Flowmeter calibration method and device, electronic equipment and storage medium

Abstract

The invention discloses a flowmeter calibration method and device, electronic equipment and a storage medium. The method comprises the steps that in the process that fluid flow collection is conducted on feed liquid added to target tobacco based on a target flowmeter, to-be-processed data of at least one dimension of the feed liquid under multiple preset durations is obtained, and the at least one dimension comprises at least one of the fluid temperature dimension, the fluid pressure dimension and the fluid flow velocity dimension; for a plurality of preset durations, inputting the to-be-processed data of at least one dimension under the current preset duration into a pre-trained uncertainty evaluation model, and determining the uncertainty of the target flowmeter under the current preset duration; when the uncertainty under the multiple preset durations meets the preset condition, the target calibration coefficient of the target flowmeter is determined on the basis of the uncertainty and the calibration coefficient determination model, and the hardware parameters of the target flowmeter are calibrated on the basis of the target calibration coefficient, so that the accuracy and efficiency of calibration of the target flowmeter are improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a flowmeter calibration method, device, electronic equipment and storage medium. Background Art

[0002] In the cigarette production process, flow meters are used to measure the flow of various fluids in the cigarette production process. For example, flow meters can be used to measure the flow of flavors, liquid materials, water, steam or compressed air in the cigarette production process. The accuracy of flow meters affects the quality of cigarette products, cigarette production efficiency, and fluid usage.

[0003] At present, the calibration of flow meters is mainly done by manually evaluating the flow data of the flow meters to obtain the uncertainty of the flow meters, and then calibrating the flow meters based on the uncertainty. However, manual evaluation may not only lead to inaccurate evaluation results, but also cause a certain lag in determining the uncertainty of the flow meters, which will affect the calibration of the flow meters and make it impossible to guarantee the real-time accuracy of the flow meters in the cigarette production process, thus affecting the overall production efficiency and production quality. Summary of the invention

[0004] The present invention provides a flow meter calibration method, device, electronic equipment and storage medium, which improve the accuracy and efficiency of calibrating a target flow meter.

[0005] According to one aspect of the present invention, a flow meter calibration method is provided, the method comprising:

[0006] In the process of collecting the fluid flow of the liquid added to the target tobacco based on the target flow meter, based on at least one sensor deployed at a preset position of the target flow meter, obtaining the to-be-processed data of at least one dimension of the liquid at multiple preset time lengths, wherein the at least one dimension includes at least one of the fluid temperature dimension, the fluid pressure dimension, and the fluid flow rate dimension;

[0007] For the to-be-processed data of at least one dimension under multiple preset time lengths, the to-be-processed data of at least one dimension under the current preset time length is input into a pre-trained uncertainty evaluation model to determine the uncertainty of the target flow meter under the current preset time length;

[0008] When the uncertainty under multiple preset time lengths meets the preset conditions, a model is determined based on the uncertainty and the calibration coefficient, a target calibration coefficient of the target flow meter is determined, and the hardware parameters of the target flow meter are calibrated based on the target calibration coefficient.

[0009] According to another aspect of the present invention, there is provided a flow meter calibration device, the device comprising:

[0010] A data acquisition module, used for acquiring the to-be-processed data of at least one dimension of the liquid added to the target tobacco under multiple preset time lengths based on at least one sensor deployed at a preset position of the target flow meter during the process of collecting the fluid flow of the liquid added to the target tobacco based on the target flow meter, wherein the at least one dimension includes at least one of the fluid temperature dimension, the fluid pressure dimension and the fluid flow rate dimension;

[0011] The uncertainty evaluation module is used to input the to-be-processed data of at least one dimension under multiple preset time lengths into a pre-trained uncertainty evaluation model to determine the uncertainty of the target flow meter under the current preset time length;

[0012] The flow meter calibration module is used to determine the target calibration coefficient of the target flow meter based on the uncertainty and calibration coefficient when the uncertainty under multiple preset time lengths meets the preset conditions, and calibrate the hardware parameters of the target flow meter based on the target calibration coefficient.

[0013] According to another aspect of the present invention, there is provided an electronic device, the electronic device comprising:

[0014] at least one processor; and

[0015] a memory communicatively connected to at least one processor; wherein,

[0016] The memory stores a computer program that can be executed by at least one processor. The computer program is executed by at least one processor so that the at least one processor can execute the flow meter calibration method according to any embodiment of the present invention.

[0017] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the flow meter calibration method of any embodiment of the present invention when executed.

[0018] According to another aspect of the present invention, a computer program product is provided, comprising a computer program, wherein the computer program, when executed by a processor, implements a flow meter calibration method according to any embodiment of the present invention.

[0019] The technical solution of the embodiment of the present invention is that in the process of collecting the fluid flow of the liquid added to the target tobacco based on the target flow meter, at least one sensor deployed at the preset position of the target flow meter is used to obtain the to-be-processed data of at least one dimension of the liquid measured by the target flow meter under multiple preset time lengths. Among them, at least one dimension includes at least one of the fluid temperature dimension, the fluid pressure dimension and the fluid flow rate dimension. For the to-be-processed data of at least one dimension under the multiple preset time lengths obtained, the to-be-processed data of at least one dimension under the current preset time length is input into a pre-trained uncertainty evaluation model to determine the uncertainty of the target flow meter under the current preset time length. It is determined whether the target flow meter needs to be calibrated based on the uncertainty under multiple preset time lengths. When the uncertainty under multiple preset time lengths meets the preset conditions, the target calibration coefficient of the target flow meter is determined according to the uncertainty and calibration coefficient determination model. The hardware parameters of the target flow meter are calibrated according to the target calibration coefficient. The present invention solves the problem that the real-time accuracy of the flow meter cannot be guaranteed due to inaccurate and untimely uncertainty evaluation of the flow meter in the prior art, improves the accuracy and efficiency of the calibration of the target flow meter, and ensures the measurement reliability of the target flow meter.

