Paper cup weight intelligent weighing system and method based on artificial intelligence

Abstract

The invention relates to the technical field of intelligent weighing, and discloses a paper cup weight intelligent weighing system and method based on artificial intelligence, and the system comprises a data collection module, an intelligent analysis module, a display early warning module, an automatic calibration module and a database. The intelligent analysis module is used for judging the qualification condition of the paper cup according to the comparison result of the weight measurement data and the first weight threshold range, and compensating and correcting the weight measurement data according to the comparison result of the qualified weight measurement data and the second weight threshold range to obtain a final weighing result; and the automatic calibration module is used for automatically calibrating the measurement parameters according to the intelligent analysis result. According to the invention, the precision and reliability of paper cup weight detection can be significantly improved, the frequency of manual calibration can be reduced, the maintenance cost can be effectively reduced, the measurement parameters can be dynamically adjusted according to real-time environmental conditions, the high-precision weighing result can still be provided under complex production conditions, and the method has a wide application prospect.

Description

Technical Field

[0001] The present invention relates to the field of intelligent weighing technology, and in particular to an artificial intelligence-based intelligent weighing system and method for paper cup weight. Background Art

[0002] In modern industrial production, quality control is a key link to ensure product quality and reduce costs, especially in the manufacturing process of disposable paper cups, the weight of the paper cup directly affects its performance and manufacturing cost. If the weight of the paper cup is too light, it may lead to insufficient structural strength and affect the user experience, while if it is too heavy, it may increase production costs and reduce economic benefits. Therefore, accurate weight control is crucial for manufacturers.

[0003] Traditional weight detection systems usually rely on mechanical or electronic sensors. Although these sensors can provide basic weight measurements, their detection accuracy is affected by many external factors, especially temperature and humidity changes and material fluctuations in the production environment. These factors often lead to inaccurate weight measurements, which ultimately affect the quality judgment of paper cups.

[0004] In addition, the error control of traditional weighing systems is usually based on a preset static error range. Although this method is simple, it lacks flexibility and fails to fully consider the dynamically changing environmental conditions and material properties during the production process. Therefore, traditional systems are prone to false alarms and missed detections during detection, resulting in unstable product quality control.

[0005] At the same time, traditional weighing systems often rely on manual calibration. This calibration method is not only time-consuming and labor-intensive, but also difficult to ensure real-time precision adjustment in the rapidly changing production environment, which in turn affects the efficiency of the entire production process and the reliability of the test results. To solve these problems, with the rapid development of artificial intelligence (AI) technology, intelligent weighing systems based on machine learning have gradually been introduced into industrial manufacturing, becoming an effective means to improve detection accuracy and optimize the production process.

[0006] Therefore, with the continuous development of intelligent manufacturing technology, traditional weight detection methods have gradually failed to meet the needs of modern efficient and accurate production. The intelligent weighing system based on artificial intelligence can not only effectively deal with various uncertainties that may be encountered in the production process, but also greatly improve the weighing accuracy and production efficiency through automatic calibration and compensation technology, while ensuring product quality, reducing the production cost of the enterprise. Therefore, the present invention proposes an intelligent weighing system and method for paper cup weight based on artificial intelligence. Summary of the invention

[0007] In view of the problems in the related art, the present invention proposes an intelligent weighing system and method for paper cup weight based on artificial intelligence to overcome the above-mentioned technical problems existing in the existing related art.

[0008] To this end, the specific technical solution adopted by the present invention is as follows:

[0009] According to one aspect of the present invention, there is provided an artificial intelligence-based intelligent weighing system for paper cup weight, comprising a data acquisition module, an intelligent analysis module, a display and warning module, an automatic calibration module and a database;

[0010] Wherein, the data acquisition module is used to collect the weight measurement data of the paper cup and the environmental temperature and humidity data in real time;

[0011] The intelligent analysis module is used to determine the qualification of the paper cup according to the comparison result of the weight measurement data and the first weight threshold range, and to compensate and correct the weight measurement data according to the comparison result of the qualified weight measurement data and the second weight threshold range to obtain the final weighing result;

[0012] The display warning module is used to display the final weighing result, qualified status or abnormal situation of the paper cup, and issue a warning in the abnormal situation;

[0013] The automatic calibration module is used to automatically calibrate the measurement parameters according to the results of the intelligent analysis;

[0014] The database is used to store and manage weighing data, analysis results and system operation records.

[0015] Preferably, the intelligent analysis module includes a paper cup quality detection module, a data compensation and correction module and a weighing result output module;

[0016] The paper cup quality detection module is used to obtain weight measurement data of the paper cup, and determine the quality of the paper cup according to the comparison result between the weight measurement data and the first weight threshold range;

[0017] The data compensation and correction module is used to compensate and correct the weight measurement data according to the comparison result between the qualified weight measurement data and the second weight threshold range to obtain the final weighing result;

[0018] The weighing result output module is used to output the final weighing result and the qualification status of the paper cup.

