorted dataset;

[0075] S222. Set a sliding window, place the sliding window in the sorted set of data to be cleaned, and set a similarity threshold. When the cosine similarity between two sorted data to be cleaned in the sliding window is greater than the similarity threshold, the two sorted data to be cleaned are similar and duplicated, and the two sorted data to be cleaned are merged; otherwise, the two sorted data to be cleaned are not similar and duplicated, and other sorted data to be cleaned in the sliding window are compared until the duplicate data in the processed adhesive strength measurement data matrix is ​​cleaned, and the processed adhesive strength measurement data matrix is ​​obtained.

[0076] The KNN algorithm is used to fill in missing data in the initial bond strength measurement data. This algorithm is simple, has a good filling effect, and is not sensitive to data. Then, the improved SNM algorithm is used to clean up duplicate data in the initial bond strength measurement data and similarity is introduced. By comparing similarity, the purpose of determining whether the data is duplicated is achieved. The cleaning efficiency is high, and the detection speed is greatly shortened by using similarity.

[0077] S3. Calculate the bonding strength of the processed bonding strength measurement data, and take the average bonding strength as the bonding strength of this batch of paper cup edge sealing samples to complete the bonding strength measurement of the paper cup edge sealing.

[0078] S3 includes the following steps:

[0079] S31. Select one row from the processed bond strength measurement data matrix and denote it as the bond strength measurement data set to be calculated. ,in Represents the force sensor reading recorded for the j-th time, where Let represent the displacement sensor reading recorded for the j-th time; set the adhesive area at the paper cup sealing edge as s, then the formula for calculating the adhesive strength is as follows:

[0080] ;

[0081] Where E represents the bond strength;

[0082] S32. Calculate the bonding strength of all bonding strength measurement data in the processed bonding strength measurement data matrix in sequence, and calculate the average value of the bonding strength of all bonding strength measurement data, which is recorded as the average value of bonding strength. Use the average value of bonding strength as the bonding strength of the paper cup sealing sample in this batch to complete the bonding strength measurement of the paper cup sealing.

[0083] S4. Obtain a new batch of paper cup sealing samples, obtain a new adhesive strength measurement data matrix, train a BP neural network in combination with the new adhesive strength measurement data matrix, obtain a BP neural network model, predict the adhesive strength of subsequent batches of paper cup sealing samples, and realize the prediction of adhesive strength at the paper cup sealing edge.

[0084] S4 includes the following steps:

[0085] S41. Obtain a new batch of paper cup sealing samples. Use force sensors and displacement sensors to test the new batch of paper cup sealing samples, record the displacement distance under different force conditions, and after missing data imputation and duplicate data cleaning, obtain a new adhesive strength measurement data matrix. Combine the adhesive strength measurement data in the new adhesive strength measurement data matrix to train the BP neural network and obtain the BP neural network model. The specific steps are as follows:

[0086] S411. Set the mean square error function of the BP neural network to E, and train it according to the gradient descent direction. The number of neurons in the input layer of the BP neural network is... The number of neurons in the output layer of a BP neural network is p represents a constant and Number of neurons in the output layer of a BP neural network Select the adhesive strength measurement data from the new adhesive strength measurement data matrix, calculate the adhesive strength of a new batch of paper cup sealing samples, and then obtain the adhesive length and width at the paper cup sealing edge of the new batch of paper cup sealing samples. Combine the adhesive strength, adhesive length and width at the paper cup sealing edge to form a sample dataset.

[0087] S412. Divide the sample dataset into a training set and a test set. Input the training set into the BP neural network for training, iterating continuously until the BP neural network converges, thus obtaining a trained BP neural network. Then input the test set into the trained BP neural network, setting a precision threshold. When the error between the output result and the actual value is less than the accuracy threshold, the BP neural network model is obtained; otherwise, the weights are adjusted until the error between the output result and the actual value is less than the accuracy threshold.

[0088] S42. Obtain the bonding length and width of the paper cup sealing edge of this batch of paper cup sealing samples, and combine it with the bonding strength of the paper cup sealing edge of this batch of paper cup sealing samples to obtain a bonding strength prediction set. Input the bonding strength prediction set into the BP neural network model, and output the prediction results of the bonding strength of the paper cup sealing edge of subsequent batches of paper cup sealing samples to complete the prediction of the bonding strength of the paper cup sealing edge.

[0089] By acquiring a new batch of paper cup sealing samples, new adhesive strength measurement data is obtained. A BP neural network is then trained to obtain a BP neural network model, which enables the prediction of adhesive strength for subsequent batches of paper cup sealing samples. This neural network can effectively solve complex mathematical problems, with small prediction errors and strong prediction feasibility.