[0020] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1 is a flow chart of a flow meter calibration method provided by an embodiment of the present invention;

[0023] Figure 2 is a flow chart of a method for determining an uncertainty assessment model provided by an embodiment of the present invention;

[0024] Figure 3 is a structural schematic diagram of a flow meter calibration device provided by an embodiment of the present invention;

[0025] Figure 4 It is a schematic diagram of the structure of an electronic device for implementing the flow meter calibration method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0026] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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 should fall within the scope of protection of the present invention.

[0027] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0028] Embodiment 1

[0029] Figure 1 This is a flow chart of a flow meter calibration method provided by the first embodiment of the present invention. This embodiment is applicable to the case of real-time calibration of a target flow meter in a cigarette production process. The method can be executed by a flow meter calibration device. The flow meter calibration device can be implemented in the form of hardware and / or software. The flow meter calibration device can be configured in an electronic device such as a mobile phone, a computer or a server. Figure 1 As shown, the method includes:

[0030] S110. In the process of collecting fluid flow of the liquid added to the target tobacco based on the target flow meter, based on at least one sensor deployed at a preset position of the target flow meter, obtain the data to be processed of at least one dimension of the liquid at multiple preset time lengths.

[0031] Among them, at least one dimension includes at least one of a fluid temperature dimension, a fluid pressure dimension, and a fluid flow rate dimension.

[0032] Multiple flow meters are deployed in the cigarette production process, and these flow meters are used to measure the flow of various liquids added to tobacco during the cigarette production process. In the embodiment of the present invention, the calibration method for each flow meter is similar, so the current flow meter is used as the target flow meter. The tobacco corresponding to the current flow meter is used as the target tobacco. That is, the target flow meter collects the flow rate of the liquid fluid added to the target tobacco. It should be noted that in the embodiment of the present invention, the liquid can be a fluid such as a fragrance liquid, liquid, and water sprayed on the target tobacco, and the present invention does not limit the type of liquid.

[0033] In the process of the target flow meter measuring the flow rate of the liquid added to the target tobacco, at least one dimension of data to be processed can be collected by at least one sensor. The at least one sensor may include at least one of a temperature sensor, a pressure sensor, and a flow rate sensor. Among them, the temperature sensor is used to collect the fluid temperature data of the liquid; the pressure sensor is used to collect the fluid pressure data of the liquid; and the flow rate sensor is used to collect the fluid flow rate data of the liquid. Accordingly, the data to be processed in at least one dimension may include at least one of the fluid temperature data, the fluid pressure data, and the fluid flow rate data.

[0034] To improve the accuracy of data collection, at least one sensor can be deployed at a preset position of the target flow meter. Optionally, at least one sensor can be deployed at an upstream or downstream position of the target flow meter. It should be noted that in order to improve the richness of subsequent data, the data to be processed can be collected separately at multiple preset time lengths corresponding to the slurry fluid flow collection process of the target flow meter. The preset time length can be a period of time in the slurry fluid flow collection process set according to actual needs.

[0035] Specifically, in the process of collecting the fluid flow of the liquid added to the target tobacco through the target flow meter, at least one sensor deployed at a preset position of the target flow meter is used to obtain the data to be processed in the dimension corresponding to each sensor at multiple preset time lengths. The data to be processed may be data in at least one dimension of the fluid temperature dimension, the fluid pressure dimension, and the fluid flow rate dimension.

[0036] In an embodiment of the present invention, a method for obtaining data to be processed may be: for multiple preset time periods corresponding to a slurry fluid flow acquisition process, based on at least one sensor deployed at a preset position of a target flow meter, obtaining the original data of the slurry in at least one dimension at each preset time period; performing data preprocessing on the original data to obtain first data corresponding to the original data; for at least one dimension, determining the second data in the current dimension based on the first data in the current dimension and a processing function corresponding to the current dimension; and using the first data and the second data in at least one dimension as data to be processed.

[0037] The raw data may be data acquired by a sensor. Since the raw data may contain noise data, abnormal data or missing data, the raw data may be preprocessed. Optionally, the data preprocessing includes at least one of data cleaning, data standardization and data filling. The first data is the data obtained after the raw data is preprocessed.

[0038] The processing function may be a function for processing the first data to obtain the derivative data corresponding to the first data. The processing functions corresponding to each dimension may be the same or different. Accordingly, the derivative data corresponding to the first data is the second data mentioned in the embodiment of the present invention. The data to be processed includes the first data and the second data.

[0039] Specifically, for multiple preset time periods corresponding to the slurry fluid flow acquisition process, according to at least one sensor deployed at a preset position of the target flow meter, the raw data of the slurry in at least one dimension under each preset time period is obtained according to the preset acquisition frequency. The raw data is cleaned to remove noise data and abnormal data in the raw data. Optionally, a Python script can be used to perform data cleaning on the raw data. The raw data is standardized by a preset standardization function to obtain standardized data. Among them, the preset standardization function can be a Z-score standardization function. The standardized data is filled with data to supplement missing data. Optionally, a multiple interpolation strategy of neighboring data can be used to fill missing data. Through the above data preprocessing process, the first data corresponding to the raw data is obtained. For at least one dimension, the first data under the dimension is processed according to the processing function corresponding to each dimension to obtain the second data corresponding to each dimension. The first data and the second data of at least one dimension are used as data to be processed.

[0040] Exemplarily, a temperature sensor, a pressure sensor and a flow rate sensor deployed upstream and / or at a target flow meter are used to collect fluid temperature data, fluid pressure data and fluid flow rate data of the slurry fluid at a preset collection frequency of N times per second. The collected fluid temperature data, fluid pressure data and fluid flow rate data are stored as raw data in a corresponding database for subsequent data processing. The raw data is cleaned by a Python script to remove noise and outliers in the raw data to obtain cleaned raw data. The cleaned raw data is standardized using a Z-score standardization function to obtain standardized raw data. The standardized raw data is filled with data using a multiple interpolation strategy of neighboring data to obtain first data. For first data in different dimensions, the first data is processed by the data to be processed corresponding to each dimension to obtain second data in each dimension. The first data in at least one dimension and the second data corresponding to the first data are used as data to be processed.

