CN103575351A - Measuring method and measuring system of primary air volume of power station boiler - Google Patents

Abstract

The application discloses a method and system for measuring the primary air volume of a utility boiler, including the following steps: obtaining auxiliary variables; filtering the errors of the auxiliary variables; establishing a measurement model of the primary air volume based on the support vector regression algorithm, and substituting the auxiliary variables into the measurement The model is calculated to obtain the primary air volume. It can be seen from the above-mentioned technical scheme that in the measurement method and system disclosed in the present application, the auxiliary variable is firstly obtained, then the error in the auxiliary variable is filtered out, and finally a measurement model of the primary air volume is established based on the support vector regression algorithm, and the auxiliary variable Substituting the measurement model for calculation to obtain the primary air volume instead of using the measuring device to directly measure the primary air volume, so the measurement results will not seriously drift over time due to the high-speed erosion of the dust-containing primary wind. The problem of large deviation between the measurement result obtained by using the measurement device to directly measure the primary air volume and the actual value is solved.

Description

A method and system for measuring the primary air volume of a utility boiler

technical field

The present application relates to the technical field of thermal power generation, and more specifically, relates to a method and system for measuring the primary air volume of a utility boiler.

Background technique

Accurate measurement of the primary air volume of power plant boilers is a key factor to determine a reasonable air-to-coal ratio and improve combustion efficiency, and is also a prerequisite for the stable operation of the combustion control system. Therefore, the measurement results must be accurate and reliable. At present, the primary air volume measurement mainly uses measuring devices such as Venturi tubes, differential pressure flowmeters, and airfoil flowmeters to directly measure the primary air. The dust content of the primary air of power plant boilers is generally relatively high. The measuring device has been scoured by the high-speed dust-containing primary air for a long time, and the wear is serious, which leads to the serious drift of the measurement results with the change of running time, which makes the deviation between the measurement result obtained by using the measuring device to directly measure the primary air volume and the actual value is relatively large. big.

Contents of the invention

In view of this, the present application provides a method for measuring the primary air volume of a power plant boiler to solve the problem of a large deviation between the measurement result obtained by using a measuring device to directly measure the primary air volume and the actual value.

In order to achieve the above purpose, the proposed scheme is as follows:

A method for measuring the primary air volume of a utility boiler, comprising the following steps:

get helper variables;

filtering out errors of said auxiliary variables;

A measurement model of the primary air volume is established based on a support vector regression algorithm, and the auxiliary variable is substituted into the measurement model for calculation to obtain the primary air volume.

Preferably, said acquisition of auxiliary variables includes:

Get the primary air pressure of the coal mill, the pressure of the coal mill separator, the opening of the regulating door, the current of the primary fan, the current of the coal mill, the differential pressure of the primary air inlet and outlet of the air preheater, the instantaneous coal supply and the temperature difference of the primary air.

Preferably, the filtering out the error of the auxiliary variable includes:

Using the clipping method to remove the gross error of the auxiliary variable;

A filtering algorithm is used to reduce the random error of the auxiliary variable.

Preferably, the measurement model of the primary air volume is established based on the support vector regression algorithm, and the auxiliary variables are substituted into the measurement model for calculation to obtain the primary air volume, including:

Get the training set;

Construct kernel function;

Construct and solve the quadratic programming problem, and solve the Lagrangian coefficient;

Selecting a solution that satisfies the quadratic programming problem;

constructing said measurement model;

The auxiliary variable is substituted into the measurement model for calculation to obtain the primary air volume.

Preferably, it also includes:

The parameters of the measurement model are corrected online.

A system for measuring the primary air volume of a utility boiler, comprising:

obtaining means, for obtaining auxiliary variables;

filtering means for filtering out errors of the auxiliary variable;

The calculation device is used to establish a primary air volume model, and calculate the auxiliary variable to obtain and output the primary air volume.

