A flow interpolation and integration method and device based on integral diffusion factor

Abstract

The present invention discloses a flow interpolation and accumulation method based on integral diffusion factor. A weighted diffusion factor is introduced into the method of the present invention. The sampling volume value can be obtained by dynamically fitting a spline curve according to the sampling flow and accumulating the energy of the sampling flow detection signal corresponding to the fitting spline curve in combination with the weighted diffusion factor. The accumulation device of the present invention includes a sampler, a measuring cylinder, a flow sensor, an industrial control machine, a terminal module, a cloud server, a query terminal, and a mobile terminal. The flow sensor and the terminal module are all connected to the industrial control machine, and the terminal module, the cloud server, the query terminal, and the mobile terminal are all connected to the Internet of Things. The sampler conveys the sampled slurry to the measuring cylinder, and the sampling result of the sampler is manually conveyed to the industrial control machine. The present invention can realize the dynamic statistics of the sampling amount of liquid slurry in industrial production, and thus provide basic information support for production management and process decision-making.

Description

Technical Field

[0001] The invention belongs to the technical field of flow detection and calculation, and in particular relates to a flow interpolation and integration method and device based on an integral diffusion factor. Background Art

[0002] The number of shift sampling and the amount of shift sampling of liquid slurry in the industrial production process directly determine the sampling uniformity of the production process inspection and testing system and the representativeness of the test results. At the same time, it can reflect the advantages and disadvantages of the current production process technical indicators from multiple spatial regions and time dimensions. It is one of the necessary criteria for determining whether the production conditions are in a steady state, and it is also the basis for dynamic adjustment of production parameters.

[0003] The sampling of liquid slurry in the industrial production process is to make the sampling process as uniform as possible and reduce the error caused by sampling unevenness under the premise of ensuring the sampling volume of the sampling shift. Affected by factors such as the working environment and installation conditions, in order to improve environmental adaptability and reduce the failure rate of sampling equipment, actual slurry sampling mostly uses mechanical devices, and the operation stability and reliability of the sampling device are guaranteed by manually setting regular inspections. Due to the time-varying characteristics of working conditions such as adjustment of production process operating parameters and abnormal fluctuations in process conditions, it is difficult for the sampling volume of a shift to reach the expected indicators. Summary of the invention

[0004] In view of the problems in the prior art, in order to dynamically monitor the single sampling volume and cumulative sampling volume in the shift sampling of liquid slurry in the industrial production process, so as to achieve the purpose of dynamically adjusting the sampling cycle for the maintenance personnel in the sampling site to provide a reference for decision-making operations, the sample length, sample domain and sample scale of the sampling point space are autonomously extracted and identified according to the dynamic characteristics of the slurry sampling flow detection value. The present invention provides a flow interpolation integration method and device based on integral diffusion factor, which can realize dynamic statistics of the liquid slurry sampling volume in industrial production, and thus provide basic information support for production management and process decision-making.

[0005] The present invention adopts the following technical solutions:

[0006] A flow interpolation and accumulation method based on an integral diffusion factor, characterized in that the method comprises the following steps:

[0007] (1) detecting a generalized sampling domain according to the flow rate output by the flow sensor, and obtaining a data queue of sampling points in the sampling domain;

[0008] (2) fitting the sampling detection flow integration density function according to the sampling point data and the sampling point data queue;

[0009] (3) Integrate the sampling detection flow rate integrated density function in the sampling domain to obtain the sampling flow rate detection signal energy accumulation function;

[0010] (4) Calculate the weighted diffusion factor of the energy accumulation function of the sampled flow detection signal;

[0011] (5) Calculate the real-time sampling volume.

[0012] According to the above-mentioned flow interpolation and integration method based on integral diffusion factor, it is characterized in that the first and last data points of the sampled and detected flow integration density function in step (2) are second-order differentiable, and the first-order derivative of the first and last data points of the sampled and detected flow integration density function is 0.

