CN113468751B - Recursion Lasso-based flowmeter anomaly online monitoring method and system and storage medium - Google Patents

Abstract

The invention discloses a recursion Lasso-based flowmeter abnormity on-line monitoring method, a system and a storage medium, wherein the method comprises the following steps: establishing an autoregressive model according to historical flow data of the flow meter, and determining a model coefficient of the autoregressive model through an offline Lasso algorithm; updating the autoregressive model by a recursion Lasso algorithm by utilizing the real-time flow of the flowmeter; and determining whether the flowmeter is abnormal according to the predicted flow obtained based on the updated autoregressive model and the actual flow of the flowmeter at the next moment. The recursion Lasso-based flowmeter abnormity on-line monitoring method updates the autoregressive model through the recursion Lasso algorithm, monitors the abnormity of the flowmeter according to the predicted flow and the actual flow obtained based on the updated autoregressive model, has the characteristics of high accuracy, convenient operation, real-time tracking and the like, provides scientific, objective and reliable technical support for on-line monitoring of the humidifying water flow in the tobacco shred manufacturing process, and ensures the metering performance stability of the flowmeter.

Description

Flowmeter Abnormal Online Monitoring Method, System and Storage Medium Based on Recursive Lasso

technical field

The invention relates to the technical field of tobacco processing, in particular to a recursive Lasso-based flowmeter abnormality online monitoring method, system and storage medium.

Background technique

There will be abnormalities in the use of measuring instruments. If they are not checked and repaired in time after the abnormality occurs, the accuracy of the measuring instruments will be affected, which will bring huge hidden dangers to the subsequent industrial production process and increase The cost of administrative expenses for the business.

In order to ensure the accuracy of the measurement data of the measuring instruments, it is necessary to monitor the data measured by the measuring instruments in real time during the working process of the measuring instruments. Once an abnormality occurs, such as data jumps, etc., it is necessary to judge whether the measuring instrument is faulty or the adjustment of the production conditions; if the measuring instrument is faulty, it is necessary to check and repair the abnormality of the measuring instrument.

The online flowmeter is one of the most important sources of basic data acquisition in the traceability chain of the cigarette manufacturing process, and its detection performance needs to be kept within the allowable range during the production process. At present, the production line of cigarette enterprises lacks professionalism in the abnormal monitoring methods of flowmeters. On the one hand, they simply refer to the routine point inspection procedures, while ignoring the characteristics of online monitoring in the production process; on the other hand, they only rely on the maintenance experience of technicians Subjective prediction is somewhat arbitrary, and there is a risk of error caused by human intervention, so it is impossible to ensure whether the flowmeter is actually faulty. It can be seen that due to the lack of scientific basis and data support, the existing flowmeter anomaly monitoring methods fail to achieve a good fault warning function, and cannot detect flowmeter failures in time, so there are large hidden dangers in production.

Therefore, there is an urgent need for a recursive Lasso-based flowmeter abnormality online monitoring method, system and storage medium.

Contents of the invention

The object of the present invention is to provide a recursive Lasso-based flowmeter abnormality online monitoring method, system and storage medium to solve the above-mentioned problems in the prior art, and to be able to monitor the abnormality of the flowmeter in time.

The present invention provides a flowmeter abnormal online monitoring method based on recursive Lasso, which includes:

According to the historical flow data of flow meter, set up autoregressive model, and determine the model coefficient of described autoregressive model by off-line Lasso algorithm;

Using the real-time flow rate of the flowmeter, the autoregressive model is updated through the recursive Lasso algorithm;

Whether the flowmeter is abnormal is determined according to the predicted flow obtained based on the updated autoregressive model and the actual flow of the flowmeter at the next moment.

The above-mentioned recursive Lasso-based flowmeter abnormality online monitoring method, wherein, preferably, the autoregressive model is established according to the historical flow data of the flowmeter, and the model of the autoregressive model is determined by the offline Lasso algorithm coefficients, including:

Collect data samples of flowmeter flow to obtain historical flow data collection;

Construct an autoregressive model according to the historical flow data set;

Obtain the regression coefficient of described autoregressive model based on off-line Lasso algorithm;

The regularization parameters of the autoregressive model are determined by means of cross-validation.