Preferably, the auxiliary variables include:

The primary air pressure of the coal mill, the pressure of the coal mill separator, the opening of the regulating door, the current of the primary fan, the current of the coal mill, the differential pressure of the primary air inlet and outlet of the air preheater, the instantaneous coal supply and the temperature difference of the primary air.

Preferably, the filtering device includes:

The first filtering module is used to eliminate the gross error of the auxiliary variable by using the clipping method;

The second filtering module is used to reduce the random error of the auxiliary variable by using a filtering algorithm.

Preferably, the computing device includes:

The input module is used to obtain the training set;

The first calculation module is used to construct a kernel function;

The second calculation module is used to construct and solve the quadratic programming problem, and obtain the Lagrangian coefficient;

A selection module is used to select a solution that satisfies the quadratic programming problem;

A construction module for constructing the measurement model;

The third calculation module is used to substitute the auxiliary variable into the measurement model for calculation, and output the primary air volume.

Preferably, it also includes:

A correction device is used for online correction of the parameters of the calculation device.

It can be seen from the above technical solution that in the method for measuring the primary air volume of a utility boiler disclosed in the present application, the auxiliary variable is first obtained, then the error in the auxiliary variable is filtered out, and finally the measurement model of the primary air volume is established based on the support vector regression algorithm , Substituting the auxiliary variable into the measurement model for calculation to obtain the primary air volume, instead of using the measuring device to directly measure the primary air volume, so the measurement results will not change over time due to the high-speed erosion of the dust-containing primary wind and serious drift, so as to solve the problem of large deviation between the measurement result obtained by using the measuring device to directly measure the primary air volume and the actual value.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments of the present application. Those skilled in the art can also obtain other drawings based on these drawings without creative work.

Fig. 1 is a flow chart of a method for measuring the primary air volume of a utility boiler disclosed in an embodiment of the present application;

Fig. 2 is a flow chart of a method for measuring the primary air volume of a utility boiler disclosed in another embodiment of the present application;

Fig. 3 is a flow chart of a method for measuring the primary air volume of a utility boiler disclosed in another embodiment of the present application;

Fig. 4 is a structural diagram of a measuring system for primary air volume of a utility boiler disclosed in another embodiment of the present application;

Fig. 5 is a structural diagram of a measuring system for the primary air volume of a utility boiler disclosed in another embodiment of the present application;

Fig. 6 is a structural diagram of a measuring system for primary air volume of a utility boiler disclosed in another embodiment of the present application.

Detailed ways

The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all of them. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of this application.

Embodiment one

Fig. 1 is a flow chart of a method for detecting primary air volume of a utility boiler disclosed in Embodiment 1 of the present application.

As shown in Figure 1, the method for detecting the primary air volume of a utility boiler disclosed in this embodiment includes the following steps:

S101: Acquiring auxiliary variables;

S102: Filter out errors of auxiliary variables;

S103: Establish a measurement model of the primary air volume based on the support vector regression algorithm, substitute auxiliary variables into the measurement model for calculation, and obtain the primary air volume.

It can be seen from the above technical solution that in the detection method of the primary air volume of the power plant boiler disclosed in this embodiment, the auxiliary variable is obtained first, then the error in the auxiliary variable is filtered, and finally the measurement model of the primary air volume is established based on the support vector regression algorithm , Substituting the auxiliary variables into the measurement model for calculation to obtain the primary air volume instead of using the measuring device to directly measure the primary air volume, so the measurement results will not seriously drift over time due to the high-speed erosion of the dust-containing primary wind , so as to solve the problem of large deviation between the measurement result obtained by using the measurement device to directly measure the primary air volume and the actual value.

Embodiment two

Fig. 2 is a flow chart of a method for measuring the primary air volume of a utility boiler disclosed in another embodiment of the present application.