[0019] Preferably, the paper cup quality detection module includes a first weight threshold range setting module and a qualified condition determination module;

[0020] Wherein, the first weight threshold range setting module is used to calculate the first weight threshold range of the paper cup according to the production design parameters of the paper cup and the allowable error range;

[0021] The qualified condition determination module is used to obtain the weight measurement data of the paper cup and determine whether the weight measurement data meets a preset first weight threshold range. If not, the paper cup corresponding to the weight measurement data is determined to be an unqualified product. If so, the paper cup corresponding to the weight measurement data is determined to be a qualified product.

[0022] Preferably, the data compensation and correction module includes a second weight threshold range setting module and a data compensation and correction module;

[0023] The second weight threshold range setting module is used to determine the corresponding theoretical weight value by combining the material parameters and environmental parameters of the current paper cup with machine learning technology, and set the second weight threshold range according to the theoretical weight value;

[0024] The data compensation and correction module is used to determine whether the weight measurement data of the qualified paper cup meets the preset second weight threshold range. If so, the weight measurement data is used as the final weighing result. If not, the weight measurement data is compensated and corrected using the output result of the machine learning technology, and the compensated weight measurement data is used as the final weighing result.

[0025] Preferably, the second weight threshold range setting module determines the corresponding theoretical weight value by combining the material parameters and environmental parameters of the current paper cup with machine learning technology, and sets the second weight threshold range according to the theoretical weight value, including:

[0026] Construct and train a hybrid quantum classical neural network model based on historical data to obtain a trained hybrid quantum classical neural network model;

[0027] The trained hybrid quantum classical neural network model is used to predict the theoretical weight value corresponding to the material parameters and environmental parameters of the current paper cup;

[0028] Based on statistical techniques and the error range of the sensor, the second weight threshold range is set in combination with the theoretical weight value.

[0029] Preferably, the constructing and training a hybrid quantum classical neural network model based on historical data to obtain a trained hybrid quantum classical neural network model comprises:

[0030] Obtain the material parameters of paper cups, environmental parameters of production workshops and weight measurement data based on the historical database, and clean and normalize the acquired data;

[0031] The features related to the weight of the paper cup are extracted according to the material parameters and the environmental parameters, and the principal component analysis method is used to select the potential key features related to the weight of the paper cup to obtain the training data set;

[0032] A hybrid quantum classical neural network model is constructed, and the training data set is used to train and optimize the hybrid quantum classical neural network model to obtain a trained hybrid quantum classical neural network model.

[0033] Preferably, the hybrid quantum classical neural network model is composed of a front classical neural network and a quantum neural network;

[0034] Among them, the front classical neural network is used to process the material parameters and environmental parameter characteristics of the paper cup; the quantum neural network is used to capture the complex nonlinear interaction between environmental parameters and material parameters.

[0035] Preferably, the second weight threshold range is set based on the statistical technology and the error range of the sensor in combination with the theoretical weight value, including:

[0036] Obtain the theoretical weight value and sensor error range, calculate the measurement error based on the actual measurement data in the historical data and the corresponding theoretical data, and calculate the average value and standard deviation based on the measurement error;

[0037] The range of the measurement error is set based on the mean value and standard deviation of the measurement error, and the second weight threshold range is set in combination with the error range of the sensor and the theoretical weight value;

[0038] Among them, the second weight threshold range is:

[0039] [W t -(E s +2σ e ),W t +(E s +2σ e )]

[0040] Where W t Indicates the theoretical weight value, E s represents the sensor error range, σ e Represents standard deviation.

[0041] Preferably, the automatic calibration module includes an error feedback module, a calibration strategy generation module and a sensor calibration module;

[0042] Wherein, the error feedback module is used to receive the measurement error data fed back by the intelligent analysis module;

[0043] The calibration strategy generation module is used to generate a calibration strategy based on the measurement error data combined with the real-time environmental temperature and humidity data, and the calibration strategy includes adjusting the sensitivity, zero offset or environmental compensation parameters of the sensor;

[0044] The sensor calibration module is used to execute the calibration strategy and adjust the parameters of the sensor.

[0045] According to another aspect of the present invention, there is provided an artificial intelligence-based intelligent weighing method for paper cup weight, the artificial intelligence-based intelligent weighing method for paper cup weight comprising the following steps:

[0046] S1. Real-time collection of paper cup weight measurement data and environmental temperature and humidity data;

[0047] S2, judging the qualification of the paper cup according to the comparison result of the weight measurement data and the first weight threshold range, and compensating and correcting the weight measurement data according to the comparison result of the qualified weight measurement data and the second weight threshold range to obtain the final weighing result;

[0048] S3, display the final weighing result, qualified status or abnormal situation of the paper cup, and issue an early warning in abnormal situation;

[0049] S4. Automatically calibrate the measurement parameters according to the results of intelligent analysis, and store and manage weighing data, analysis results and system operation records.