[0013] According to the flow interpolation and integration method based on the integral diffusion factor, it is characterized in that in step (2), according to {x i ,f(x i )}, (i=0,…,n), cubic spline interpolation is used to fit the sampling detection flow integration density function S(x):

[0014]

[0015] in:

[0016] x i , (i=0,…,n) is the i-th time value of the sampling point data queue in the sampling domain;

[0017] f(x i )=y i , (i=0,…,n) is the i-th sampled flow signal value in the sampling domain;

[0018] h i , (i=0,…,n-1) is the time step, which is the difference between the current moment and the previous moment, h i =x i+1 -x i ;

[0019] M i , (i=0,…,n) is the second-order derivative of the sampling detection flow integrated density function at the i-th moment, and M i Satisfy the formula:

[0020]

[0021] μ i ,λ i , (i=1,…,n-1) is the current time step h i and the adjacent time step h i-1 ,h i+1 The linear equation expression that determines the constant coefficient, d i , (i=1,…,n-1) is the number of the previous time step h i-1 , current time step h iand the i-th sampling flow signal value y in the sampling domain i and adjacent sampling flow signal value y i-1 ,y i+1 Determine the constant coefficients of the linear equation expression.

[0022] According to the above-mentioned flow interpolation and accumulation method based on integral diffusion factor, it is characterized in that the expression of the energy accumulation function P(.) of the single sampling flow detection signal in step (3) is: Where: n p is the sample length in the sampling domain; c0 is the reference offset coefficient of the sampling flow detection signal, c0=4.

[0023] According to the above-mentioned flow interpolation and accumulation method based on integral diffusion factor, it is characterized in that the calculation formula of the weighted diffusion factor k of the energy accumulation function of the sampled flow detection signal in step (4) is:

[0024]

[0025] in:

[0026] n k is the forgetting sequence length of the diffusion factor;

[0027] c(i) is the forgetting coefficient of the i-th weighted diffusion factor, i=0,…,n k ;n k ≥0;

[0028] P(i) is the energy accumulation function value of the i-th sampled flow detection signal;

[0029] v(i) is the actual measured value of the i-th sampling flow corresponding to P(i).

[0030] According to the above-mentioned flow interpolation and accumulation method based on integral diffusion factor, it is characterized in that the calculation formula of the volume v(t) of the real-time sampling flow in step (5) is: v(t)=kP(t), where: P(t) is the energy accumulation value of the current sampling flow detection signal, and t is the current time.

[0031] A flow interpolation integration device based on the above-mentioned integral diffusion factor-based flow interpolation integration method is characterized in that the device comprises a sampling machine (1), a measuring cylinder (2), a flow sensor (4), an industrial control machine (5), a terminal module (6), a cloud server (8), a query terminal (9), and a mobile terminal (10); the flow sensor (4) and the terminal module (6) are all connected to the industrial control machine (5); the terminal module (6), the cloud server (8), the query terminal (9), and the mobile terminal (10) are all connected to the Internet of Things (7); the sampling machine (1) conveys the sampled slurry to the measuring cylinder (2) by gravity, and the sampling result of the sampling machine (1) is conveyed to the industrial control machine (5).

[0032] According to the above-mentioned flow interpolation integration method based on integral diffusion factor, the integration device is characterized in that the industrial control machine (5) includes a sampling signal density fitting model module, a sampling signal energy integration module, a sampling quantity output module, a parameter library, and a protocol analysis packaging component; the terminal module (6) is an NB-IoT terminal module.

[0033] Compared with the prior art, the beneficial technical effects of the present invention are as follows: (1) The present invention is applicable to the field of flow mass / volume accumulation balance of flow detection instruments in production processes such as mining, mineral processing, chemical industry and inspection and testing, and is mainly used to solve the problems of correction and accurate calculation of the accumulated value of liquid slurry sampling flow. (2) The present invention provides a cubic interpolation fitting method for the sampling point energy accumulation function, which automatically fits the sampling flow signal density function in the sampling domain from the remote terminal flow detection signal queue, combined with the corresponding sampling time sequence; the sampling flow density function forms the sampling point energy accumulation function by integrating in the sampling domain space. (3) The present invention provides a weighted diffusion factor accumulation method composed of a forgetting coefficient sequence, which can effectively eliminate or avoid the accumulation errors caused by process parameter fluctuations, flow meter system drift, and external disturbances. (4) The present invention can realize the accurate accumulation of the sampling flow of liquid slurry under the micro-flow state. The algorithm of the present invention is simple, practical and easy to implement. It makes it possible to accumulate statistics of sampling volume shifts, and provides basic data support for inspection and testing personnel to adjust the parameters of the sampling device in real time. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a schematic diagram of the device of the present invention and its implementation process;