As shown in Figure 2, the method for measuring the primary air volume of a utility boiler disclosed in this embodiment includes the following steps:

S201: Acquiring auxiliary variables;

First select the auxiliary variable to be obtained. In this embodiment, auxiliary variables are selected based on the combination of mechanism and principal component analysis.

The so-called mechanism is to analyze the factors affecting the primary air volume according to the principle of fluid mechanics, so as to determine the influence of each factor on the air volume. The so-called principal component analysis is to find a group of variables, and carry as much useful information of the original variables as possible when the number of variables is small, such as using the cumulative contribution rate method of the principal components to determine the number of principal components.

According to the results of the above analysis, the auxiliary variables selected in this embodiment include the primary air pressure of the coal mill, the pressure of the separator of the coal mill, the opening of the regulating door, the current of the primary fan, the current of the obtained coal mill, and the primary air intake of the air preheater. Outlet differential pressure, instantaneous coal supply and primary air temperature difference.

S202: Using the clipping method to eliminate the gross error of the auxiliary variable;

S203: Using a filtering algorithm to reduce the random error of the auxiliary variable;

S204: Obtain a training set;

Given a training set {(p ₁ ,y ₁ ),…,(p _k ,y _k )} where the input p _i ∈ R ⁿ , the output y _i ∈ R ^k , i=1,2…k; the training set is from Part of the sample data obtained from the field test of a specific unit, where p _i ∈ R ⁿ is the selected auxiliary variable, and y _i ∈ R ^k is the leading variable (here is the primary air volume).

The training set includes training samples and test samples. 2000 sets of data are selected from the DCS historical data provided by a domestic power plant as training samples, and 200 sets of data are selected as test samples. Use the normalization function to normalize the training samples and test samples.

S205: Construct a kernel function;

Construct an appropriate kernel function. In this embodiment, the radial basis kernel function (RBF) is preferred as the kernel function, and an appropriate precision ε>0 and penalty parameter C>0 are selected. The precision and penalty parameters generally adopt the default values of the model algorithm directly.

S206: Construct and solve the quadratic programming problem, and obtain the Lagrangian coefficient;

$\stackrel{<span\; class="notranslate">\; ~</span>}{\mathrm{<span\; class="notranslate">\; a</span>}}<span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; (</span>{\stackrel{<span\; class="notranslate">\; ~</span>}{\mathrm{<span\; class="notranslate">\; a</span>}}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>{\stackrel{<span\; class="notranslate">\; ~</span>}{\mathrm{<span\; class="notranslate">\; a</span>}}}_{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; ,</span>{\stackrel{<span\; class="notranslate">\; ~</span>}{\mathrm{<span\; class="notranslate">\; a</span>}}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; (</span>{{\stackrel{<span\; class="notranslate">\; ~</span>}{\mathrm{<span\; class="notranslate">\; a</span>}}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>{{\stackrel{<span\; class="notranslate">\; ~</span>}{\mathrm{<span\; class="notranslate">\; a</span>}}}_{\mathrm{<span\; class="notranslate">\; k</span>}}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; T</span>}}$

$\underset{\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; a</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; \&Element;</span>{\mathrm{<span\; class="notranslate">\; R</span>}}^{\mathrm{<span\; class="notranslate">\; k</span>}}}{\mathrm{<span\; class="notranslate">\; min</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 2</span>}}\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; k</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; K</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; \&epsiv;</span>}\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; k</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; k</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span>$

$\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; .</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; .</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; k</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>{{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0,0</span>}<span\; class="notranslate">\; \&le;</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; ,</span>{{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; C</span>}<span\; class="notranslate">\; ,</span>$

i,j=1,2,...,k

为使用拉格朗日法求解二次规划问题（即最值问题）时引入的拉格朗日常数，核函数采用K(p_i,p)=exp(-γ*|p_i-p|²)，式中γ为常数，可以给定任意的初始值或使用默认的初始值，训练时会自动寻优得到更合理的值，其中p_i为训练时获得的支持向量,p为实测的辅助变量归一化后组成的向量，