[0050] Compared with the prior art, the present invention provides an intelligent weighing system and method for paper cup weight based on artificial intelligence, which has the following beneficial effects:

[0051] (1) The present invention can not only significantly improve the accuracy and reliability of paper cup weight detection by introducing machine learning technology, but also effectively reduce the frequency of manual calibration through the automatic calibration function, thereby effectively reducing maintenance costs. In addition, the present invention can also dynamically adjust the measurement parameters according to real-time environmental conditions to ensure that high-precision weighing results can still be provided under complex production conditions, and has broad application prospects.

[0052] (2) The present invention can not only detect the qualified status of paper cups according to the first weight threshold range through the intelligent analysis module, but also identify abnormal measurement data in qualified paper cups according to the second weight threshold range, and use machine learning technology to compensate and correct it, thereby ensuring the accuracy of the final weighing result. This automatic compensation function enables the system to adapt to various uncertain factors that may occur in production, reduce the need for manual intervention, and improve production efficiency.

[0053] (3) The present invention introduces a hybrid quantum classical neural network model, which can accurately predict the theoretical weight value of the paper cup based on material parameters and environmental conditions. Therefore, it can not only set the second weight threshold range by combining statistical technology and the error range of the sensor, but also use the accurately predicted theoretical weight value to compensate and correct the real-time weight measurement data. Compared with the traditional weighing method, the present invention can dynamically set the weight threshold range according to real-time environmental temperature and humidity conditions, greatly improving the accuracy and adaptability of detection.

[0054] (4) The present invention has an automatic calibration module, which can dynamically generate a calibration strategy and adjust the sensor parameters in real time, including sensitivity, zero offset, and environmental compensation parameters, based on the results of intelligent analysis, combined with measurement errors and real-time environmental data. This function ensures that the system always maintains a high-precision weighing effect during long-term operation, and reduces measurement errors caused by equipment aging or environmental changes. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0056] Figure 1 is a structural block diagram of an intelligent weighing system for paper cup weight based on artificial intelligence according to an embodiment of the present invention;

[0057] Figure 2 The present invention is a flowchart of an intelligent weighing method for paper cup weight based on artificial intelligence.

[0058] In the figure:

[0059] 1. Data acquisition module; 2. Intelligent analysis module; 3. Display and warning module; 4. Automatic calibration module; 5. Database. DETAILED DESCRIPTION

[0060] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0061] According to an embodiment of the present invention, a paper cup weight intelligent weighing system and method based on artificial intelligence are provided.

[0062] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to one embodiment of the present invention, there is provided an intelligent weighing system for paper cup weight based on artificial intelligence, comprising a data acquisition module 1, an intelligent analysis module 2, a display and warning module 3, an automatic calibration module 4 and a database 5;

[0063] Wherein, the data acquisition module 1 is used to collect the weight measurement data of the paper cup and the environmental temperature and humidity data in real time;

[0064] Specifically, in this embodiment, the weight measurement data of the paper cup and the ambient temperature and humidity data are collected in real time through the weighing sensor and the temperature and humidity sensor. After the data collection is completed, the original data will be filtered, noise eliminated, etc. to improve the accuracy and stability of the data. Common preprocessing techniques include low-pass filtering, Kalman filtering, etc.

[0065] The intelligent analysis module 2 is used to determine the qualification of the paper cup according to the comparison result of the weight measurement data and the first weight threshold range, and to compensate and correct the weight measurement data according to the comparison result of the qualified weight measurement data and the second weight threshold range to obtain the final weighing result;

[0066] Specifically, the intelligent analysis module 2 includes a paper cup quality detection module, a data compensation and correction module, and a weighing result output module;

[0067] The paper cup quality detection module is used to obtain weight measurement data of the paper cup, and determine the quality of the paper cup according to the comparison result between the weight measurement data and the first weight threshold range;

[0068] The paper cup quality detection module includes a first weight threshold range setting module and a qualified condition determination module;

[0069] Wherein, the first weight threshold range setting module is used to calculate the first weight threshold range of the paper cup according to the production design parameters of the paper cup and the allowable error range;

[0070] In this embodiment, the step of calculating the first weight threshold range of the paper cup according to the production design parameters of the paper cup and the allowable error range can be divided into the following specific steps:

[0071] 1. Determine the production design parameters of paper cups

[0072] Design weight of paper cup: Each paper cup has its predetermined design weight, which is usually determined by the design parameters such as the material, size, thickness, etc. of the paper cup. d It is the ideal weight expected during production.

[0073] Material parameters: The materials used in paper cups may have slight batch differences, so it is necessary to record the design values ​​of the material such as density and thickness, which have a direct impact on the final weight of the paper cup.

[0074] 2. Set the allowable error range

[0075] The error range allowed by the production process: During the production process, there is a certain error range for equipment and materials. For example, due to reasons such as equipment accuracy and uneven material distribution, the weight of the produced paper cups will fluctuate to a certain extent.

[0076] Error range setting: Usually, the allowable error range E a It is set according to production requirements and quality standards. The error range can be expressed as a percentage (such as ±5%) or an absolute weight value (such as ±0.5 grams). This value is usually determined based on industry standards, customer needs and material properties.

[0077] 3. Calculate the first weight threshold range

[0078] Calculation of upper and lower limits: Combined with the design weight W of the paper cup d and the allowable error range E a , the upper and lower limits of the first weight threshold range can be calculated.