The above-mentioned recursive Lasso-based flowmeter abnormality online monitoring method, wherein, preferably, the optimal condition of the offline Lasso algorithm is determined by the following formula, and the optimal condition of the offline Lasso algorithm is determined according to the Regression coefficients and regularization parameters for an autoregressive model:

Among them, p is the lag coefficient of the autoregressive model, y _n represents the flow value to be predicted at the current moment, y _ni represents the flow value at the historical moment, α represents the regression coefficient of the autoregressive model, μ _n-1 represents the regularity of the autoregressive model parameterization.

The above-mentioned recursive Lasso-based flowmeter abnormal online monitoring method, wherein, preferably, the real-time flow of the flowmeter is used to update the autoregressive model through the recursive Lasso algorithm, specifically including:

Using the model coefficients of the autoregressive model determined by the offline Lasso algorithm as the initial model coefficients of the autoregressive model in the update process;

Whenever a new real-time flow of the flowmeter is read, the model parameters of the autoregressive model are updated according to the new real-time flow of the flowmeter, and the autoregressive model is updated by the following formula:

Among them, Y _n-1 represents the historical flow data collection of flow measured by the flowmeter, Z _n-1 represents the autoregressive item of the autoregressive model, y _n represents the flow value to be predicted at the current moment, and Z _n represents the autoregressive value corresponding to time n. Regression item, t is the parameter of the recursive Lasso algorithm, and its value is 0-1. After the new data is read in, the value of t will change from 0 to 1. When t=1, it means that the new data is completely read in. The parameters α and μ in the regression model are also updated, and the new data at the next moment is ready to be read in. After the new data at the next moment is read in, the value of t changes from 0 to 1 again.

In the online monitoring method for flowmeter abnormalities based on recursive Lasso as described above, preferably, the flowmeter is determined according to the predicted flow rate obtained based on the updated autoregressive model and the actual flow rate of the flowmeter at the next moment. Is it abnormal, including:

According to the updated autoregressive model, the flow rate of the flow meter at the next moment is predicted in real time to obtain the predicted flow rate;

Obtain an estimated curve diagram of the predicted flow according to the obtained predicted flow corresponding to different moments;

According to the deviation between the predicted flow and the actual flow of the flowmeter at the next moment, a residual map is obtained;

Whether the flowmeter is abnormal is determined according to the residual error graph.

According to the recursive Lasso-based flowmeter abnormality online monitoring method as described above, preferably, the determining whether the flowmeter is abnormal according to the residual map specifically includes:

When the ordinate of the residual graph exceeds a preset threshold, it is determined that the flow data of the flowmeter is abnormal;

According to the production records, it is determined whether the abnormality of the flow data of the flowmeter is caused by the abnormality of the flowmeter or the replacement of the shredded stem product batch.

The above-mentioned recursive Lasso-based flowmeter abnormality online monitoring method, wherein, preferably, the recursive Lasso-based flowmeter abnormality online monitoring method is used to monitor whether the humidification water flow in the shredded tobacco process is abnormal,

The recursive Lasso-based flowmeter abnormal online monitoring method also includes:

Based on the preset flow threshold of the cut stem product batch and the predicted flow, the specification of the cut stem product at the current moment is determined.

The present invention also provides a flowmeter abnormal online monitoring system based on recursive Lasso using the above method, including:

The autoregressive model building module is used to establish an autoregressive model according to the historical flow data of the flowmeter, and determine the model coefficients of the autoregressive model by an off-line Lasso algorithm;

The autoregressive model update module is used to update the autoregressive model through the recursive Lasso algorithm by using the real-time flow of the flowmeter;

The flow monitoring module is used to determine whether the flow meter is abnormal according to the predicted flow obtained based on the updated autoregressive model and the actual flow of the flow meter at the next moment.

The present invention also provides a computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when it is run on a computer, the computer executes the above-mentioned recursive Lasso-based flowmeter abnormality online monitoring method.

The present invention also provides a computer program product, which is characterized in that, when the computer program product is run on a terminal device, the terminal device is made to execute the above-mentioned recursive Lasso-based flowmeter abnormality online monitoring method.