[0041] In the embodiment of the present invention, the second data in different dimensions are determined in different ways. Specifically, it can be: for the fluid temperature dimension and the fluid flow velocity dimension corresponding to the target flow meter, based on the first data of the current dimension in at least one time slice of a preset duration, determine the first data difference corresponding to the first data; determine the derivative of the first data difference with respect to the time slice, and use the derivative as the second data in the current dimension.

[0042] The first data difference may be a difference between first data corresponding to a start time and an end time in a time slice.

[0043] Specifically, for the fluid temperature dimension, the first data difference of each time slice is determined based on the first data under at least one time slice of the preset duration. The second-order derivative of the first data difference relative to the time slice is determined. It should be noted that since there is at least one time slice under the preset duration, at least one second-order derivative result corresponding to at least one time slice can be obtained. Then, the at least one second-order derivative result can be averaged, and the averaged result is used as the second data corresponding to the current preset duration.

[0044] Among them, the second-order derivative function can be expressed as follows:

[0045]

[0046] Wherein, T represents the first data difference corresponding to the fluid temperature dimension, t represents the time slice, and T” represents the second-order derivative result.

[0047] For the fluid velocity dimension, the first data difference of each time slice is determined based on the first data of at least one time slice of the preset duration. The first-order derivative of the first data difference relative to the time slice under the fluid velocity dimension is determined. The at least one first-order derivative result is averaged, and the averaged result is used as the second data corresponding to the current preset duration.

[0048] Among them, the first-order derivative function is expressed as follows,

[0049]

[0050] Wherein, V represents the first data difference corresponding to the fluid velocity dimension, t represents the time slice, and V' represents the first-order derivative result.

[0051] Optionally, for the fluid flow rate dimension and the fluid temperature dimension, the first data difference corresponding to the first data can be determined based on the first data corresponding to the current dimension at the start time and the end time of the preset duration, and the second-order derivative result of the first data difference relative to the preset duration is used as the second data in the current dimension.

[0052] For the fluid pressure dimension corresponding to the target flow meter, determine the frequency domain data corresponding to the first data based on the first data under at least one time slice under a preset time length, the number of data collection times corresponding to the time slice, and the frequency domain processing function; and use the frequency domain data as the second data under the fluid pressure dimension.

[0053] The number of data acquisition times may be determined based on the time slice and the preset acquisition frequency. The frequency domain processing function may be a function determined based on the fast Fourier transform and used to calculate the frequency spectrum feature data corresponding to the first data. The frequency spectrum feature data corresponding to the first data is the frequency domain data, that is, the second data under the fluid pressure dimension.

[0054] Specifically, for the fluid pressure dimension, the first data and the number of data collection times in at least one time slice under a preset time length are substituted into the frequency domain processing function to obtain the frequency domain data corresponding to the first data, that is, the second data in the fluid pressure dimension.

[0055] Optionally, the first data in at least one time slice under a preset time length may be subjected to a fast Fourier transform process to obtain frequency domain signal data corresponding to the first data. Power spectrum data may be obtained based on the absolute value square data corresponding to the frequency domain signal data. Feature extraction may be performed on the power spectrum data to obtain spectrum feature data, i.e., frequency domain data.

[0056] Specifically, the first data in at least one time slice under the preset time length is processed by fast Fourier transform to obtain frequency domain signal data corresponding to the first data. The fast Fourier transform function can be expressed as follows:

[0057]

[0058] Among them, X(k) represents the frequency domain signal data, n represents the nth sampling point in the time slice, and N represents the number of data acquisition times in the time slice, that is, the number of sampling points. x(n) represents the fluid pressure data at the nth sampling point, j represents the imaginary unit, and k represents the discrete frequency index.

[0059] According to the frequency domain signal data and the amplitude spectrum determination function, the amplitude spectrum data corresponding to the frequency domain signal data is determined. The amplitude spectrum determination function is expressed as follows:

[0060] As=|X(k)

[0061] Wherein, As represents the amplitude spectrum data.

[0062] The power spectrum data is obtained according to the amplitude spectrum data and the power spectrum determination function. The power spectrum determination function is expressed as follows:

[0063] Ps=As 2 =|X(k) 2

[0064] Wherein, Ps represents power spectrum data.

[0065] Perform feature extraction processing on the power spectrum data corresponding to at least one time slice within the current preset duration to obtain the frequency spectrum feature data of the current preset duration, that is, frequency domain data. For example, based on the power spectrum data corresponding to at least one time slice within the current preset duration, determine the frequency corresponding to the maximum power spectrum data, or determine the frequency range of energy distribution in the power spectrum.

[0066] S120. For the data to be processed in at least one dimension under multiple preset time lengths, input the data to be processed in at least one dimension under the current preset time length into a pre-trained uncertainty evaluation model to determine the uncertainty of the target flow meter under the current preset time length.

[0067] Among them, the uncertainty assessment model can be used to predict the measurement uncertainty of the target flow meter under the current preset time length. Optionally, the uncertainty assessment model can be a multi-layer neural network model determined based on the Python programming language and the TensorFlow machine learning model. Uncertainty is the abbreviation of measurement uncertainty, which can be understood as the degree of doubt about the correctness or accuracy of the flow measurement results of the target flow meter. Uncertainty indicates the possible difference between the flow measurement results of the target flow meter and the actual flow. The smaller the uncertainty, the more accurate the target flow meter. Correspondingly, the larger the uncertainty, the less accurate the target flow meter is, and the target flow meter can be calibrated.

[0068] Specifically, for the data to be processed in at least one dimension under multiple preset time lengths, the data to be processed in at least one dimension under each preset time length is processed by a pre-trained uncertainty assessment model to obtain the uncertainty of the target flow meter corresponding to each preset time length.