为训练后的线性模型的常数项，然后采用训练样本进行模型的训练、参数选择及模型验证；

In order to use the Lagrange method to solve the quadratic programming problem (that is, the most value problem), the Lagrange daily number introduced, the kernel function uses K(p _i ,p)=exp(-γ*|p _i -p| ² ), where γ is a constant, any initial value can be given or the default initial value can be given, and a more reasonable value will be automatically optimized during training, where p _i is the support vector obtained during training, and p is the measured auxiliary A vector of normalized variables,

is the constant term of the trained linear model, and then use the training samples to carry out model training, parameter selection and model verification;

S207: Select a solution satisfying the quadratic programming problem;

或

计算

Choose a solution that satisfies the above constraints

or

calculate

$\tilde{b}={\mathrm{the\; y}}_j-\underset{i=1}{\overset{k}{\mathrm{\&Sigma;}}}({{\tilde{a}}_i}^{*}-{\tilde{a}}_i)K(p_i,p_j)+\mathrm{\&epsiv;}$ or $\tilde{b}={\mathrm{the\; y}}_t-\underset{i=1}{\overset{k}{\mathrm{\&Sigma;}}}({{\tilde{a}}_i}^{*}-{\tilde{a}}_i)K(p_i,p_t)-\mathrm{\&epsiv;}$

S208: Construct a measurement model;

$\mathrm{<span\; class="notranslate">\; the\; y</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; g</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; k</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}<span\; class="notranslate">\; (</span>{{\stackrel{<span\; class="notranslate">\; ~</span>}{\mathrm{<span\; class="notranslate">\; a</span>}}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; -</span>{\stackrel{<span\; class="notranslate">\; ~</span>}{\mathrm{<span\; class="notranslate">\; a</span>}}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; K</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\stackrel{<span\; class="notranslate">\; ~</span>}{\mathrm{<span\; class="notranslate">\; b</span>}}$

Where Y is the primary air volume.

S209: Substituting the auxiliary variable into the measurement model for calculation to obtain the primary air volume.

Embodiment three

Fig. 3 is a flow chart of a method for measuring primary air volume of a utility boiler disclosed in another embodiment of the present application.

As shown in Figure 3, the method for measuring the primary air volume of a utility boiler disclosed in this embodiment includes the following steps:

S301: Acquiring auxiliary variables;

S302: Using a clipping method to eliminate gross errors of auxiliary variables;

S303: Using a filtering algorithm to reduce the random error of the auxiliary variable.

S304: Obtain a training set;

S305: Construct a kernel function;

S306: Construct and solve the quadratic programming problem, and obtain the Lagrangian coefficient;

S307: Select a solution satisfying the quadratic programming problem;

S308: Construct a measurement model

S309: Substituting the auxiliary variable into the measurement model for calculation to obtain the primary air volume.

S310: Perform online correction to the parameters of the measurement model.

The parameters of the measurement model are corrected through a large number of sample data obtained in online operation.

It can be seen from the above technical solution that this embodiment adds step S310 on the basis of the previous embodiment, which is used for online correction of the parameters of the measurement model, so that the obtained primary air volume is closer to the actual air volume, The accuracy of measurement can be improved.

Embodiment four

FIG. 4 is a structural diagram of a system for measuring the primary air volume of a utility boiler disclosed in another embodiment of the present application, including an acquisition device 10 , a filter device 20 and a calculation device 30 .

The acquiring device 10 is used for acquiring auxiliary variables.

The filtering device 20 is used to filter out errors of auxiliary variables.

The calculation device 30 is used to establish a primary air volume measurement model, and calculate auxiliary variables to obtain and output the primary air volume.