The present invention provides a flowmeter abnormality online monitoring method based on recursive Lasso, which updates the autoregressive model through the recursive Lasso algorithm, and monitors the abnormality of the flowmeter according to the predicted flow and actual flow obtained based on the updated autoregressive model It has the characteristics of high accuracy, convenient operation and real-time tracking. It provides scientific, objective and reliable technical support for the on-line monitoring of humidification water flow in the tobacco shred process, thereby ensuring the stability of the measurement performance of the flowmeter, and The invention is also applicable to detection scenarios of other measuring instruments, and has wide application prospects.

Description of drawings

In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be further described below in conjunction with accompanying drawing, wherein:

Fig. 1 is the flowchart of the embodiment of the flow meter abnormal online monitoring method based on recursive Lasso provided by the present invention;

Fig. 2 is the estimation graph of the predicted flow provided by the embodiment of the present invention;

FIG. 3 is a residual diagram provided by an embodiment of the present invention;

Fig. 4 is a structural block diagram of an embodiment of an online monitoring system for flowmeter abnormality based on recursive Lasso provided by the present invention.

detailed description

Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. The description of the exemplary embodiments is illustrative only, and in no way restricts the disclosure, its application or uses. The present disclosure can be implemented in many different forms and is not limited to the embodiments described here. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art. It should be noted that relative arrangements of parts and steps, compositions of materials, numerical expressions and numerical values set forth in these embodiments should be interpreted as illustrative only and not as limiting, unless specifically stated otherwise.

"First", "second" and similar words used in the present disclosure do not indicate any order, quantity or importance, but are only used to distinguish different parts. Words like "comprising" or "comprising" mean that the elements preceding the word cover the elements listed after the word, and do not exclude the possibility of also covering other elements. "Up", "Down" and so on are only used to indicate the relative positional relationship. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

In the present disclosure, when it is described that a specific component is located between a first component and a second component, there may or may not be an intervening component between the specific component and the first component or the second component. When it is described that a specific component is connected to other components, the specific component may be directly connected to the other component without an intermediate component, or may not be directly connected to the other component but has an intermediate component.

All terms (including technical terms or scientific terms) used in the present disclosure have the same meaning as understood by one of ordinary skill in the art to which the present disclosure belongs, unless otherwise specifically defined. It should also be understood that terms defined in, for example, general-purpose dictionaries should be interpreted as having meanings consistent with their meanings in the context of the relevant technology, and should not be interpreted in idealized or extremely formalized meanings, unless explicitly stated herein Defined like this.

Techniques, methods and devices known to those of ordinary skill in the relevant art may not be discussed in detail, but where appropriate, techniques, methods and devices should be considered part of the description.

Aiming at the flowmeter, which is a sensor device that is frequently used in the cigarette production process, based on the actual application scene and objective scientific ideas, the present invention proposes an online monitoring method for flowmeter abnormality based on recursive Lasso to monitor the flowmeter online, and timely Find faults and reduce losses.

The online monitoring method for flowmeter abnormality based on recursive Lasso of the present invention is used to monitor whether the flow rate of humidifying water in the tobacco shred process is abnormal. As shown in Figure 1, the online monitoring method for abnormal flowmeter based on recursive Lasso provided by this embodiment In the actual implementation process, the specific steps are as follows:

Step S1. Establish an autoregressive model according to the historical flow data of the flowmeter, and determine the model coefficients of the autoregressive model through an offline Lasso algorithm.

In one embodiment of the online monitoring method for flowmeter abnormalities based on recursive Lasso of the present invention, the step S1 may specifically include:

Step S11, collecting data samples of the flow rate of the flow meter to obtain a historical flow data set.

Specifically, the data collected by the humidification water flow meter in the cut stem feeding unit is collected according to a preset time length as a data source for online monitoring, so as to obtain a set of historical flow data.

Step S12, constructing an autoregressive model according to the historical flow data set.

Step S13, calculating the regression coefficient of the autoregressive model based on the offline Lasso algorithm.

Step S14, using a cross-validation method to determine the regularization parameters of the autoregressive model.