[0069] S130. When the uncertainty under multiple preset time lengths meets preset conditions, determine the model based on the uncertainty and the calibration coefficient, determine the target calibration coefficient of the target flow meter, and calibrate the hardware parameters of the target flow meter based on the target calibration coefficient.

[0070] The preset condition may be that the uncertainty range corresponding to the uncertainty under multiple preset time lengths does not meet the preset uncertainty range of the target flow meter. The calibration coefficient determination model may be a mathematical model for determining the calibration coefficient of the target flow meter. The target calibration coefficient may be a coefficient of a hardware parameter that needs to be calibrated for the current target flow meter, determined based on the uncertainty under multiple preset time lengths. The hardware parameter may be a parameter of at least one hardware component in the target flow meter.

[0071] Specifically, the uncertainty range corresponding to the target flow meter is determined according to the uncertainties under multiple preset time lengths. When the uncertainty range belongs to the preset uncertainty range, it means that the accuracy of the target flow meter meets the requirements and no calibration is required. However, when the uncertainty range does not meet the preset uncertainty range, it means that the target flow meter needs to be calibrated. The uncertainty under multiple preset conditions is processed according to the calibration coefficient determination model to obtain the target calibration coefficient of the target flow meter. The hardware parameters of the target flow meter are calibrated by the target calibration coefficient to improve the measurement accuracy of the target flow meter.

[0072] In an embodiment of the present invention, the target calibration coefficient may be determined by: determining the mean data corresponding to the uncertainty under multiple preset time lengths, and determining a function based on the mean data and the confidence interval to obtain a data range to be compared corresponding to the uncertainty under multiple preset time lengths; when the data range to be compared does not satisfy the preset data range, determining an error value corresponding to the data range to be compared and the preset data range; determining the target calibration coefficient corresponding to the target flow meter based on the error value, the historical calibration coefficient corresponding to the target flow meter, the preset adjustment weight coefficient, and the calibration coefficient determination model.

[0073] Among them, the mean data can be understood as the mean of uncertainty under multiple preset time lengths. The confidence interval determination function can be used to determine the function of the uncertainty range of the target flow meter under a preset confidence level. Optionally, the preset confidence level can be a confidence level set according to actual needs. For example, the confidence level can be 95%. The data range to be compared is the current uncertainty range of the target flow meter. The data range to be compared can be a pre-set standard range of uncertainty of the target flow meter. The error value can be a deviation value of the uncertainty of the target flow meter determined based on the data range to be compared and the data range to be compared. The historical calibration coefficient can be understood as the calibration number corresponding to the last calibration of the target flow meter. The preset adjustment weight coefficient can be an adjustment parameter determined according to actual needs.

[0074] Specifically, the mean data is determined according to the uncertainty under multiple preset time lengths. The mean data can be determined by the following function.

[0075]

[0076] Among them, x i represents the uncertainty corresponding to the i-th preset duration, n represents the number of preset durations, Represents mean data. According to the mean data and the uncertainty corresponding to each preset time length, the standard deviation data corresponding to the uncertainty under multiple preset time lengths is determined. Substitute the mean data, standard deviation data and the number of preset time lengths into the confidence interval determination function to obtain the data range to be compared corresponding to the target flow meter. Among them, the confidence interval determination function can be expressed as follows:

[0077]

[0078] in, represents mean data, n represents the number of preset time lengths, s represents standard deviation data, and z represents the z value corresponding to the confidence level. The z value can be determined by looking up the table. For example, when the confidence level is 95%, the z value is approximately 1.96.

[0079] After obtaining the data range to be compared through the confidence interval determination function, determine whether the data range to be compared meets the preset data range. If the data range to be compared meets the preset data range, the target flow meter does not need to be calibrated. If the data range to be compared does not meet the preset data range, it means that the target flow meter needs to be calibrated. According to the data range to be compared and the preset data range, determine the error value. According to the error value, the historical calibration coefficient and the preset adjustment weight coefficient, use the calibration coefficient determination model to obtain the target calibration coefficient.

[0080] Optionally, the calibration coefficient determination model may be a data model for determining the calibration coefficient, which may be expressed as follows:

[0081] k new =k old +Δk*sign(e)

[0082] Among them, k old represents the historical calibration coefficient, Δk represents the preset adjustment weight coefficient, e represents the error value, k new Optionally, a coefficient range of the target calibration coefficient may be set so that the target calibration coefficient belongs to the corresponding coefficient range, thereby avoiding over-calibration of the target flow meter.

[0083] Optionally, when the data range to be compared does not satisfy the preset data range, it also includes: generating warning information based on the data range to be compared and the data to be processed corresponding to the target flow meter, and performing warning processing based on the warning information.

[0084] The preset data range may be a pre-set standard range of uncertainty of the target flow meter. The warning information may be information for reminding relevant personnel to calibrate or replace the target flow meter. Optionally, the warning information may be sent in a variety of ways such as broadcast, SMS or email.

[0085] Specifically, when the data range to be compared does not meet the preset data range, a warning message can be generated based on the data range to be compared and the data to be processed of the target flow meter, so as to remind the corresponding staff to calibrate or replace the target flow meter through the warning message.

[0086] The technical solution of this embodiment is that in the process of collecting the fluid flow of the liquid added to the target tobacco based on the target flow meter, at least one sensor deployed at the preset position of the target flow meter is used to obtain the data to be processed of at least one dimension of the liquid measured by the target flow meter under multiple preset time lengths. Among them, at least one dimension includes at least one of the fluid temperature dimension, the fluid pressure dimension and the fluid flow rate dimension. For the data to be processed of at least one dimension under the multiple preset time lengths obtained, the data to be processed of at least one dimension under the current preset time length is input into a pre-trained uncertainty evaluation model to determine the uncertainty of the target flow meter under the current preset time length. It is determined whether the target flow meter needs to be calibrated based on the uncertainty under multiple preset time lengths. When the uncertainty under multiple preset time lengths meets the preset conditions, the target calibration coefficient of the target flow meter is determined according to the uncertainty and calibration coefficient determination model. The hardware parameters of the target flow meter are calibrated according to the target calibration coefficient. The present invention solves the problem that the real-time accuracy of the flow meter cannot be guaranteed due to inaccurate and untimely uncertainty evaluation of the flow meter in the prior art, improves the accuracy and efficiency of the calibration of the target flow meter, and ensures the measurement reliability of the target flow meter.