It can be seen from the above technical solution that in the measurement system of the primary air volume of the utility boiler disclosed in this embodiment, the auxiliary variable is obtained first, then the error in the auxiliary variable is filtered out, and finally the measurement of the primary air volume is established based on the support vector regression algorithm Model, the auxiliary variable is substituted into the measurement model to calculate the primary air volume, instead of using the measuring device to directly measure the primary air volume, so the measurement results will not change over time due to the high-speed erosion of the dust-containing primary wind. Drift, so as to solve the problem of large deviation between the measurement result obtained by using the measurement device to directly measure the primary air volume and the actual value.

Embodiment five

Fig. 5 is a structural diagram of a measuring system for primary air volume of a utility boiler disclosed in another embodiment of the present application.

As shown in FIG. 5 , the measuring system for the primary air volume of a utility boiler disclosed in this embodiment includes: an acquisition device 10 , a filter device 20 and a calculation device 30 , wherein the filter device 20 includes a first filter module 21 and a second filter module 22 ; The calculation device 30 includes an input module 31 , a first calculation module 32 , a second calculation module 33 , a selection module 34 , a construction module 35 and a third calculation module 36 . in:

The acquiring device 10 is used for acquiring auxiliary variables, and in the process of acquiring variables, the auxiliary variables to be acquired are first selected. In this embodiment, auxiliary variables are selected based on the combination of mechanism and principal component analysis.

The so-called mechanism is to analyze the factors affecting the primary air volume according to the principle of fluid mechanics, so as to determine the influence of each factor on the air volume. The so-called principal component analysis is to find a group of variables, and carry as much useful information of the original variables as possible when the number of variables is small, such as using the cumulative contribution rate method of the principal components to determine the number of principal components.

According to the results of the above analysis, the auxiliary variables selected in this embodiment include the primary air pressure of the coal mill, the pressure of the separator of the coal mill, the opening of the regulating door, the current of the primary fan, the current of the obtained coal mill, and the primary air intake of the air preheater. Outlet differential pressure, instantaneous coal supply and primary air temperature difference.

The first filtering device 21 is used for receiving the auxiliary variable, and adopting a clipping method to eliminate gross errors of the auxiliary variable.

The second filtering device 22 is used to receive the auxiliary variable after gross errors have been eliminated, and use a filtering algorithm to reduce the random error of the auxiliary variable.

Input module 30 is used for obtaining training set;

Training set {(p ₁ ,y ₁ ),…,(p _k ,y _k )} where input p _i ∈ R ⁿ , output y _i ∈ R ^k , i=1,2…k; Part of the sample data obtained from the field test, where p _i ∈ R ⁿ is the selected auxiliary variable, and y _i ∈ R ^k is the leading variable (here is the primary air volume).

The training set includes training samples and test samples. 2000 sets of data are selected from the DCS historical data provided by a domestic power plant as training samples, and 200 sets of data are selected as test samples. Use the normalization function to normalize the training samples and test samples.

The first calculating module 32 is used for constructing kernel function;

Construct an appropriate kernel function. In this embodiment, the radial basis kernel function (RBF) is preferred as the kernel function, and an appropriate precision ε>0 and penalty parameter C>0 are selected. The precision and penalty parameters generally adopt the default values of the model algorithm directly.

The second calculation module 33 is used to construct and solve the quadratic programming problem, and solve the Lagrangian coefficient;

$\stackrel{<span\; class="notranslate">\; ~</span>}{\mathrm{<span\; class="notranslate">\; a</span>}}<span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; (</span>{\stackrel{<span\; class="notranslate">\; ~</span>}{\mathrm{<span\; class="notranslate">\; a</span>}}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>{\stackrel{<span\; class="notranslate">\; ~</span>}{\mathrm{<span\; class="notranslate">\; a</span>}}}_{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; ,</span>{\stackrel{<span\; class="notranslate">\; ~</span>}{\mathrm{<span\; class="notranslate">\; a</span>}}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; (</span>{{\stackrel{<span\; class="notranslate">\; ~</span>}{\mathrm{<span\; class="notranslate">\; a</span>}}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>{{\stackrel{<span\; class="notranslate">\; ~</span>}{\mathrm{<span\; class="notranslate">\; a</span>}}}_{\mathrm{<span\; class="notranslate">\; k</span>}}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; T</span>}}$