[0079] Upper limit value W u =W d +E a ;

[0080] Lower limit value W l =W d -E a ;

[0081] 4. Adjust the error range according to actual production conditions

[0082] Consider the accuracy of the production equipment: The accuracy of the production equipment may affect the accuracy of the measurement. Therefore, when setting the allowable error, the error range of the equipment itself needs to be considered. m If the error of the equipment is large, it may be necessary to appropriately expand the allowable error range to prevent the determination result from being affected by the fluctuation of the equipment.

[0083] Adjusted error range: According to the accuracy of the production equipment, the adjusted error range can be expressed as: E t =E a +E m ;

[0084] Then recalculate the weight threshold range of the paper cup:

[0085] Upper limit value W u =W d +E t ;

[0086] Lower limit value W l =W d -E t ;

[0087] 5. Get the first weight threshold range

[0088] Final weight threshold range: [W l ,W u ]=[W d -E t,W d +E t ];

[0089] In order to better understand the technical solution of setting the first weight threshold range of the present invention, an example is given below:

[0090] Assume the design weight of the paper cup is W d 10 grams, allowable error E a The error E of the production equipment is 0.5 g. m 0.1 gram.

[0091] Then: total error E t =0.5+0.1=0.6 g;

[0092] Upper limit value W u =10+0.6=10.6 grams;

[0093] Lower limit value W l =10-0.6=9.4 grams.

[0094] The final first weight threshold range is: [9.4, 10.6].

[0095] The qualified condition determination module is used to obtain the weight measurement data of the paper cup and determine whether the weight measurement data meets a preset first weight threshold range. If not, the paper cup corresponding to the weight measurement data is determined to be an unqualified product. If so, the paper cup corresponding to the weight measurement data is determined to be a qualified product.

[0096] The data compensation and correction module is used to compensate and correct the weight measurement data according to the comparison result between the qualified weight measurement data and the second weight threshold range to obtain the final weighing result;

[0097] The data compensation and correction module includes a second weight threshold range setting module and a data compensation and correction module;

[0098] The second weight threshold range setting module is used to determine the corresponding theoretical weight value by combining the material parameters and environmental parameters of the current paper cup with machine learning technology, and set the second weight threshold range according to the theoretical weight value;

[0099] The second weight threshold range setting module determines the corresponding theoretical weight value by using machine learning technology in combination with the material parameters and environmental parameters of the current paper cup, and sets the second weight threshold range according to the theoretical weight value, including:

[0100] 1) constructing and training a hybrid quantum classical neural network model based on historical data to obtain a trained hybrid quantum classical neural network model; the constructing and training a hybrid quantum classical neural network model based on historical data to obtain a trained hybrid quantum classical neural network model includes:

[0101] Based on the historical database, the material parameters of the paper cup (including the type, thickness, material, etc. of the paper cup, which directly affect the theoretical weight of the paper cup), the environmental parameters of the production workshop (such as temperature, humidity, etc., which may affect the weight of the paper material and the accuracy of the sensor) and the weight measurement data are obtained, and the obtained data are cleaned and normalized;

[0102] The features related to the weight of the paper cup are extracted according to the material parameters and the environmental parameters, and the principal component analysis method is used to select the potential key features related to the weight of the paper cup to obtain the training data set;

[0103] Specifically, the features related to the weight of paper cups are extracted according to the material parameters and environmental parameters, and the principal component analysis (PCA) method is used to select the potential key features related to the weight of paper cups. The steps to obtain the training data set are as follows:

[0104] Extract features related to the weight of paper cups: Material parameters and environmental parameters have direct or indirect effects on the weight of paper cups. These influencing factors need to be extracted through feature engineering. Common feature extraction methods include: linear relationship features (directly extracting the linear effects of material density, thickness, etc. on weight), nonlinear relationship features (extracting the nonlinear relationship between temperature and humidity and weight, for example, extracting potential influencing factors through interaction terms or high-order terms, such as the square of temperature, the product of density and thickness, etc.), and combined features (combining multiple features to generate new features to represent complex influencing factors, such as the product of density and material thickness may better represent the final weight of the paper cup);

[0105] Apply principal component analysis: Principal component analysis (PCA) is a statistical technique for dimensionality reduction and feature selection. It converts the raw data into a small number of unrelated principal components through linear transformation, helping us find the most important features related to the target variable (paper cup weight). The specific steps are as follows:

[0106] Calculate the covariance matrix: Calculate the covariance matrix C for the standardized data of all material parameters and environmental parameters:

[0107]

[0108] Where, X i represents the eigenvector of the i-th sample, n represents the total number of samples, α is the mean of the eigenvector, and T represents the transpose;

[0109] Eigenvalue decomposition: Perform eigenvalue decomposition on the covariance matrix to obtain a set of eigenvectors (principal components) and corresponding eigenvalues. The eigenvalue represents the variance contribution of each principal component in the original data.