[0035] Figure 2 is the input / output structure of the method of the present invention;

[0036] Figure 3 The figure is a calculation flow chart of the method of the present invention. DETAILED DESCRIPTION

[0037] The integration method of the flow interpolation integration device based on the integral diffusion factor of the present invention firstly adopts cubic spline interpolation to fit the sampling detection flow integration density function in the sampling point sample domain according to the instantaneous flow detection signal value output by the sampling point. The single sampling integral integration value, that is, the single sampling flow detection signal energy accumulation function, can be obtained by performing segmented integration on the sample scale in the sampling point sample domain by the sampling detection flow integration density function. The diffusion factor of the sampling flow integration can be obtained according to the ratio of the single sampling flow detection signal energy accumulation function to the corresponding measured sampling volume measured by the measuring cylinder. Since the sampling flow has independence and irrelevance in the spatial scale, in order to reduce or eliminate the perturbation error of the flow integration volume measurement deviation, the sampling device measurement deviation and the unknown factors on the site as much as possible, the weighted diffusion factor is introduced in the method of the present invention. According to the dynamic fitting spline curve of the sampling flow and the energy accumulation of the sampling flow detection signal corresponding to the fitting spline curve, the sampling volume value of the sub-sampling can be obtained in combination with the weighted diffusion factor. In the present invention, the sample length is time-varying and is obtained by the recursive dynamic iteration of the maximum and minimum method. The sample domain can be determined according to the sample length and the first element, and the sample scale is a one-to-one correspondence between the sample domain element tensor. The method of the present invention is as follows:

[0038] A flow interpolation and accumulation method based on an integral diffusion factor comprises the following steps:

[0039] (1) According to the generalized sampling domain of flow detection output by the flow sensor, the flow sensor is a non-contact flow sensor. The sampling point data queue f(x) in the sampling domain can be obtained through the sampling autonomous identification rule. The first and last data points of the sampling detection flow integration density function are second-order differentiable, and the first and last data points of the sampling detection flow integration density function are 0. Therefore, it can be known that S'(x0) = f'(x0), S'(x n )=f'(x n )=0, f0'=S'(x0) is the first-order derivative of the sampling detection flow integrated density function expression at x0; f n '=S'(x n ) is the sampling detection flow integrated density function expression in x n The first derivative at .

[0040] (2) According to {x i ,f(x i )}, (i=0,…,n), cubic spline interpolation is used to fit the sampling detection flow integration density function S(x):

[0041] In formula (2):

[0042] y i =f(x i ), (i=0,…,n), formula (1)

[0043] x i , (i=0,…,n) is the i-th time value of the sampling point data queue in the sampling domain;

[0044] f(x i )=y i , (i=0,…,n) is the i-th sampled flow signal value in the sampling domain;

[0045] h i , (i=0,…,n) is the time step, which is the difference between the current moment and the previous moment, h i =x i+1 -x i ;

[0046] By taking the derivative of formula (2), we can get:

[0047]

[0048] S'(x) is the first-order derivative of the sampling detection flow integrated density function expression at the xth point;

[0049] S'(x i ), (i=0,…,n) is the first-order derivative of the sampling detection flow integrated density function expression corresponding to the i-th moment;

[0050] By the node continuity constraint S'(x i -0)=S'(x i +0) we can see that:

[0051]

[0052] S'(x i -0)=S' - (x i ),S'(x i +0)=S' + (x i ) are the sampling detection flow integration density function segment nodes x i The first-order derivatives before and after ;

[0053] By rearranging formula (4), we can get:

[0054]

[0055] Multiply both ends of equation (5) by We can get:

[0056]

[0057] make The expression (6) can be transformed into:

[0058] μ i M i-1 +2M i +λ i M i+1 =d i ,i=1,…,n-1, formula (8)