$\underset{\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; a</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; \&Element;</span>{\mathrm{<span\; class="notranslate">\; R</span>}}^{\mathrm{<span\; class="notranslate">\; k</span>}}}{\mathrm{<span\; class="notranslate">\; min</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; 2</span>}}\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; k</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; K</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; \&epsiv;</span>}\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; k</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; k</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span>$

$\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; .</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; .</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; k</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>{{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0,0</span>}<span\; class="notranslate">\; \&le;</span>{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; ,</span>{{\mathrm{<span\; class="notranslate">\; a</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; C</span>}<span\; class="notranslate">\; ,</span>$

i,j=1,2,...,k

为使用拉格朗日法求解二次规划问题（即最值问题）时引入的拉格朗日常数，核函数采用K(p_i,p)=exp(-γ*|p_i-p|²)，式中γ为常数，可以给定任意的初始值或使用默认的初始值，训练时会自动寻优得到更合理的值，其中p_i为训练时获得的支持向量,p为实测的辅助变量归一化后组成的向量，

为训练后的线性模型的常数项，然后采用训练样本进行模型的训练、参数选择及模型验证；

In order to use the Lagrange method to solve the quadratic programming problem (that is, the most value problem), the Lagrange daily number introduced, the kernel function uses K(p _i ,p)=exp(-γ*|p _i -p| ² ), where γ is a constant, any initial value can be given or the default initial value can be given, and a more reasonable value will be automatically optimized during training, where p _i is the support vector obtained during training, and p is the measured auxiliary A vector of normalized variables,

is the constant term of the trained linear model, and then use the training samples to carry out model training, parameter selection and model verification;

Selection module 34 is used for selecting the solution that satisfies the quadratic programming problem;

或计算

Choose a solution that satisfies the above constraints

or calculate

$\tilde{b}={\mathrm{the\; y}}_j-\underset{i=1}{\overset{k}{\mathrm{\&Sigma;}}}({{\tilde{a}}_i}^{*}-{\tilde{a}}_i)K(p_i,p_j)+\mathrm{\&epsiv;}$ or $\tilde{b}={\mathrm{the\; y}}_t-\underset{i=1}{\overset{k}{\mathrm{\&Sigma;}}}({{\tilde{a}}_i}^{*}-{\tilde{a}}_i)K(p_i,p_t)-\mathrm{\&epsiv;}$

The construction module 35 is used to construct the measurement model.

$\mathrm{<span\; class="notranslate">\; the\; y</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; g</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; k</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}<span\; class="notranslate">\; (</span>{{\stackrel{<span\; class="notranslate">\; ~</span>}{\mathrm{<span\; class="notranslate">\; a</span>}}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; -</span>{\stackrel{<span\; class="notranslate">\; ~</span>}{\mathrm{<span\; class="notranslate">\; a</span>}}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; K</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; p</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>\stackrel{<span\; class="notranslate">\; ~</span>}{\mathrm{<span\; class="notranslate">\; b</span>}}$

Where y is the primary air volume.

The third calculation module 36 is used for substituting auxiliary variables into the measurement model for calculation to obtain and output the primary air volume.

Embodiment six

Fig. 6 is a structural diagram of a measuring system for primary air volume of a utility boiler disclosed in another embodiment of the present application.

As shown in FIG. 6 , the measurement system for the primary air volume of the utility boiler disclosed in this embodiment is based on the previous embodiment with an additional correction device 40 , which is connected to the computing device 30 .

The correction device 40 corrects the parameters of the measurement model through a large amount of sample data obtained during online operation.