[0110] Select the main principal components: Select the first k principal components based on the size of the eigenvalues. These principal components retain most of the variance of the data, which means that they have the strongest ability to explain the data. Usually, the principal components that can explain more than 90% of the variance are selected.

[0111] Transform data: Project the original material parameter and environmental parameter data onto these principal components to form a new feature data set. This reduced feature set will contain the potential key features that have the greatest impact on the weight of the paper cup.

[0112] Generate training data set: According to the principal components extracted by PCA, a new feature matrix after dimensionality reduction is obtained as the training data set for the machine learning model:

[0113] feature matrix: new feature matrix X PCA It is obtained by projecting the original material parameters and environmental parameters onto the principal components.

[0114] Target variable: The actual weighing data of the paper cup is used as the target variable Y.

[0115] The final training data set is (X PCA ,Y), used for subsequent machine learning model training.

[0116] A hybrid quantum classical neural network model is constructed, and the training data set is used to train and optimize the hybrid quantum classical neural network model to obtain a trained hybrid quantum classical neural network model.

[0117] Specifically, the hybrid quantum classical neural network model is composed of a front classical neural network and a quantum neural network, wherein the front classical neural network is used to process the material parameters and environmental parameter characteristics of the paper cup; the quantum neural network is used to capture the complex nonlinear interaction between environmental parameters and material parameters.

[0118] 1. Classical Neural Network (CNN): Mainly used to process conventional input data, such as the material parameters (density, thickness, type, etc.) and environmental parameters (temperature, humidity, etc.) of paper cups. The design of this part can be flexibly adjusted, but generally uses the following components:

[0119] Input layer: Input material parameters and environmental parameters. Assuming there are N material and environmental characteristics, the size of the input layer is N.

[0120] Hidden layer: The hidden layer of a classic neural network is used to preliminarily process input features. You can use fully connected layers or convolutional layers. For example, use multiple fully connected layers with activation functions (such as ReLU, tanh, etc.) to extract key features.

[0121] Output to the quantum layer: The output of the classical network is usually a feature vector, which is used as the input of the quantum circuit. In this part, the classical neural network is responsible for extracting relatively high-dimensional features for quantum processing. Assuming that the output size of the classical network is M, this vector can be mapped to the quantum state.

[0122] 2. Quantum Neural Network (QNN): The quantum part uses quantum circuits to perform quantum state encoding and quantum operations on the high-dimensional features extracted by the classical network, and captures complex nonlinear relationships through the evolution of quantum states, that is, capturing the complex nonlinear interactions between environmental parameters and material parameters.

[0123] Quantum state encoding: The output feature vector of the classical neural network will be encoded into a quantum state. The encoding method of the quantum state is usually amplitude encoding or angle encoding, and the specific choice depends on the number of qubits. For example, if the classical network outputs a feature vector of length M, it can be mapped to a state of log2M qubits through amplitude encoding.

[0124] Quantum circuit design: Quantum circuits are composed of a series of parameterized quantum gates. These quantum gates simulate complex nonlinear relationships by adjusting parameters. Common quantum gates include:

[0125] Hadamard gate: Generates a superposition state for initializing quantum bits.

[0126] Revolving gate (Rx, Ry, Rz): used to adjust the angle of the quantum bit, combined with classical parameters.

[0127] CNOT gate: Generates quantum entanglement to capture the correlation between features.

[0128] The combination of quantum gates constitutes a quantum circuit, and the quantum state evolves in the circuit. By parameterizing these quantum gates, classical optimization algorithms can be used to update these parameters and optimize the output of the quantum network.

[0129] Quantum state measurement: The output in quantum computing is obtained by measuring the quantum state. The measurement operation converts the probability distribution of the quantum state into classical data (such as a binary result of 0 or 1); by measuring the state of the quantum bit multiple times, a probability distribution is obtained, which is further converted into a classical data representation.

[0130] Classical post-processing: The output of the quantum circuit is measured and then combined with the processing results of the classical neural network. The quantum output can be fed back into the classical network to adjust the prediction value.

[0131] 3. Hybrid training process:

[0132] Training the classical part: The weights of the classical neural network part are trained using the traditional back propagation algorithm. The material and environmental parameters of the paper cup are input, and the intermediate high-dimensional feature vector is output.

[0133] Training the quantum part: The quantum part is trained using a variational quantum algorithm (VQA) to optimize the quantum circuit by adjusting the parameters of the quantum gates. Common quantum optimization algorithms include quantum natural gradient descent (QNG). The circuit is executed through a quantum circuit simulator (or quantum computer) and the loss function (such as mean square error or cross entropy) is calculated. The parameters of the quantum gates are updated through a classical optimizer to gradually reduce the prediction error.

[0134] Joint Optimization: The training of a hybrid quantum-classical model is an iterative process, with the classical and quantum parts optimized alternately. The classical optimizer (such as Adam, SGD) updates the weights of the classical neural network, and the quantum optimizer adjusts the parameters in the quantum circuit.

[0135] Evaluation and Validation: Use an independent validation dataset to evaluate the performance of the hybrid model and judge the generalization ability of the model based on the prediction error (such as mean square error, accuracy, etc.). Cross-validate the model to ensure that the model is robust to predictions under different production conditions and environments.