Specifically, the optimal condition of the offline Lasso algorithm is determined by the following formula, and the regression coefficient and the regularization parameter of the autoregressive model are determined according to the optimal condition of the offline Lasso algorithm:

Among them, p is the lag coefficient of the autoregressive model, y _n represents the flow value to be predicted at the current moment, y _ni represents the flow value at the historical moment, α represents the regression coefficient of the autoregressive model, μ _n-1 represents the regularity of the autoregressive model parameterization.

其中ε_i表示流量计测量的噪声数据。那么Lasso的最优解就变为：The l ₁ norm is introduced into the autoregressive model, assuming that the flow data measured by the flowmeter is y∈R ^n-1 , and the regression coefficient of the model is p,

Where _εi represents the noise data measured by the flowmeter. Then the optimal solution of Lasso becomes:

Among them, μ _n-1 is the regularization parameter. The form of the solution of formula (1) is sparse, and only a small number of elements are non-zero. The purpose of the Lasso algorithm is to combine some unimportant Variables are set to zero to highlight some non-zero important variables, which are useful variables for prediction, so it can be determined which historical moments of flow are helpful to predict the current moment of flow based on these non-zero variables.

The index of the non-zero elements in the regression coefficient vector α defines an "effective set". In order to make the notation easier, the "effective set" is placed in front, for example, α ^T =(α ₁ ^T ,0 ^T ), v ^T = (v ₁ ^T ,v ₂ ^T ). Among them, for all i, the higher-order function v _1i =sgn(α _1i ) is satisfied; for all j, -1≤v _2j ≤1, where i and j are represented separately for different sets index, which avoids confusion. At the same time, according to the "effective set", Y is divided into Y=(Y ₁ Y ₂ ), where Y represents the set of y _n-1 to y _np in formula (1), and the elements in each set correspond to a regression The coefficient α, _Y1 means the element in Y corresponding to which the regression coefficient α is not zero, and _Y2 refers to the element in Y corresponding to which the regression coefficient α is zero. If the obtained solution is unique, then Y ₁ ^T Y ₁ is reversible, then the optimal condition is

It should be noted that α can be computed in closed form given the “active set” and the sign of the coefficients in the eigenvector α.

Step S2, using the real-time flow rate of the flowmeter to update the autoregressive model through a recursive Lasso (Least absolute shrinkage and selection operator) algorithm.

Fine-tuning the model parameters of the autoregressive model according to the new data read in can improve the predictive effect of the autoregressive model. In one embodiment of the online monitoring method for flowmeter abnormality based on recursive Lasso of the present invention, the step S2 may specifically include:

Step S21, using the model coefficients of the autoregressive model determined by the offline Lasso algorithm as the initial model coefficients of the autoregressive model in the updating process.

Step S22, each time a new real-time flow rate of the flowmeter is read, the model parameters of the autoregressive model are updated according to the new real-time flow rate of the flowmeter, and the autoregressive model is updated by the following formula:

Among them, Y _n-1 represents the historical flow data collection of flow measured by the flowmeter, Z _n-1 represents the autoregressive item of the autoregressive model, y _n represents the flow value to be predicted at the current moment, and Z _n represents the autoregressive value corresponding to time n. Regression item, t is the parameter of the recursive Lasso algorithm, and its value is 0-1. After the new data is read in, the value of t will change from 0 to 1. When t=1, it means that the new data is completely read in. The parameters α and μ in the regression model are also updated, and the new data at the next moment is ready to be read in. After the new data at the next moment is read in, the value of t changes from 0 to 1 again.

Specifically, assuming that the solution α ^(n-1) of formula (1) at time n-1 has been calculated, new observation data y _n is obtained, then y _n and historical data can be used to predict time n+1 data y _n+1 , then the goal is to calculate the value of α ⁽ⁿ⁾ . This leads to the solution of recursive Lasso. Denote z _n ＝(y _n ,y _n-1 ,...,y _np ), Z _n ＝(z ₁ ,z ₂ ,...,z _n ) ^T , Y _n-1 ＝(y ₁ ,y ₂ ,...y _n-1 ) ^T , where z _n represents the historical time flow required to predict the current flow, Z _n represents the collection of all previous historical time flow, Y _n-1 represents all historical time flow, Then the optimized objective function becomes:

The entire update path becomes α ^(n-1) = α(0,μ _n-1 ) to α ⁽ⁿ⁾ =α(1,μ _n ). According to this update path, the update method can be divided into two steps:

Step 1: when t=0, update the regularization parameters from μ _n−1 to μ _n , which is equivalent to calculating the regularization path between the two using the minimum angle regression method.