It can be seen from the above technical solution that this embodiment adds a correction device 40 on the basis of the previous embodiment, which is used for online correction of the parameters of the measurement model, so that the obtained primary air volume is closer to the actual air volume. The accuracy of measurement can be improved.

Finally, it should also be noted that in this text, relational terms such as first and second etc. are only used to distinguish one entity or operation from another, and do not necessarily require or imply that these entities or operations, any such actual relationship or order exists. Furthermore, the term "comprises", "comprises" or any other variation thereof is intended to cover a non-exclusive inclusion such that a process, method, article, or apparatus comprising a set of elements includes not only those elements, but also includes elements not expressly listed. other elements of or also include elements inherent in such a process, method, article, or device. Without further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article or apparatus comprising said element.

Each embodiment in this specification is described in a progressive manner, each embodiment focuses on the difference from other embodiments, and the same and similar parts of each embodiment can be referred to each other.

The above description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the application. Therefore, the present application will not be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims (10)

1. A method for measuring primary air volume of a utility boiler, characterized in that, comprises the steps: get helper variables; filtering out errors of said auxiliary variables; A measurement model of the primary air volume is established based on a support vector regression algorithm, and the auxiliary variable is substituted into the measurement model for calculation to obtain the primary air volume. 2. measurement method as claimed in claim 1, is characterized in that, described acquisition auxiliary variable comprises: Get the primary air pressure of the coal mill, the pressure of the coal mill separator, the opening of the regulating door, the current of the primary fan, the current of the coal mill, the differential pressure of the primary air inlet and outlet of the air preheater, the instantaneous coal supply and the temperature difference of the primary air. 3. measurement method as claimed in claim 1, is characterized in that, described filtering out the error of described auxiliary variable, comprises: Using the clipping method to remove the gross error of the auxiliary variable; A filtering algorithm is used to reduce the random error of the auxiliary variable. 4. measurement method as claimed in claim 1, is characterized in that, described based on support vector regression algorithm and sets up the measurement model of primary wind volume, and described auxiliary variable is substituted into measurement model and calculates, obtains described primary wind volume, comprises : Get the training set; Construct kernel function; Construct and solve the quadratic programming problem, and solve the Lagrangian coefficient; Selecting a solution that satisfies the quadratic programming problem; constructing said measurement model; The auxiliary variable is substituted into the measurement model for calculation to obtain the primary air volume. 5. The measurement method according to any one of claims 1 to 4, further comprising: The parameters of the measurement model are corrected online. 6. A measuring system for the primary air volume of a utility boiler, characterized in that it comprises: obtaining means, for obtaining auxiliary variables; filtering means for filtering out errors of the auxiliary variable; The calculation device is used to establish a primary air volume model, and calculate the auxiliary variable to obtain and output the primary air volume. 7. The measurement system of claim 6, wherein said auxiliary variables comprise: The primary air pressure of the coal mill, the pressure of the coal mill separator, the opening of the regulating door, the current of the primary fan, the current of the coal mill, the differential pressure of the primary air inlet and outlet of the air preheater, the instantaneous coal supply and the temperature difference of the primary air. 8. The measurement system according to claim 6, wherein said filtering means comprises: The first filtering module is used to eliminate the gross error of the auxiliary variable by using the clipping method; The second filtering module is used to reduce the random error of the auxiliary variable by using a filtering algorithm. 9. The measurement system of claim 6, wherein said computing means comprises: The input module is used to obtain the training set; The first calculation module is used to construct a kernel function; The second calculation module is used to construct and solve the quadratic programming problem, and obtain the Lagrangian coefficient; A selection module is used to select a solution that satisfies the quadratic programming problem; A construction module for constructing the measurement model; The third calculation module is used to substitute the auxiliary variable into the measurement model for calculation, and output the primary air volume. 10. The measurement system according to claims 6-9, further comprising: A correction device is used for online correction of the parameters of the calculation device.

Images