[0136] 2) Using the trained hybrid quantum classical neural network model to predict the theoretical weight value corresponding to the material parameters and environmental parameters of the current paper cup;

[0137] Specifically, the material parameters and environmental parameters of the current paper cup are input into the trained hybrid quantum classical neural network model, and the model is used to output the theoretical weight value corresponding to the material parameters and environmental parameters of the current paper cup.

[0138] 3) Based on statistical techniques and the error range of the sensor, the second weight threshold range is set in combination with the theoretical weight value.

[0139] In this embodiment, the weighing sensor is usually affected by temperature fluctuations, mechanical noise, humidity changes, etc. These errors may cause the measured value to deviate from the actual weight. Each sensor has an error range (such as ±0.01g) provided by the manufacturer, and this error range can be used as an important basis for setting the second weight threshold.

[0140] Specifically, the second weight threshold range is set based on the statistical technology and the error range of the sensor in combination with the theoretical weight value, including:

[0141] Obtain the theoretical weight value and sensor error range, calculate the measurement error based on the actual measurement data in the historical data and the corresponding theoretical data, and calculate the average value and standard deviation based on the measurement error;

[0142] The formula for calculating the average value is:

[0143]

[0144] The formula for calculating the standard deviation is:

[0145]

[0146] In the formula, μ e represents the average value, ε i represents the i-th measurement error value, n represents the total number of measurement errors, σ e Represents standard deviation.

[0147] The range of the measurement error is set based on the mean value and standard deviation of the measurement error, and the second weight threshold range is set in combination with the error range of the sensor and the theoretical weight value;

[0148] Among them, the second weight threshold range is:

[0149] [W t -(E s +2σ e ),W t +(E s +2σ e )]

[0150] Where W t Indicates the theoretical weight value, E s represents the sensor error range, σ e Represents standard deviation.

[0151] In order to better understand the technical solution of setting the second weight threshold range of the present invention, an example is given below:

[0152] Predicted theoretical weight W t =10 g;

[0153] Sensor error E s =0.01 g;

[0154] The average error μ in historical data e = 0 g (assuming the system has no bias error);

[0155] Standard deviation e=0.02 g;

[0156] Then the second weight threshold range is:

[0157] [10-(0.01+2×0.02),10+(0.01+2×0.02)]=[9.94,10.06];

[0158] Therefore, the second weight threshold range of the paper cup is 9.94 grams to 10.06 grams.

[0159] The data compensation and correction module is used to determine whether the weight measurement data of the qualified paper cup meets the preset second weight threshold range. If so, the weight measurement data is used as the final weighing result. If not, the weight measurement data is compensated and corrected using the output result of the machine learning technology (that is, the real-time weight measurement data is compensated and corrected using the theoretical weight value predicted by the model), and the compensated weight measurement data is used as the final weighing result.

[0160] The weighing result output module is used to output the final weighing result and the qualification status of the paper cup.

[0161] The display warning module 3 is used to display the final weighing result, qualified status or abnormal conditions of the paper cup, and issue a warning in abnormal conditions; the module displays the analysis results on the user interface, such as displaying the final weighing result, qualified status or abnormal conditions (such as overweight, leakage, etc.) of the paper cup. When the system detects an abnormal weight, the display warning module will issue an alarm and prompt the operator to handle or intervene. It can remind through a graphical interface or light, sound, etc.

[0162] The automatic calibration module 4 is used to automatically calibrate the measurement parameters according to the results of the intelligent analysis;

[0163] Specifically, the automatic calibration module (4) includes an error feedback module, a calibration strategy generation module and a sensor calibration module;

[0164] Among them, the error feedback module is used to receive the measurement error data fed back by the intelligent analysis module; this module receives the error information provided by the intelligent analysis module by monitoring the deviation between the measurement result and the theoretical weight value. If the measurement result exceeds the preset second weight threshold range, the error feedback module records the error information.

[0165] The calibration strategy generation module is used to generate a calibration strategy based on the measurement error data combined with the real-time environmental temperature and humidity data, and the calibration strategy includes adjusting the sensitivity, zero offset or environmental compensation parameters of the sensor; the function of this module is to dynamically formulate a correction plan based on the current error data and historical calibration information to ensure that the measurement parameters can adapt to the current environmental changes.

[0166] The sensor calibration module is used to execute the calibration strategy and adjust the parameters of the sensor. The sensor calibration module executes the calibration plan formulated by the calibration strategy generation module and directly adjusts the measurement parameters of the weighing sensor. The specific adjustments include:

[0167] Sensitivity Adjustment: Adjust the sensitivity of the sensor to ensure that more subtle weight changes can be captured.

[0168] Zero drift correction: Correct the sensor zero drift caused by long-term use or changes in temperature and humidity.

[0169] Environmental compensation parameters: According to environmental parameters such as temperature, humidity, and air pressure, the internal compensation mechanism of the sensor is appropriately adjusted to ensure accurate measurement in different environments.