[0087] Embodiment 2

[0088] Figure 2 This is a flow chart of a method for determining an uncertainty assessment model provided by the second embodiment of the present invention. This embodiment is based on the above embodiment. Before processing the data to be processed according to the uncertainty assessment model, a trained uncertainty assessment model can be first trained. The specific implementation method can refer to the technical solution of this embodiment. Among them, the technical terms that are the same as or corresponding to the above embodiment are not repeated here. Figure 2 As shown, the method includes:

[0089] S210: Acquire multiple training samples, wherein the training samples include: sample data and theoretical uncertainty corresponding to the sample data.

[0090] Among them, the sample data may include sample first data and sample second data corresponding to at least one dimension under a historical preset time length. The historical preset time length may be a period of time set according to actual needs during the historical collection of the flow rate of the slurry fluid. The sample first data is data obtained after preprocessing the data collected by at least one sensor under the historical preset time length. The sample second data may be data obtained after processing the sample first data through a processing function of the corresponding dimension. The theoretical uncertainty may be the uncertainty of the target flow meter actually measuring the slurry fluid under the historical preset time length.

[0091] Specifically, before training the uncertainty assessment model to be trained, multiple training samples can be obtained to train the model through the training samples. In order to improve the accuracy of the uncertainty assessment model, as many and rich training samples as possible can be obtained. That is, the first sample data of at least one dimension under multiple historical preset time lengths are obtained. And the first sample data is processed by the processing function under each dimension to obtain the second sample data corresponding to the first sample data. The first sample data and the second sample data are used as sample data to train the uncertainty assessment model to be trained through the sample data.

[0092] S220: Input the sample data in the training sample into the uncertainty assessment model to be trained to obtain the prediction uncertainty.

[0093] It should be noted that, for each training sample, the method of S220 can be used to train it, so as to obtain an uncertainty assessment model.

[0094] The model parameters in the uncertainty assessment model are default values. The model parameters in the uncertainty assessment model to be trained are corrected through training samples to obtain a trained uncertainty assessment model. The predicted uncertainty can be the uncertainty of the target flow meter output after the sample data is input into the uncertainty assessment model to be trained.

[0095] Specifically, the sample data in the training samples are input into the uncertainty assessment model to be trained to obtain the prediction uncertainty, so as to correct the model parameters of the uncertainty assessment model to be trained through the prediction uncertainty.

[0096] Exemplarily, a multi-layer neural network model determined based on Python language and TensorFlow machine learning model is used as an example for explanation. The uncertainty assessment model to be trained may include an input layer, a hidden layer, and an output layer. The input layer can be used to determine m input features corresponding to sample data of at least one dimension. The hidden layer may include 12 neurons for feature extraction processing of the input features. Optionally, the hidden layer may use a ReLU activation function to process the input features. The output layer may include 1 neuron for outputting prediction uncertainty. Among them, the output layer may use a Sigmoid activation function.

[0097] The sample data of the training sample is input into the uncertainty assessment model to be trained, and the sample data is processed through the input layer to obtain m input features corresponding to the sample data of at least one dimension. According to the number of sample data N and the m input features corresponding to each sample data, an N×m matrix X is obtained. The matrix X is linearly transformed through the hidden layer to obtain the feature z to be used (1) . The process of linear transformation can be shown as follows.

[0098] z (1) =XW (1) +b (1)

[0099] Where X represents the input data of the hidden layer, which is a matrix with dimension N×m. (1) Represents the weight matrix from the input layer to the hidden layer, which is a matrix of dimension m×12. (1) Represents the bias vector of the hidden layer, with a dimension of 1×12. (1) is the feature to be used, which is a matrix with a dimension of N×12. The feature to be used is processed by the ReLU activation function to obtain the activation value a of the hidden layer (1) .

[0100] a (1) =ReLU(z (1) )

[0101] Among them, ReLU(z (1) )=max(0,z (1)). Perform linear transformation on the activation value of the hidden layer to obtain the features to be processed of the output layer. Use the Sigmoid activation function to process the features to be processed of the output layer to obtain the prediction uncertainty. The specific process can be as follows:

[0102] z (2) =a (1) W (2) +b (2)

[0103] Among them, a (1) W represents the activation value of the hidden layer, with a dimension of N×12. (2) Represents the weight matrix from the hidden layer to the output layer, which is a matrix of dimension 12×1. (2) is the bias vector of the output layer, with a dimension of 1×1, z (2) Represents the features to be processed in the output layer, which is a matrix with dimension N×1. The features to be processed z in the output layer are activated by the Sigmoid activation function. (2) Processing to obtain the prediction uncertainty Right now,

[0104]

[0105] in, represents the prediction uncertainty, a (2) represents the activation value of the output layer, z (2) Represents the features to be processed in the output layer.

[0106] S230. Determine the loss value based on the theoretical uncertainty and the predicted uncertainty.

[0107] The loss value is determined based on the loss function and is used to characterize the degree of difference between the theoretical uncertainty and the predicted uncertainty. Optionally, the loss function may be a mean square error function.

[0108] Specifically, the theoretical uncertainty and the predicted uncertainty are processed according to the loss function to obtain the difference between the theoretical uncertainty and the actual uncertainty, that is, the loss value, so as to correct the model parameters of the uncertainty assessment model to be trained.