[0059] Expanding for i=1,…,n, we can get:

[0060]

[0061] From formula (3), we can see that When i=0, we can get:

[0062]

[0063] When i=n, we can get:

[0064]

[0065] Formula (11) multiplied by Formula (12) multiplied by Transformation yields:

[0066]

[0067] Combining equations (9), (13), and (14), let λ0 = 1, μ n =1, we get The system of equations:

[0068]

[0069] Where: M i , (i=0,…,n) is the second-order derivative of the sampling detection flow integrated density function expression at the i-th moment; μ i ,λ i , (i=1,…,n-1) is the current time step h i and the adjacent time step h i-1 ,h i+1 Determine the constant coefficient of the linear equation expression; d i , (i=1,…,n-1) is the number of the previous time step h i-1 , current time step h i and the i-th sampling flow signal value y in the sampling domain i and adjacent sampling flow signal value y i-1 ,y i+1 Determine the constant coefficients of the linear equation expression.

[0070] In formula (14), μi ,λ i ,d i , (i=0,…,n) can be obtained by solving the expressions in equations (6), (14), and (15). Solving the linear equations according to equation (15) yields Combined with formula (2), the expression of the integrated density function of the sampling detection flow at the sampling point can be obtained.

[0071] (3) Integrate the sampling detection flow integrated density function in the sampling domain to obtain the sampling flow detection signal energy accumulation function; the expression of the single sampling flow detection signal energy accumulation function P(.) is:

[0072]

[0073] Where: n p is the sample length in the sampling domain; c0 is the reference offset coefficient of the sampling flow detection signal, c0=4.

[0074] (4) Calculate the weighted diffusion factor k of the energy accumulation function of the sampled flow detection signal:

[0075]

[0076]

[0077] in:

[0078] n k is the forgetting sequence length of the diffusion factor;

[0079] c(i) is the forgetting coefficient of the i-th weighted diffusion factor;

[0080] P(i) is the energy accumulation function value of the i-th sampled flow detection signal, mA;

[0081] v(i) is the actual measured value of the i-th sampling flow rate corresponding to P(i), mL.

[0082] (5) Calculate the real-time sampling volume. The calculation formula for the real-time sampling volume v(t) is:

[0083] v(t)=kP(t), formula (19)

[0084] Where: P(t) is the current sampling flow detection signal energy accumulation value, mA; t is the current moment; v(t) is the current slurry sampling volume calculation value, mL.

[0085] like Figure 2 As shown in the figure, the input parameters of the flow signal integration method are (x, y), f0', f n '、n k, v(.), where x is the time series element in the sampling point sample domain, y is the flow sampling sequence element in the sampling point sample domain, f0',f n ' is the boundary constraint condition of the sampled flow signal sequence, n k is the length of the forgetting coefficient sequence, v(.) is the offline measurement value of the sampling volume at the sampling point. The output of the flow signal integration method is the sampling volume at the sampling point.

[0086] The calculation process of sampling point sampling volume accumulation is as follows: Figure 3 As shown, the details are as follows:

[0087] ① The on-site industrial control machine obtains the flow sampling signal and the corresponding timestamp sequence of the sampling signal in real time through the IO channel (x i ,y i ) / (x i ,f(x i )) and its sample domain length n;

[0088] ②According to the flow constraint judgment rules of the sampling point, obtain the sampling domain and sampling point sample data information

[0089] Q(.)={x i ,f(x i ),i=0,…,n}, and initialize the boundary constraints f0'=0,f n ' = 0;

[0090] ③Calculate h i =x i+1 -x i ,y i =f(x i ), (i=1,…,n-1) and

[0091]

[0092] ④Calculation and

[0093] ⑤Let λ0=1,μ n =1, solve The equation system can be obtained Cubic spline interpolation coefficient sequence;

[0094] ⑥ General Substitute the interpolation coefficient into the following formula, where

[0095] i=0,…,n-1, The sampling detection flow integrated density function S(x) can be obtained;

[0096] ⑦ The sampling detection flow integrated density function S(x) in the sample domain x∈[x i,x i+1 ], (i=0,…,n-1) and integrate to obtain the energy accumulation function P(x) of the sampling flow detection signal;