[0170] The database 5 is used to store and manage weighing data, analysis results and system operation records.

[0171] Specifically, the database stores historical weighing data, training data for machine learning models, calibration logs, etc., which facilitates subsequent queries, statistical analysis, and model optimization. It can also be used to track system operation status, help analyze long-term trends, or discover potential problems. Through the accumulation of this data, the system can continue to improve and adapt to different usage scenarios.

[0172] According to another embodiment of the present invention, Figure 2 As shown, a method for intelligently weighing paper cup weight based on artificial intelligence is provided, and the method for intelligently weighing paper cup weight based on artificial intelligence comprises the following steps:

[0173] S1. Real-time collection of paper cup weight measurement data and environmental temperature and humidity data;

[0174] S2, judging the qualification of the paper cup according to the comparison result of the weight measurement data and the first weight threshold range, and compensating and correcting the weight measurement data according to the comparison result of the qualified weight measurement data and the second weight threshold range to obtain the final weighing result;

[0175] S3, display the final weighing result, qualified status or abnormal situation of the paper cup, and issue an early warning in abnormal situation;

[0176] S4. Automatically calibrate the measurement parameters according to the results of intelligent analysis, and store and manage weighing data, analysis results and system operation records.

[0177] In summary, with the help of the above technical scheme of the present invention, the present invention can not only significantly improve the accuracy and reliability of paper cup weight detection by introducing machine learning technology, but also effectively reduce the frequency of manual calibration through the automatic calibration function, thereby effectively reducing maintenance costs. In addition, the present invention can also dynamically adjust the measurement parameters according to real-time environmental conditions to ensure that high-precision weighing results can still be provided under complex production conditions, and has broad application prospects. At the same time, the present invention can not only detect the qualified status of paper cups according to the first weight threshold range through the intelligent analysis module, but also identify abnormal measurement data in qualified paper cups according to the second weight threshold range, and use machine learning technology to compensate and correct it to ensure the accuracy of the final weighing result. This automatic compensation function enables the system to adapt to various uncertainties that may occur in production, reducing the need for manual intervention and improving production efficiency. At the same time, the present invention can accurately predict the theoretical weight value of the paper cup based on material parameters and environmental conditions by introducing a hybrid quantum classical neural network model, so that not only can the second weight threshold range be set in combination with statistical techniques and the error range of the sensor, but also the real-time weight measurement data can be compensated and corrected using the accurately predicted theoretical weight value. Compared with the traditional weighing method, the present invention can dynamically set the weight threshold range according to real-time environmental temperature and humidity and other conditions, greatly improving the accuracy and adaptability of the detection. At the same time, the present invention has an automatic calibration module, which can dynamically generate a calibration strategy and adjust the parameters of the sensor in real time, including sensitivity, zero offset and environmental compensation parameters, based on the results of intelligent analysis, combined with measurement errors and real-time environmental data. This function ensures that the system always maintains a high-precision weighing effect during long-term operation, reducing measurement errors caused by equipment aging or environmental changes.

[0178] The technical features of the above-mentioned embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification. A person of ordinary skill in the art can understand that all or part of the steps in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a program. The program can be stored in a computer-readable storage medium. When the program is executed, it includes the steps described in the above method, and the storage medium, such as ROM / RAM, a disk, an optical disk, etc.

[0179] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. An artificial intelligence-based intelligent weighing system for paper cup weight, characterized in that: The paper cup weight intelligent weighing system based on artificial intelligence comprises a data acquisition module (1), an intelligent analysis module (2), a display and early warning module (3), an automatic calibration module (4) and a database (5); The data acquisition module (1) is used to collect the weight measurement data of the paper cup and the environmental temperature and humidity data in real time; The intelligent analysis module (2) is used to determine the qualification of the paper cup according to the comparison result of the weight measurement data and the first weight threshold range, and to compensate and correct the weight measurement data according to the comparison result of the qualified weight measurement data and the second weight threshold range to obtain a final weighing result; The display and warning module (3) is used to display the final weighing result, qualified status or abnormal situation of the paper cup, and issue a warning in the event of an abnormal situation; The automatic calibration module (4) is used to automatically calibrate the measurement parameters according to the results of the intelligent analysis; The database (5) is used to store and manage weighing data, analysis results and system operation records.

2. According to claim 1, the paper cup weight intelligent weighing system based on artificial intelligence is characterized in that: The intelligent analysis module (2) comprises a paper cup quality detection module, a data compensation and correction module and a weighing result output module; The paper cup quality detection module is used to obtain weight measurement data of the paper cup, and determine the quality of the paper cup according to the comparison result between the weight measurement data and the first weight threshold range; The data compensation and correction module is used to compensate and correct the weight measurement data according to the comparison result between the qualified weight measurement data and the second weight threshold range to obtain the final weighing result; The weighing result output module is used to output the final weighing result and the qualification status of the paper cup.