[0090] Example 2

[0091] This embodiment also discloses a system for measuring the adhesive strength at the edge of a paper cup based on multiple sensors, specifically including: an initial adhesive strength measurement data acquisition module, an initial adhesive strength measurement data processing module, a paper cup edge-sealing adhesive strength calculation module, and a paper cup edge-sealing adhesive strength prediction module;

[0092] The initial bond strength measurement data acquisition module is used to measure the initial bond strength of paper cup sealing samples using multiple sensors;

[0093] The initial bond strength measurement data processing module is used to fill in missing data and clean up duplicate data in the initial bond strength measurement data;

[0094] The adhesive strength calculation module at the edge of the paper cup is used to calculate the adhesive strength of the paper cup edge-sealing sample using adhesive strength measurement data;

[0095] The paper cup edge sealing adhesive strength prediction module is used to predict the adhesive strength of subsequent batches of paper cup edge sealing samples using a BP neural network.

[0096] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0097] The preferred embodiments of the invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention.

Claims

1. A method for measuring the adhesive strength at the sealing edge of a paper cup based on multiple sensors, characterized in that, Includes the following steps: S1. Obtain a batch of paper cup sealing samples, maintain the measurement environment using temperature and humidity sensors, and measure the shear strength, tensile strength and peel strength data of the paper cup sealing samples using force and displacement sensors to obtain the initial adhesive strength measurement data matrix. S2. Fill in missing data and clean up duplicate data in the initial bond strength measurement data matrix to obtain the processed bond strength measurement data matrix; S3. Calculate the bonding strength of the processed bonding strength measurement data, and take the average bonding strength as the bonding strength of this batch of paper cup edge sealing samples to complete the bonding strength measurement of the paper cup edge sealing. S4. Obtain a new batch of paper cup sealing samples to get a new adhesive strength measurement data matrix. Combine the new adhesive strength measurement data matrix to train a BP neural network to obtain a BP neural network model. Use this model to predict the adhesive strength of subsequent batches of paper cup sealing samples, thus achieving the prediction of adhesive strength at the paper cup sealing edge.

2. The method for measuring the adhesive strength at the sealing edge of a paper cup based on multiple sensors according to claim 1, characterized in that, S1 includes the following steps: S11. Obtain a batch of paper cup sealing samples, place the paper cup sealing samples in a measurement environment, use a temperature sensor and a humidity sensor to measure the temperature and humidity of the measurement environment, and divide the paper cup sealing samples into paper cup sealing stretching samples, paper cup sealing shearing samples and paper cup sealing peeling samples. S12. Apply stretching, shearing and peeling operations to the paper cup edge-stretching sample, paper cup edge-shearing sample and paper cup edge-peeling sample until the paper cup edge falls off, and use force sensor and displacement sensor to record the displacement distance under different force states, respectively, to obtain paper cup edge-stretching strength data set, paper cup edge-shearing strength data set and paper cup edge-peeling strength data set. S13. Combine the paper cup edge sealing tensile strength data set, paper cup edge sealing shear strength data set, and paper cup edge sealing peel strength data set to obtain the initial adhesive strength measurement data matrix.

3. The method for measuring the adhesive strength at the sealing edge of a paper cup based on multiple sensors according to claim 2, characterized in that, S2 includes the following steps: S21. Use the KNN algorithm to fill in the missing data in the sample data of the initial bond strength measurement data matrix to obtain the processed bond strength measurement data matrix; S22. Use the improved SNM algorithm to perform repeated data cleaning on the sample data in the processed adhesive strength measurement data matrix to obtain the processed adhesive strength measurement data matrix.

4. The method for measuring the adhesive strength at the sealing edge of a paper cup based on multiple sensors according to claim 3, characterized in that, S21 includes the following steps: S211. Select sample data from the initial bond strength measurement data matrix to obtain a first tensile strength data set, a first shear strength data set, and a first peel strength data set; Calculate the Euclidean distances from the missing data in the first tensile strength dataset to the other sample data to obtain the first Euclidean distance set; calculate the Euclidean distances from the missing data in the first shear strength dataset to the other sample data to obtain the second Euclidean distance set; calculate the Euclidean distances from the missing data in the first peel strength dataset to the other sample data to obtain the third Euclidean distance set. S212. Set the number of nearest neighbors k, find the k smallest Euclidean distances in the first Euclidean distance set, the second Euclidean distance set, and the third Euclidean distance set respectively, and add them to the first nearest neighbor set, the second nearest neighbor set, and the third nearest neighbor set respectively; The missing data in the first tensile strength data set is filled using the first tensile strength data corresponding to the most frequent Euclidean distance in the first nearest neighbor set; the missing data in the first shear strength data set is filled using the first shear strength data corresponding to the most frequent Euclidean distance in the second nearest neighbor set; the missing data in the first peel strength data set is filled using the first peel strength data corresponding to the most frequent Euclidean distance in the third nearest neighbor set; all missing data in the initial bond strength measurement data matrix are filled sequentially to obtain the processed bond strength measurement data matrix.