[0097] ⑧ Use a metering bucket to measure the sampling volume offline, and confirm the completion of the operation after entering relevant data information (sampling volume, fuzzy sampling time) on the on-site industrial control computer operation terminal;

[0098] ⑨ Repeat steps ① to ⑧ until the number of offline measurements is greater than or equal to the forgetting coefficient n k ;

[0099] ⑩Calculate the weighted forgetting coefficient

[0100] Calculate the weighted diffusion factor

[0101] If a new offline measurement sampling volume is entered into the on-site industrial control computer system, the integration function module automatically performs steps ① to Operate calculations to complete parameter updates in the cumulative model;

[0102] According to the sampling point determination rule, the current sampling domain data information is automatically tracked and acquired, the integrated sampling volume v(t)=k·p(t) of the current sampling point is calculated, and the single sampling volume calculation is completed.

[0103] The flow interpolation and accumulation device based on the integral diffusion factor of the present invention comprises three parts, namely, a flow data acquisition mechanism, a remote transmission terminal, and a flow accumulation module. Among them, the flow data acquisition mechanism comprises a sampling mechanism and a non-contact flow sensor, and the flow accumulation module runs in a data server and is the core of the method involved in the present invention.

[0104] See also Figure 1 The flow interpolation integration method based on the integral diffusion factor of the present invention includes a sampling machine 1, a measuring cylinder 2, a flow sensor 4, an industrial control machine 5, a terminal module 6, a cloud server 8, a query terminal 9, and a mobile terminal 10. The flow sensor 4 and the terminal module 6 are connected to the industrial control machine 5, and the terminal module 6, the cloud server 8, the query terminal 9, and the mobile terminal 10 are connected to the Internet of Things 7. The sampling machine 1 delivers the sampled slurry to the measuring cylinder 2, and the sampling result of the sampling machine 1 is delivered to the industrial control machine 5. The industrial control machine 5 includes a sampling signal density fitting model module, a sampling signal energy integration module, a sampling amount output module, a parameter library, and a protocol analysis packaging component. The sampling signal density fitting model module identifies the sampling queue {x corresponding to the sampling start point and the end point according to the flow detection signal output by the flow sensor i ,f(x i)}, (i=0,…,n), the cubic spline fitting interpolation method is used to obtain the continuously differentiable sampling density fitting model expression; the sampling signal energy integration is the integration intensity of the cubic spline interpolation function of the flow signal, and its corresponding module is obtained by continuous integration in the sampling queue area based on the model expression of the sampling signal density fitting module and the sampling flow reference deviation signal. The sampling signal energy integration module calculates the sampling flow integral diffusion factor online based on the dynamic historical integration vector and forgetting coefficient in the fixed-length sliding window from the integration parameter table in the parameter library, and then obtains the integrated sampling volume from the energy value of the current flow signal. The sampling volume output module is a flow interpolation integration device application that provides the integrated sampling volume to the outside in two ways. For third-party program interprocess calls, the application provides SDK development interface and call parameter table; for third-party device communication, the application provides ModbusRTU communication protocol. The parameter library is located in the SQLCE database in the industrial control machine. It is automatically created after the application is installed. It mainly includes historical data tables and integrated parameter tables. The historical data table contains information such as sampling point length, sampling sample elements, sampling start and end time, sampling integrated quantity, and test sampling quantity. The integrated parameter table includes information such as cubic spline interpolation coefficients, integrated energy, dynamic historical integrated vectors, and forgetting coefficients, which are called by the sampling volume integrated module. The protocol parsing and encapsulation component mainly encapsulates the current sampling flow signal, corresponding flow collection time, sampling integrated quantity and other perception and measurement data and additional information of the Internet of Things module according to batch transmission rules and uploads them to the cloud server platform. Terminal module 6 is an NB-IoT terminal module.