The second step: when μ=μ _n , calculate t from 0 to 1.

The solution process of the above second step is as follows: firstly, it needs to be proved that α(t, μ) is a piecewise smooth function for t. In order to make the sign more convenient, let α(t)=α(t,μ), if you already know the "effective set" and the sign of the coefficient in α, you can calculate the solution of Lasso. In the first step, the "effective set" and the signs of the coefficients in α have been calculated, and when t∈[0,t ^* ) (where t ^* means at t∈[0,t ^* ) when α(t ) is a smooth curve), the "effective set" and the signs of the coefficients in α remain unchanged, and Lasso's solution α(t) is smooth. A point where the "effective set" changes is called a turning point. Next, we will analyze how to calculate this point:

When t=t ^* , update the "effective set" and the sign of the coefficient in α, and keep it unchanged until the next turning point arrives, and iterate this process until t=1, so that the required solution α(t ).

Thus, the online update algorithm of Lasso with added observations is obtained:

STEP1: Calculate α ^(n-1) = α(0, μ _n-1 ) to α ⁽ⁿ⁾ = α(1, μ _n );

是

的子矩阵，其中的列是“有效集”，初始化

初始化转折点t'＝0；STEP2: Initialize the non-zero coefficients of α(0,μ _n ) to the "effective set", let v=sgn(α(0,μ _n )), let v ₁ and z _n,1 be v and z _n according to the "effective set"Set" sub-vectors,

yes

A submatrix of , the columns of which are the "active set", initialized

Initialize turning point t'=0;

STEP3: Calculate the next turning point t'. If the turning point is smaller than the previous turning point or the turning point is greater than 1, jump to STEP5,

The first case is: first the i-th element in α ₁ (t') becomes 0; then remove i from the "effective set"; then set v _i to 0;

The second case is: first, the absolute value of the jth element of ω ₂ (t') reaches 1; then j is added to the "effective set"; then, if the element reaches 1 (or -1), then v _j is set to 1 (or -1).

和z_n,1，更新

STEP4: Update v ₁ according to the updated "effective set",

and z _n,1 , update

的有效集给出。STEP5: Calculate the final result when t=1, where the value of α ⁽ⁿ⁾ is determined by

The effective set of is given.

When there are too many observations in the data set, the calculation time of STEP1 will be longer, and in the subsequent calculation process, the observation point data that is farther away from the current time has little effect on predicting the value at the current time, so consider after a period of time Eliminate the initial historical data, so that the calculation time of the model is stable within a certain range.

Assuming that after the solution α ⁽ⁿ⁾ at time n has been calculated, the original data needs to be eliminated before the new observation data is read in, thus leading to the solution of recursive Lasso elimination of historical data. Write z ₁ =(y ₀ ,y _-1 ,...,y _-p ), Z=(z ₂ ,...,z _n ) ^T , Y=(y ₂ ,y ₃ ,...,y _n ) ^T , then the optimized objective function becomes:

The entire update path becomes α ⁽ⁿ⁾ = α(1,μ _n ) to α ^(n') = α(0,μ _n' ), according to this update path, the update method can be divided into two steps: STEP1: When When t=1, the regularization parameters are updated from μ _n to μ _n ', which is equivalent to calculating the regularization path between the two using the minimum angle regression method.

STEP2: When μ=μ _n' , calculate t from 1 to 0.

At this time, the calculation steps of STEP2 are the same as the steps of adding observation data.

Step S3. Determine whether the flowmeter is abnormal according to the predicted flow obtained based on the updated autoregressive model and the actual flow of the flowmeter at the next moment.

In one embodiment of the online monitoring method for flowmeter abnormality based on recursive Lasso of the present invention, the step S3 may specifically include:

Step S31 , according to the updated autoregressive model, predict the flow rate of the flow meter at the next moment in real time, and obtain the predicted flow rate.

Step S32 , according to the obtained predicted flow corresponding to different time points, an estimated graph of the predicted flow is obtained.