[0109] Exemplarily, in combination with the above example, the loss function is taken as a mean square error function as an example for explanation.

[0110]

[0111] Where N represents the number of sample data, y i represents the theoretical uncertainty corresponding to the i-th sample data, Represents the prediction uncertainty corresponding to the i-th sample data. L represents the loss value.

[0112] S240: Based on at least one preset optimization algorithm and loss value, modify the model parameters of the uncertainty assessment model to be trained to obtain a trained uncertainty assessment model.

[0113] In general, the model parameters of the uncertainty assessment model to be trained are initial parameters or default parameters. When the uncertainty assessment model to be trained is trained, the model parameters in the model can be corrected based on the output results of the uncertainty assessment model to be trained, that is, the loss value of the uncertainty assessment model to be trained can be corrected to obtain a trained uncertainty assessment model. Among them, the preset optimization algorithm can be used to adjust the model parameters and the number of iterations to optimize the model performance. Optionally, the preset optimization algorithm can be an optimization algorithm such as the AdaGrad optimization algorithm.

[0114] Specifically, when the model parameters in the uncertainty assessment model to be trained are corrected using at least one preset optimization algorithm and loss value, the convergence of the loss function can be used as a training goal, such as whether the training error is less than the preset error, or whether the error change tends to be stable, or whether the current number of iterations is equal to the preset number. If the detection reaches the convergence condition, such as the training error of the loss function is less than the preset error, or the error change trend tends to be stable, it indicates that the training of the uncertainty assessment model to be trained is completed, and the iterative training can be stopped at this time. If it is detected that the convergence condition is not met at present, other training samples can be further obtained to continue training the uncertainty assessment model to be trained until the training error of the loss function is within the preset range. When the training error of the loss function reaches convergence, the uncertainty assessment model that has been trained can be obtained, that is, after the data to be processed is input into the uncertainty assessment model, the uncertainty of the target flow meter corresponding to the data to be processed can be accurately obtained.

[0115] Exemplarily, in combination with the above example, the gradient of the loss function relative to the linear transformation of the output layer is determined, and the gradient of the loss function relative to the linear transformation of the hidden layer is determined. Specifically, it can be:

[0116]

[0117] Among them, L represents the loss value, z (2) represents the features to be processed in the output layer, a (2) represents the activation value of the output layer, N represents the number of sample data, and y represents the theoretical uncertainty of the sample data.

[0118]

[0119] Among them, L represents the loss value, z (1) represents the features to be used in the hidden layer, z(2) represents the features to be processed in the output layer, W (2)T Represents the transposed matrix of the weight matrix from the hidden layer to the output layer, ReLU'(z (1) ) represents the derivative of the ReLU activation function and is defined as 1.

[0120] Determine the gradient of the loss function with respect to the weights and biases of the output layer, and determine the gradient of the loss function with respect to the weights and biases of the hidden layer.

[0121] The weight gradient of the loss function to the output layer is:

[0122] Among them, L represents the loss value, W (2) represents the weight matrix from the hidden layer to the output layer, a (1) represents the transpose of the activation value of the hidden layer, z (2) Represents the features to be processed in the output layer.

[0123] The bias gradient of the loss function to the output layer is:

[0124] Among them, L represents the loss value, b (2) is the bias vector of the output layer, N represents the number of sample data, z (2) Represents the features to be processed in the output layer.

[0125] The weight gradient of the loss function with respect to the hidden layer is:

[0126] Among them, L represents the loss value, W (1) represents the weight matrix from the input layer to the hidden layer, X represents the transpose of the input data of the hidden layer, and z (1) is the feature to be used in the hidden layer.

[0127] The bias gradient of the loss function to the hidden layer is:

[0128] Among them, L represents the loss value, b (1) represents the bias vector of the hidden layer, N represents the number of sample data, z (1) is the feature to be used in the hidden layer.

[0129] The model parameters of the uncertainty assessment model to be trained are modified by the AdaGrad optimization algorithm. Specifically, the gradient square accumulation matrix G is initialized. (1) and G (2) is a zero matrix, and the gradient square accumulation matrix is ​​updated, that is:

[0130]

[0131] Among them, G(1) ' and G (2) ' represents the updated gradient square accumulation matrix.

[0132] Use the AdaGrad optimization algorithm to update the weights and biases of the hidden layer and output layer. Specifically, it can be:

[0133]

[0134] Among them, W (1) ' represents the updated weight matrix of the hidden layer, b (1) ' represents the updated bias vector of the hidden layer, W (2) ' represents the updated weight matrix of the output layer, b (2) ' represents the bias vector of the updated output layer, η represents the learning rate, and ε is a constant used to maintain data stability.

[0135] The above method is used to modify the model parameters in the uncertainty assessment model to be trained by using at least one preset optimization algorithm and loss value to obtain a trained uncertainty assessment model.

[0136] Optionally, after obtaining the trained uncertainty assessment model, the k-fold cross validation and correlation coefficient (R 2 ) and the corresponding test samples to evaluate the model performance of the trained uncertainty evaluation model, so that when the performance evaluation results meet the requirements, the data to be processed can be processed based on the trained uncertainty evaluation model. Optionally, the performance and algorithm of the uncertainty evaluation model can be continuously optimized based on the actual use information and user feedback information of the uncertainty evaluation model. Regularly calibrate the sensor and target flow meter to ensure the accuracy of the collected data.

[0137] The technical solution of this embodiment obtains a plurality of training samples, and trains the uncertainty assessment model to be trained by the plurality of training samples to obtain the prediction uncertainty. The loss value is determined by the prediction uncertainty and the theoretical uncertainty in the training samples. The model parameters of the uncertainty assessment model to be trained are corrected according to at least one loss value to obtain a trained uncertainty assessment model, and the uncertainty of the target flow meter is obtained by processing the processed data by the trained uncertainty assessment model, thereby improving the accuracy and efficiency of the uncertainty assessment, thereby improving the accuracy and efficiency of the calibration of the target flow meter, and ensuring the measurement reliability of the target flow meter.