[0105] A sampling tube is installed on the sampler 1. The working principle of the sampler is that the flushing water from the process flows into the sampler through the water valve. When the water level in the sampler reaches a certain height, the counterweight balance of the sampler is broken, the sampler overturns, and the internal flushing water flows out. At this time, the sampling tube connected to the sampler cuts the liquid slurry waterfall surface for sampling. After the flushing water flows out, the sampler returns to its original balance position, and the sampling slurry in the sampling tube flows into the sampling box, completing a sampling. The sampling time interval and sampling frequency can be controlled by adjusting the opening of the water valve above the sampler.

[0106] The measuring cylinder 2 is a measuring cylinder for offline measuring the actual sampling volume, and is used to measure the volume of a single sampling slurry irregularly as a necessary parameter for the recursive calculation of the weighted diffusion factor. Figure 1 3 is the on-site operator, who mainly inputs the sampling flow measurement results and the corresponding sampling fuzzy time domain into the on-site industrial control computer as necessary parameters for the recursive calculation of the weighted diffusion factor. Figure 1 The middle component 4 is a flow sensor, which is used to dynamically sense the changes in the sampled flow and output the measurement results in the form of analog signals to the IO channel of the on-site industrial control computer.

[0107] The on-site industrial control machine 5 is mainly used for the functions of model application operation, data input and output, and real-time display of key parameters in the flow integration method. In the present invention, the on-site industrial control machine adopts a customized paperless recorder, pre-installed WinCE operating system, and the development environment of application programs and functional modules such as sampling signal density fitting model, sampling flow signal energy integration module, sampling volume output module and protocol analysis packaging component is Visual Studio 2015. The variables in the parameter library include sampling point characteristic parameters (peak corresponding timestamp in the sampling domain, sampling domain length, samples in the sampling domain, boundary constraints, cubic spline interpolation density fitting coefficients, integration function coefficients, weighting factors (sub-item weighting factors and forgetting sequences), sampling volume offline measurement sequences, etc.). The parameter library variables are stored in the corresponding data table created and deployed in the SQLCE database. The protocol packaging and analysis component mainly uses a custom protocol to encapsulate data information such as real-time detection signals of flow sensors, sampling point characteristic parameters (sampling point peaks, sampling integration), etc., and uploads them to the enterprise cloud server through batch transmission via the NB-IoT module deployed on site.

[0108] The terminal module 6 is an NB-IoT terminal module, which mainly uploads the batch information sent by the on-site industrial control machine through the RS485 serial port to the enterprise cloud server 8. The Internet of Things 7 is an Internet of Things service provided by operators (China Telecom, China Mobile, China Unicom), which realizes seamless connection with the enterprise cloud platform server by configuring the terminal module channel information. The cloud server 8 is used to store the sampling point data information, analyze and process the relevant data, and provide Web publishing to the query terminal 9 in the network and APP instant information push function to the mobile terminal 10.

[0109] The problems to be solved by the flow interpolation and accumulation method and device based on integral diffusion factor of the present invention are as follows: (1) Using the data information such as the ordered flow queue and its timestamp output by the flow sensor and the remote transmission terminal, a generalized sample domain of the sampled flow is constructed. According to the sampling point boundary constraint specifications and conditions, the flow detection signal is subjected to a cubic spline interpolation algorithm to determine a fitting function expression of the sampling flow density in the sampling domain; (2) the sampling flow density fitting curve in the sampling domain is segmentedly integrated in the sampling space domain to obtain the energy accumulation function expression of the sampling flow detection signal, and the single energy absolute accumulation of the sampling flow detection signal can be obtained by summing up in the sampling space. According to the baseline distribution of the sampling signal output offset characteristic in the sampling domain, the single energy relative value of the sampling flow detection signal can be obtained; (3) In order to reduce and eliminate the influence of manual measurement deviation, linear and nonlinear output drift of the detection instrument device itself, process parameter fluctuation, and external environmental disturbance on the integration model, the present invention introduces a weighted diffusion factor into the integration model, that is, the single energy relative value constructed according to the flow detection signal and the offline sampling volume measurement ratio measured by the test cylinder, and the ratio is a diffusion factor constituting the integration model. The weighted diffusion factor of the integration model can be obtained by weighting the multiple diffusion factors with independent and unrelated characteristics at different times of the same sampling system using parameters based on the forgetting coefficient; (4) According to the generalized sampling domain formed by the sampling detection value of the current sampling point and other related parameters, the sampling points and their sampling domains are tracked and marked in real time, and the sampling cubic spline interpolation is performed in the marked sampling domain to finally form the relative value of the sampling energy based on the flow detection signal, which is converted into the current sampling flow integration value through the weighted diffusion factor of the integration model.