5. The method for measuring the adhesive strength at the sealing edge of a paper cup based on multiple sensors according to claim 4, characterized in that, S22 includes the following steps: S221. Select sample data from the processed bond strength measurement data matrix to obtain a data set to be cleaned. Calculate the attribute values ​​of the data to be cleaned in the data set to be cleaned and sort the attribute values ​​to obtain an attribute value set. Map the attributes in the attribute value set to a vector space to obtain a feature vector set corresponding to the data set to be cleaned. Calculate the cosine similarity of all data to be cleaned in the data set to be cleaned based on the feature vectors in the feature vector set. S222. Set a sliding window, place the sliding window in the dataset to be cleaned, and set a similarity threshold. By comparing the cosine similarity of the data to be cleaned in the sliding window with the similarity threshold, the duplicate data in the sliding window is cleaned until all the duplicate data in the processed adhesive strength measurement data matrix is ​​cleaned, and the processed adhesive strength measurement data matrix is ​​obtained.

6. The method for measuring the adhesive strength at the sealing edge of a paper cup based on multiple sensors according to claim 5, characterized in that, S3 includes the following steps: S31. Based on the adhesive strength measurement data in the processed adhesive strength measurement data matrix, set the adhesive area at the paper cup sealing edge as s, and calculate the adhesive strength according to the adhesive strength calculation formula. S32. Calculate the bonding strength of all bonding strength measurement data in the processed bonding strength measurement data matrix in sequence, and calculate the average value of the bonding strength of all bonding strength measurement data, which is recorded as the average value of bonding strength. Use the average value of bonding strength as the bonding strength of the paper cup sealing sample in this batch to complete the bonding strength measurement of the paper cup sealing.

7. The method for measuring the adhesive strength at the sealing edge of a paper cup based on multiple sensors according to claim 6, characterized in that, S4 includes the following steps: S41. Obtain a new batch of paper cup sealing samples. Use force sensors and displacement sensors to test the new batch of paper cup sealing samples respectively, record the displacement distance under different force states, and after missing data filling and duplicate data cleaning, obtain a new adhesive strength measurement data matrix. Train the BP neural network to obtain the BP neural network model. S42. Obtain the bonding length and width of the paper cup sealing edge of this batch of paper cup sealing samples, and combine it with the bonding strength of the paper cup sealing edge of this batch of paper cup sealing samples to obtain a bonding strength prediction set. Input the bonding strength prediction set into the BP neural network model, and output the prediction results of the bonding strength of the paper cup sealing edge of subsequent batches of paper cup sealing samples to complete the prediction of the bonding strength of the paper cup sealing edge.

8. The method for measuring the adhesive strength at the sealing edge of a paper cup based on multiple sensors according to claim 7, characterized in that, Training the BP neural network to obtain the BP neural network model includes the following steps: Let the mean squared error function of the BP neural network be E, and let it be trained according to the gradient descent direction. The number of neurons in the input layer of the BP neural network is... The number of neurons in the output layer of a BP neural network is p represents a constant and Number of neurons in the output layer of a BP neural network Select the adhesive strength measurement data from the new adhesive strength measurement data matrix, calculate the adhesive strength of a new batch of paper cup sealing samples, and then obtain the adhesive length and width at the paper cup sealing edge of the new batch of paper cup sealing samples. Combine the adhesive strength, adhesive length and width at the paper cup sealing edge to form a sample dataset. The sample dataset is divided into a sample training set and a sample test set. The sample training set is input into the BP neural network for training. The training is iterated continuously until the BP neural network converges, and the trained BP neural network is obtained. Then, the sample test set is input into the trained BP neural network, and the accuracy threshold is set to... When the error between the output result and the actual value is less than the accuracy threshold, the BP neural network model is obtained; otherwise, the weights are adjusted until the error between the output result and the actual value is less than the accuracy threshold.

9. A system for implementing the multi-sensor-based method for measuring the adhesive strength at the sealing edge of a paper cup as described in any one of claims 1-8, characterized in that, Specifically, it includes: The module includes: initial bond strength measurement data acquisition module, initial bond strength measurement data processing module, paper cup edge sealing bond strength calculation module, and paper cup edge sealing bond strength prediction module. The initial bond strength measurement data acquisition module is used to measure the initial bond strength of paper cup sealing samples using multiple sensors; The initial bond strength measurement data processing module is used to fill in missing data and clean up duplicate data in the initial bond strength measurement data; The adhesive strength calculation module at the edge of the paper cup is used to calculate the adhesive strength of the paper cup edge-sealing sample using adhesive strength measurement data; The paper cup edge sealing adhesive strength prediction module is used to predict the adhesive strength of subsequent batches of paper cup edge sealing samples using a BP neural network.

Images