[0138] Embodiment 3

[0139] Figure 3 Schematic diagram of the structure of a flow meter calibration device provided by the third embodiment of the present invention. Figure 3As shown, the device includes: a data acquisition module 310 , an uncertainty evaluation module 320 and a flow meter calibration module 330 .

[0140] The data acquisition module 310 is used to acquire the to-be-processed data of at least one dimension of the slurry added to the target tobacco under multiple preset time lengths based on at least one sensor deployed at a preset position of the target flow meter during the process of collecting the fluid flow of the slurry added to the target tobacco based on the target flow meter, wherein the at least one dimension includes at least one of the fluid temperature dimension, the fluid pressure dimension and the fluid flow rate dimension; the uncertainty assessment module 320 is used to input the to-be-processed data of at least one dimension under the current preset time length into a pre-trained uncertainty assessment model for the to-be-processed data of at least one dimension under the multiple preset time lengths, and determine the uncertainty of the target flow meter under the current preset time length; the flow meter calibration module 330 is used to determine the target calibration coefficient of the target flow meter based on the uncertainty and calibration coefficient determination model when the uncertainty under multiple preset time lengths meets the preset conditions, and calibrate the hardware parameters of the target flow meter based on the target calibration coefficient.

[0141] The technical solution of this embodiment is that in the process of collecting the fluid flow of the liquid added to the target tobacco based on the target flow meter, at least one sensor deployed at the preset position of the target flow meter is used to obtain the data to be processed of at least one dimension of the liquid measured by the target flow meter under multiple preset time lengths. Among them, at least one dimension includes at least one of the fluid temperature dimension, the fluid pressure dimension and the fluid flow rate dimension. For the data to be processed of at least one dimension under the multiple preset time lengths obtained, the data to be processed of at least one dimension under the current preset time length is input into a pre-trained uncertainty evaluation model to determine the uncertainty of the target flow meter under the current preset time length. It is determined whether the target flow meter needs to be calibrated based on the uncertainty under multiple preset time lengths. When the uncertainty under multiple preset time lengths meets the preset conditions, the target calibration coefficient of the target flow meter is determined according to the uncertainty and calibration coefficient determination model. The hardware parameters of the target flow meter are calibrated according to the target calibration coefficient. The present invention solves the problem that the real-time accuracy of the flow meter cannot be guaranteed due to inaccurate and untimely uncertainty evaluation of the flow meter in the prior art, improves the accuracy and efficiency of the calibration of the target flow meter, and ensures the measurement reliability of the target flow meter.

[0142] On the basis of the above embodiment, optionally, the data acquisition module includes: a raw data acquisition unit, which is used to obtain the raw data of the slurry in at least one dimension under each preset time length for multiple preset time lengths corresponding to the slurry fluid flow acquisition process based on at least one sensor deployed at a preset position of the target flow meter; a first data determination unit, which is used to perform data preprocessing on the raw data to obtain first data corresponding to the raw data; a second data determination unit, which is used to determine, for at least one dimension, the second data in the current dimension based on the first data in the current dimension and the processing function corresponding to the current dimension; and a data to be processed determination unit, which is used to use the first data and the second data in at least one dimension as data to be processed.

[0143] Optionally, a second data determination unit is used to determine, for the fluid temperature dimension and the fluid flow rate dimension corresponding to the target flow meter, the first data corresponding to the first data of the current dimension under at least one time slice of a preset time length; determine the derivative of the first data difference with respect to the time slice, and use the derivative as the second data under the current dimension.

[0144] Optionally, a second data determination unit is used to determine the frequency domain data corresponding to the first data for the fluid pressure dimension corresponding to the target flow meter based on the first data in at least one time slice under a preset time length, the number of data collection times corresponding to the time slice, and the frequency domain processing function; and use the frequency domain data as the second data in the fluid pressure dimension.

[0145] Optionally, the device also includes: a model training module, used to obtain multiple training samples, wherein the training samples include: sample data and theoretical uncertainty corresponding to the sample data; inputting the sample data in the training samples into the uncertainty assessment model to be trained to obtain predicted uncertainty; determining a loss value based on the theoretical uncertainty and the predicted uncertainty; and correcting model parameters of the uncertainty assessment model to be trained based on at least one preset optimization algorithm and the loss value to obtain a trained uncertainty assessment model.

[0146] Optionally, the flow meter calibration module includes: a target calibration coefficient determination unit, used to determine the mean data corresponding to the uncertainty under multiple preset time lengths, and determine the function based on the mean data and the confidence interval to obtain the data range to be compared corresponding to the uncertainty under multiple preset time lengths; when the data range to be compared does not meet the preset data range, determine the error value corresponding to the data range to be compared and the preset data range; based on the error value, the historical calibration coefficient corresponding to the target flow meter, the preset adjustment weight coefficient and the calibration coefficient determination model, determine the target calibration coefficient corresponding to the target flow meter.

[0147] Optionally, when the data range to be compared does not satisfy the preset data range, the device further includes: an information warning module, which is used to generate warning information based on the data range to be compared and the data to be processed corresponding to the target flow meter, and perform warning processing based on the warning information.

[0148] The flow meter calibration device provided in the embodiment of the present invention can execute the flow meter calibration method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0149] Embodiment 4

[0150] Figure 4 1 is a schematic diagram of the structure of an electronic device provided in Embodiment 4 of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0151] like Figure 4 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0152] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0153] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a flow meter calibration method.

[0154] In some embodiments, the flow meter calibration method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the flow meter calibration method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to perform the flow meter calibration method in any other appropriate manner (e.g., by means of firmware).

[0155] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0156] The computer program for implementing the flow meter calibration method of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program can be executed entirely on the machine, partially on the machine, partially on the machine as a stand-alone software package and partially on a remote machine, or entirely on a remote machine or server.