3. The paper cup weight intelligent weighing system based on artificial intelligence according to claim 2 is characterized in that: The paper cup quality detection module includes a first weight threshold range setting module and a qualified condition determination module; Wherein, the first weight threshold range setting module is used to calculate the first weight threshold range of the paper cup according to the production design parameters of the paper cup and the allowable error range; The qualified condition determination module is used to obtain the weight measurement data of the paper cup and determine whether the weight measurement data meets a preset first weight threshold range. If not, the paper cup corresponding to the weight measurement data is determined to be an unqualified product. If so, the paper cup corresponding to the weight measurement data is determined to be a qualified product.

4. The paper cup weight intelligent weighing system based on artificial intelligence according to claim 2 is characterized in that: The data compensation and correction module includes a second weight threshold range setting module and a data compensation and correction module; The second weight threshold range setting module is used to determine the corresponding theoretical weight value by combining the material parameters and environmental parameters of the current paper cup with machine learning technology, and set the second weight threshold range according to the theoretical weight value; The data compensation and correction module is used to determine whether the weight measurement data of the qualified paper cup meets the preset second weight threshold range. If so, the weight measurement data is used as the final weighing result. If not, the weight measurement data is compensated and corrected using the output result of the machine learning technology, and the compensated weight measurement data is used as the final weighing result.

5. The paper cup weight intelligent weighing system based on artificial intelligence according to claim 4 is characterized in that: The second weight threshold range setting module determines the corresponding theoretical weight value by using machine learning technology in combination with the material parameters and environmental parameters of the current paper cup, and sets the second weight threshold range according to the theoretical weight value, including: Construct and train a hybrid quantum classical neural network model based on historical data to obtain a trained hybrid quantum classical neural network model; The trained hybrid quantum classical neural network model is used to predict the theoretical weight value corresponding to the material parameters and environmental parameters of the current paper cup; Based on statistical techniques and the error range of the sensor, the second weight threshold range is set in combination with the theoretical weight value.

6. The paper cup weight intelligent weighing system based on artificial intelligence according to claim 5 is characterized in that: The hybrid quantum classical neural network model is constructed and trained based on historical data to obtain a trained hybrid quantum classical neural network model, which includes: Obtain the material parameters of paper cups, environmental parameters of production workshops and weight measurement data based on the historical database, and clean and normalize the acquired data; The features related to the weight of the paper cup are extracted according to the material parameters and the environmental parameters, and the principal component analysis method is used to select the potential key features related to the weight of the paper cup to obtain the training data set; A hybrid quantum classical neural network model is constructed, and the training data set is used to train and optimize the hybrid quantum classical neural network model to obtain a trained hybrid quantum classical neural network model.

7. The paper cup weight intelligent weighing system based on artificial intelligence according to claim 6 is characterized in that: The hybrid quantum classical neural network model is composed of a front classical neural network and a quantum neural network; Among them, the front classical neural network is used to process the material parameters and environmental parameter characteristics of the paper cup; the quantum neural network is used to capture the complex nonlinear interaction between environmental parameters and material parameters.

8. The paper cup weight intelligent weighing system based on artificial intelligence according to claim 5 is characterized in that: The second weight threshold range is set based on the statistical technology and the error range of the sensor in combination with the theoretical weight value, including: Obtain the theoretical weight value and sensor error range, calculate the measurement error based on the actual measurement data in the historical data and the corresponding theoretical data, and calculate the average value and standard deviation based on the measurement error; The range of the measurement error is set based on the mean value and standard deviation of the measurement error, and the second weight threshold range is set in combination with the error range of the sensor and the theoretical weight value; Among them, the second weight threshold range is: [W t -(E s +2s e ),W t +(E s +2s e )] Where W t Indicates the theoretical weight value, E s represents the sensor error range, σ e Represents standard deviation.

9. The paper cup weight intelligent weighing system based on artificial intelligence according to claim 1, characterized in that: The automatic calibration module (4) comprises an error feedback module, a calibration strategy generation module and a sensor calibration module; Wherein, the error feedback module is used to receive the measurement error data fed back by the intelligent analysis module; The calibration strategy generation module is used to generate a calibration strategy based on the measurement error data combined with the real-time environmental temperature and humidity data, and the calibration strategy includes adjusting the sensitivity, zero offset or environmental compensation parameters of the sensor; The sensor calibration module is used to execute the calibration strategy and adjust the parameters of the sensor.

10. An artificial intelligence-based intelligent weighing method for paper cup weight, implemented based on the artificial intelligence-based intelligent weighing system for paper cup weight as described in any one of claims 1 to 9, characterized in that: The paper cup weight intelligent weighing method based on artificial intelligence comprises the following steps: S1. Real-time collection of paper cup weight measurement data and environmental temperature and humidity data; S2, judging the qualification of the paper cup according to the comparison result of the weight measurement data and the first weight threshold range, and compensating and correcting the weight measurement data according to the comparison result of the qualified weight measurement data and the second weight threshold range to obtain the final weighing result; S3, display the final weighing result, qualified status or abnormal situation of the paper cup, and issue an early warning in abnormal situation; S4. Automatically calibrate the measurement parameters according to the results of intelligent analysis, and store and manage weighing data, analysis results and system operation records.

Images