The dotted line in Fig. 2 is the estimated curve diagram of the predicted flow.

Step S33 , according to the deviation between the predicted flow rate and the actual flow rate of the flowmeter at the next moment, a residual map is obtained.

The deviation here is the absolute value of the difference between the predicted value and the actual flow. Among them, the residual graph is used to characterize whether the flow is abnormal in a predetermined usage environment. In some embodiments of the present invention, the obtained residual diagram is shown in Figure 3, and the straight lines in the diagram are 50kg/h, 25kg/h, and 10kg/h from top to bottom.

Step S34. Determine whether the flowmeter is abnormal according to the residual map.

Use the deviation value to estimate the change in the online flow working process, and when the deviation value is too large, it is determined that the flow has abnormal operation. In one embodiment of the online monitoring method for flowmeter abnormality based on recursive Lasso of the present invention, the step S34 may specifically include:

Step S341, when the ordinate of the residual graph exceeds a preset threshold, it is determined that the flow data of the flow meter is abnormal.

In Figure 3, when the flowmeter fails at the 7000th moment, the predicted value deviates greatly from the actual value, and the residual error (the vertical axis of the residual error graph) is greater than 50 kg/h, which is considered as abnormal flow data (Fig. The abnormal point in is around 100kg/h).

Step S342 , according to the production records, it is determined whether the abnormality of the flow data of the flowmeter is caused by the abnormality of the flowmeter or the replacement of the cut stem product batch.

If the cut stem product batch is changed, it is determined that the reason for the abnormal flow data of the flowmeter is that the production has been adjusted; if the cut stem product batch has not been changed, it is determined that the cause of the abnormal flow data of the flowmeter is the If an abnormality occurs, the flowmeter needs to be maintained in time. In the present invention, the status performance of the flow meter is tracked in real time by using the residual graph, and when the flow is abnormal, it can be monitored in time and an alarm can be issued.

Further, in some embodiments of the present invention, the recursive Lasso-based flowmeter abnormality online monitoring method also includes:

Step S4, based on the preset flow threshold of the cut-stem product batch and the predicted flow, determine the specification of the cut-stem product at the current moment.

Based on the preset flow threshold of the cut-stem product batch and the estimated curve drawn based on the predicted flow at multiple times, the product batch entering the cut-stem feeding unit at the current moment is determined. Since each batch of cut stem products is different, the flow rate of humidification water will also be different. Therefore, by setting the range of flow rate of humidification water, the specification of cut stem products can be judged according to the flow rate estimated by the autoregressive model.

The shredded stems in the data samples are divided into two types, one is ordinary stems, and the other is special golden leaf stems. When the material supply is stable, the humidifying water flow rate of common stalks is within the range of 110-130kg/h, and the humidifying water flow rate of special golden leaf stalks is within the range of 140-170kg/h. Therefore, the product specifications of the shredded stems can be judged according to the predicted flow value. In Figure 2, the dotted line is the predicted value, and the solid curve is the actual value; the solid line is the dividing line of 140kg/h. If the material supply is stable and the predicted flow rate is less than 140kg/h, it is judged as a common stem, and the measured flow rate is greater than 140kg/h. It is judged as a special stalk for golden leaves.

The recursive Lasso-based flowmeter abnormality online monitoring method provided by the embodiment of the present invention updates the autoregressive model through the recursive Lasso algorithm, and monitors the flowmeter according to the predicted flow rate and actual flow rate obtained based on the updated autoregressive model. Abnormal conditions, with the characteristics of high accuracy, convenient operation and real-time tracking, provide scientific, objective and reliable technical support for the on-line monitoring of humidification water flow in the tobacco shred process, thereby ensuring the stability of the measurement performance of the flowmeter. Moreover, the present invention is also applicable to detection scenarios of other measuring instruments, and has wide application prospects.

Correspondingly, as shown in Fig. 4, the present invention also provides a flowmeter abnormality online monitoring system based on recursive Lasso, including:

The autoregressive model building module 1 is used to set up an autoregressive model according to the historical flow data of the flowmeter, and determine the model coefficients of the autoregressive model by an off-line Lasso algorithm;

The autoregressive model updating module 2 is used to update the autoregressive model through the recursive Lasso algorithm by utilizing the real-time flow rate of the flowmeter;

The flow monitoring module 3 is configured to determine whether the flow meter is abnormal according to the predicted flow obtained based on the updated autoregressive model and the actual flow of the flow meter at the next moment.