[0157] Embodiment 5

[0158] Embodiment 5 of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to cause a processor to execute a flow meter calibration method, the method comprising:

[0159] In the process of collecting the fluid flow of the slurry added to the target tobacco based on the target flow meter, based on at least one sensor deployed at a preset position of the target flow meter, the to-be-processed data of at least one dimension of the slurry under multiple preset time lengths are obtained, wherein the at least one dimension includes at least one of the fluid temperature dimension, the fluid pressure dimension and the fluid flow rate dimension; for the to-be-processed data of at least one dimension under multiple preset time lengths, the to-be-processed data of at least one dimension under the current preset time length are input into a pre-trained uncertainty assessment model to determine the uncertainty of the target flow meter under the current preset time length; when the uncertainty under multiple preset time lengths meets the preset conditions, the target calibration coefficient of the target flow meter is determined based on the uncertainty and calibration coefficient determination model, and the hardware parameters of the target flow meter are calibrated based on the target calibration coefficient.

[0160] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0161] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0162] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0163] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0164] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0165] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A flow meter calibration method, characterized in that: include: In the process of collecting the fluid flow of the liquid added to the target tobacco based on the target flow meter, based on at least one sensor deployed at a preset position of the target flow meter, obtaining the data to be processed of at least one dimension of the liquid at multiple preset time lengths, wherein the at least one dimension includes at least one of a fluid temperature dimension, a fluid pressure dimension, and a fluid flow rate dimension; For the data to be processed in at least one dimension under the multiple preset time lengths, input the data to be processed in at least one dimension under the current preset time length into a pre-trained uncertainty evaluation model to determine the uncertainty of the target flow meter under the current preset time length; When the uncertainty under the multiple preset time lengths meets the preset conditions, a model is determined based on the uncertainty and the calibration coefficient, the target calibration coefficient of the target flow meter is determined, and the hardware parameters of the target flow meter are calibrated based on the target calibration coefficient.

2. The method according to claim 1, characterized in that The step of obtaining the to-be-processed data of at least one dimension of the slurry at multiple preset time lengths includes: For a plurality of preset time periods corresponding to the slurry fluid flow rate collection process, based on at least one sensor deployed at a preset position of the target flow meter, obtaining raw data of the slurry in at least one dimension at each preset time period; Performing data preprocessing on the original data to obtain first data corresponding to the original data; For at least one dimension, determining second data under the current dimension according to first data under the current dimension and a processing function corresponding to the current dimension; The first data and the second data in the at least one dimension are used as the data to be processed.

3. The method according to claim 2, characterized in that The determining, according to the first data in the current dimension and the processing function corresponding to the current dimension, the second data in the current dimension comprises: For the fluid temperature dimension and the fluid flow velocity dimension corresponding to the target flow meter, determine a first data difference corresponding to the first data based on the first data of the current dimension under at least one time slice of the preset time length; Determine a derivative of the first data difference with respect to the time slice, and use the derivative as the second data in the current dimension.

4. The method according to claim 2, characterized in that: The determining, according to the first data in the current dimension and the processing function corresponding to the current dimension, the second data in the current dimension comprises: For the fluid pressure dimension corresponding to the target flow meter, based on the first data of at least one time slice under the preset time length, the number of data collection times corresponding to the time slice, and the frequency domain processing function, determine the frequency domain data corresponding to the first data; The frequency domain data is used as the second data under the fluid pressure dimension.

5. The method according to claim 1, characterized in that The method further comprises: Acquire a plurality of training samples, wherein the training samples include: sample data and theoretical uncertainty corresponding to the sample data; Inputting sample data in the training sample into the uncertainty assessment model to be trained to obtain prediction uncertainty; Determining a loss value based on the theoretical uncertainty and the predicted uncertainty; Based on at least one preset optimization algorithm and the loss value, the model parameters of the uncertainty assessment model to be trained are modified to obtain a trained uncertainty assessment model.

6. The method according to claim 1, characterized in that When the uncertainty under the plurality of preset time lengths satisfies a preset condition, determining a target calibration coefficient of the target flow meter based on the uncertainty and a calibration coefficient determination model includes: Determine mean data corresponding to the uncertainties under the multiple preset time lengths, and determine a function based on the mean data and the confidence interval to obtain a data range to be compared corresponding to the uncertainties under the multiple preset time lengths; In the case that the data range to be compared does not satisfy the preset data range, determining an error value corresponding to the data range to be compared and the preset data range; Based on the error value, the historical calibration coefficient corresponding to the target flow meter, the preset adjustment weight coefficient and the calibration coefficient determination model, the target calibration coefficient corresponding to the target flow meter is determined.

7. The method according to claim 6, characterized in that When the to-be-compared data range does not satisfy a preset data range, the method further includes: Based on the data range to be compared and the data to be processed corresponding to the target flow meter, early warning information is generated, and early warning processing is performed based on the early warning information.

8. A flow meter calibration device, characterized in that: include: A data acquisition module, used for acquiring, in the process of collecting the fluid flow of the liquid added to the target tobacco based on the target flow meter, at least one sensor deployed at a preset position of the target flow meter, data to be processed of at least one dimension of the liquid under multiple preset time lengths, wherein the at least one dimension includes at least one of a fluid temperature dimension, a fluid pressure dimension, and a fluid flow rate dimension; An uncertainty evaluation module, for inputting the to-be-processed data of at least one dimension under the multiple preset time lengths into a pre-trained uncertainty evaluation model, and determining the uncertainty of the target flow meter under the current preset time length; The flow meter calibration module is used to determine the target calibration coefficient of the target flow meter based on the uncertainty and calibration coefficient when the uncertainty under the multiple preset time lengths meets the preset conditions, and calibrate the hardware parameters of the target flow meter based on the target calibration coefficient.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so as to enable the at least one processor to perform the flow meter calibration method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the flow meter calibration method according to any one of claims 1 to 7 when executed.

Images