Claims

1. A flow interpolation and integration method based on integral diffusion factor, characterized in that: The method comprises the following steps: (1) detecting a generalized sampling domain according to the flow rate output by the flow sensor, and obtaining a data queue of sampling points in the sampling domain; (2) fitting the sampling detection flow integration density function according to the sampling point data and the sampling point data queue; (3) Integrate the sampling detection flow rate integrated density function in the sampling domain to obtain the sampling flow rate detection signal energy accumulation function; (4) Calculate the weighted diffusion factor k of the energy accumulation function of the sampled flow detection signal. The calculation formula is: in: n k is the forgetting sequence length of the diffusion factor; c(i) is the forgetting coefficient of the i-th weighted diffusion factor, P(i) is the energy accumulation function value of the i-th sampled flow detection signal; v(i) is the actual measured value of the i-th sampling flow corresponding to P(i); (5) Calculate the real-time sampling amount, the calculation formula is: v(t)=kP(t), in: P(t) is the accumulated energy value of the current sampled flow detection signal, and t is the current moment.

2. The flow interpolation and integration method based on the integral diffusion factor according to claim 1 is characterized in that: In step (2), the first and last data points of the sampled detection flow integrated density function are second-order differentiable, and the first-order derivative of the first and last data points of the sampled detection flow integrated density function is 0.

3. The flow interpolation and integration method based on the integral diffusion factor according to claim 2 is characterized in that: In step (2), according to {x i |f(x i ), i = 0, ..., n}, and cubic spline interpolation is used to fit the sampling detection flow integrated density function S(x): in: x i is the i-th time value of the sampling point data queue in the sampling domain, where i = 0,…,n; f(x i )=y i is the i-th sampled flow signal value in the sampling domain, where i = 0,…,n; h i is the time step, which is the difference between the current moment and the previous moment, where i = 0,…,n-1, h i =x i+1 -x i ; M i , i=0,…,n is the second-order derivative of the sampling detection flow integrated density function at the i-th moment, and M i Satisfy the formula: μ i ,λ i , i=1,…,n-1 is the current time step h i and the adjacent time step h i-1 ,h i+1 The linear equation expression that determines the constant coefficient, d i , i=1,…,n-1 is the number of the previous time step h i-1 , current time step h i and the i-th sampling flow signal value y in the sampling domain i and adjacent sampling flow signal value y i-1 ,y i+1 Determine the constant coefficients of the linear equation expression.

4. The flow interpolation and accumulation method based on the integral diffusion factor according to claim 1 is characterized in that: The expression of the energy accumulation function P(.) of the single sampling flow detection signal in step (3) is: Where: n p is the sample length in the sampling domain; c0 is the reference offset coefficient of the sampling flow detection signal, c0=4.

5. An integration device based on the flow interpolation integration method based on the integral diffusion factor according to any one of claims 1 to 4, characterized in that: The device comprises a sampling machine (1), a measuring cylinder (2), a flow sensor (4), an industrial control machine (5), a terminal module (6), a cloud server (8), a query terminal (9), and a mobile terminal (10); the flow sensor (4) and the terminal module (6) are all connected to the industrial control machine (5); the terminal module (6), the cloud server (8), the query terminal (9), and the mobile terminal (10) are all connected to the Internet of Things (7); the sampling machine (1) conveys the sampled slurry to the measuring cylinder (2) by gravity, and the sampling result of the sampling machine (1) is conveyed to the industrial control machine (5).

6. The integrator of the flow interpolation integrator method based on the integral diffusion factor according to claim 5, characterized in that: The industrial control machine (5) comprises a sampling signal density fitting model module, a sampling signal energy accumulation module, a sampling quantity output module, a parameter library, and a protocol analysis packaging component; the terminal module (6) is an NB-IoT terminal module.

Images