The recursive Lasso-based flowmeter abnormality online monitoring system provided by the embodiment of the present invention uses the autoregressive model update module to update the autoregressive model through the recursive Lasso algorithm, and the flow monitoring module is based on the prediction obtained based on the updated autoregressive model It has the characteristics of high accuracy, convenient operation and real-time tracking. It provides scientific, objective and reliable technical support for the on-line monitoring of humidification water flow in the tobacco shredded process, and can further The metering performance stability of the flowmeter is guaranteed, and the present invention is also applicable to detection scenarios of other metering instruments, and has broad application prospects.

It should be understood that the division of the various components of the recursive Lasso-based flowmeter abnormality online monitoring system shown in Figure 4 is only a division of logical functions. In actual implementation, it can be fully or partially integrated into a physical entity, or physically separate. And these components can all be implemented in the form of software called by the processing element; they can also be implemented in the form of hardware; some components can also be implemented in the form of software called by the processing element, and some components can be implemented in the form of hardware. For example, a certain above-mentioned module may be a separately established processing element, or may be integrated into a certain chip of the electronic device for implementation. The implementation of other components is similar. In addition, all or part of these components can be integrated together, or implemented independently. In the process of implementation, each step of the above method or each of the above components can be completed by an integrated logic circuit of hardware in the processor element or an instruction in the form of software.

The present invention also provides a computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when it is run on a computer, the computer executes the above-mentioned recursive Lasso-based flowmeter abnormality online monitoring method.

The present invention also provides a computer program product, which is characterized in that, when the computer program product is run on a terminal device, the terminal device is made to execute the above-mentioned recursive Lasso-based flowmeter abnormality online monitoring method.

In the above embodiments, all or part may be implemented by software, hardware, firmware or any combination thereof. When implemented using software, it may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on the computer, the processes or functions according to the embodiments of the present application will be generated in whole or in part. The computer can be a general purpose computer, a special purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from a website, computer, server or data center Transmission to another website site, computer, server or data center via wired (eg coaxial cable, optical fiber, Digital Subscriber Line (DSL)) or wireless (eg infrared, wireless, microwave, etc.). The computer-readable storage medium may be any available medium that can be accessed by a computer, or a data storage device such as a server or a data center integrated with one or more available media. The available medium may be a magnetic medium (for example: floppy disk, hard disk, magnetic tape), an optical medium (for example: Digital Versatile Disc (Digital Versatile Disc, DVD)), or a semiconductor medium (for example: Solid State Disk (Solid State Disk, SSD) )Wait.

So far, the embodiments of the present disclosure have been described in detail. Certain details known in the art have not been described in order to avoid obscuring the concept of the present disclosure. Based on the above description, those skilled in the art can fully understand how to implement the technical solutions disclosed herein.

Although some specific embodiments of the present disclosure have been described in detail through examples, those skilled in the art should understand that the above examples are for illustration only, rather than limiting the scope of the present disclosure. Those skilled in the art should understand that the above embodiments can be modified or some technical features can be equivalently replaced without departing from the scope and spirit of the present disclosure. The scope of the present disclosure is defined by the appended claims.

Claims (5)

1. A recursion Lasso-based flowmeter abnormity on-line monitoring method is characterized by being used for monitoring whether humidifying water flow in a tobacco shredding process is abnormal or not, and comprising the following steps of:

establishing an autoregressive model according to historical flow data of the flow meter, and determining a model coefficient of the autoregressive model through an offline Lasso algorithm;

updating the autoregressive model by a recursion Lasso algorithm by utilizing the real-time flow of the flowmeter;

determining whether the flowmeter is abnormal according to the predicted flow obtained by the updated autoregressive model and the actual flow of the flowmeter at the next moment,

the method for determining the flow rate of the flowmeter comprises the following steps of establishing an autoregressive model according to historical flow rate data of the flowmeter, and determining a model coefficient of the autoregressive model through an offline Lasso algorithm, wherein the method specifically comprises the following steps:

collecting a data sample of the flow of the flowmeter to obtain a historical flow data set;

constructing an autoregressive model according to the historical flow data set;

solving a regression coefficient of the autoregressive model based on an offline Lasso algorithm;

determining the regularization parameters of the autoregressive model by using a cross-validation method,

determining the optimal condition of the offline Lasso algorithm through the following formula, and determining the regression coefficient and the regularization parameter of the autoregressive model according to the optimal condition of the offline Lasso algorithm:

where p is the hysteresis coefficient of the autoregressive model, y _n Representing the flow value to be predicted at the current moment, y _n-i Represents the flow rate value at the historical time, alpha represents the regression coefficient of the autoregressive model, mu _n-1 A regularization parameter that represents an autoregressive model,

the updating of the autoregressive model by using the real-time flow of the flow meter and the recursive Lasso algorithm specifically comprises:

taking the model coefficient of the autoregressive model determined by an offline Lasso algorithm as an initial model coefficient of the autoregressive model in an updating process;

updating the model parameters of the autoregressive model according to the new real-time flow of the flowmeter every time a new real-time flow of the flowmeter is read in, and updating the autoregressive model by the following formula:

wherein, Y _n-1 Historical flow representing flow measured by a flow meterVolume data set, Z _n-1 Autoregressive term, y, representing an autoregressive model _n Representing the flow value to be predicted at the current moment, Z _n Showing an autoregressive term corresponding to the moment n, wherein t is a parameter of a recursion Lasso algorithm, the value of the parameter is 0-1, the value of t is changed from 0 to 1 after new data is read, when t =1, the new data is completely read, parameters alpha and mu in the autoregressive model are updated, new data at the next moment is prepared to be read, after the new data at the next moment is read, the value of t is changed from 0 to 1 again,

determining whether the flowmeter is abnormal according to the predicted flow obtained by the updated autoregressive model and the actual flow of the flowmeter at the next moment, specifically comprising:

predicting the flow of the flowmeter at the next moment in real time according to the updated autoregressive model to obtain predicted flow;

obtaining an estimation curve graph of the predicted flow according to the obtained predicted flow corresponding to different moments;

obtaining a residual error map according to the deviation between the predicted flow and the actual flow of the flowmeter at the next moment;

and determining whether the flow meter is abnormal according to the residual error map.

2. The method for online monitoring of flow meter anomaly based on recursive Lasso according to claim 1, wherein the determining whether the flow meter is anomalous according to the residual error map specifically comprises:

when the ordinate of the residual error map exceeds a preset threshold value, determining that the flow data of the flowmeter is abnormal;

and determining whether the reason causing the flow data abnormity of the flowmeter is the abnormity of the flowmeter or the batch replacement of the cut stem product according to the production record.

3. The recursion Lasso-based flowmeter anomaly online monitoring method as claimed in claim 1, wherein the recursion Lasso-based flowmeter anomaly online monitoring method is used for monitoring whether the flow rate of the humidifying water in the tobacco shredding process is abnormal,

the flow meter abnormity online monitoring method based on recursion Lasso further comprises the following steps:

and determining the cut stem product specification at the current moment based on a preset flow threshold of the cut stem product batch and the predicted flow.

4. A recursive Lasso based flowmeter anomaly on-line monitoring system employing the method of any of claims 1-3, comprising:

the system comprises an autoregressive model establishing module, a flow meter and a flow control module, wherein the autoregressive model establishing module is used for establishing an autoregressive model according to historical flow data of the flow meter and determining a model coefficient of the autoregressive model through an offline Lasso algorithm;

the autoregressive model updating module is used for updating the autoregressive model through a recursion Lasso algorithm by utilizing the real-time flow of the flowmeter;

and the flow monitoring module is used for determining whether the flowmeter is abnormal or not according to the predicted flow obtained by the updated autoregressive model and the actual flow of the flowmeter at the next moment.

5. A computer-readable storage medium, having stored thereon a computer program which, when run on a computer, causes the computer to execute the recursive Lasso based flowmeter anomaly online monitoring method of any of claims 1 to 3.

Images