CN101763084B - System and method for minimizing chemical oxygen demand (COD) discharge of pesticide production waste liquid incinerator - Google Patents

Abstract

The invention discloses a system for minimizing chemical oxygen demand (COD) discharge of a pesticide production waste liquid incinerator, comprising a field intelligent instrument connected with the pesticide production waste liquid incinerator, a DCS system and a host computer. The host computer comprises a standardized processing module for collecting training samples TX of key system variables from a database and the chemical oxygen demand data Y corresponding to the training samples TX and standardized processing the training samples TX, a support vector machine (SVM) module for soft-sensor modeling and a particle swarm algorithm module for solving the following minimum problem with the pgK obtained when iteration is ended being the operating variable with minimum chemical oxygen demand. The invention also discloses a method for minimizing the chemical oxygen demand (COD) discharge of the pesticide production waste liquid incinerator. The invention provides a system and a method for minimizing the chemical oxygen demand (COD) discharge of the pesticide production waste liquid incinerator, which can measure the COD online and effectively monitor whether the COD exceeds standard.

Description

The minimizing chemical oxygen demand (COD) discharge of pesticide production waste liquid incinerator system and method

Technical field

The present invention relates to the pesticide producing field, especially, relate to a kind of pesticide production waste liquid incinerator chemical oxygen consumption (COC) (COD) that makes and discharge minimized system and method.

Background technology

Along with developing rapidly of pesticide industry, the problem of environmental pollution of emission has caused the great attention of national governments and corresponding environmental administration.The qualified discharge of research and solution agricultural chemicals organic liquid waste is controlled and harmless minimization, not only becomes the difficult point and the focus of various countries' scientific research, also is the science proposition that is related to the national active demand of social sustainable development simultaneously.

Burning method is to handle the agricultural chemicals raffinate at present and waste residue is the most effective, thorough, use the most general method.The chemical oxygen consumption (COC) of burning back waste water is the most important index that pesticide waste liquid burns hazardous emission, but it can't on-line measurement, and off-line measurement once needs four or five hours, can't in time reflect working conditions change and instruct actual production.Therefore, in actual burning process, the COD severe overweight.

Summary of the invention

Can't on-line measurement in order to overcome existing incinerator process COD, the deficiency of COD severe overweight, the invention provides and a kind ofly can carry out on-line measurement COD, effectively monitor the minimizing chemical oxygen demand (COD) discharge of pesticide production waste liquid incinerator system and method whether COD exceeds standard

The technical solution adopted for the present invention to solve the technical problems is:

A kind of minimizing chemical oxygen demand (COD) discharge of pesticide production waste liquid incinerator system comprises the field intelligent instrument, DCS system and the host computer that are connected with pesticide production waste liquid incinerator, and described DCS system comprises control station and database; Described field intelligent instrument is connected with the DCS system, and described DCS system is connected with host computer, and described host computer comprises:

The standardization module, be used for from the training sample TX of database acquisition system key variables, the chemical oxygen consumption (COC) data Y of training sample TX correspondence, training sample TX is carried out standardization, the average of each variable is 0, variance is 1, obtains input matrix X, adopts following process to finish: its formula is (1), (2), (3):

1.1) computation of mean values: $\stackrel{\‾}{\mathrm{TX}}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{TX}}_i---\left(1\right)$

1.2) basis of calculation is poor: ${\mathrm{\σ}}_x^2=\frac{1}{N-1}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}({\mathrm{TX}}_i-\stackrel{\‾}{\mathrm{TX}}),---\left(2\right)$

1.3) standardization: $X=\frac{\mathrm{TX}-\stackrel{\‾}{\mathrm{TX}}}{{\mathrm{\σ}}_x},---\left(3\right)$

Wherein, TX is a training sample, and N is a number of training, and TX is the average of training sample, σ _xStandard deviation for data sample;

The support vector machine module is used for soft sensor modeling, and its detailed process is as follows: its formula is: (4), (5):

$\underset{\mathrm{\α},{\mathrm{\α}}^{*}}{\max }\{-\frac{1}{2}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{M}{\mathrm{\Σ}}}({\mathrm{\α}}_i-{\mathrm{\α}}_i^{*})({\mathrm{\α}}_j-{\mathrm{\α}}_j^{*})K(x_i,x_j)-\mathrm{\ϵ}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}({\mathrm{\α}}_i+{\mathrm{\α}}_i^{*})+\underset{i=1}{\overset{M}{\mathrm{\Σ}}}{y}_i({\mathrm{\α}}_i-{\mathrm{\α}}_i^{*})\}---\left(4\right)$

Constraint condition: $\underset{i=1}{\overset{M}{\mathrm{\Σ}}}({\mathrm{\α}}_i-{\mathrm{\α}}_i^{*})=0$

0≤α _i≤γ

$0\≤{\mathrm{\α}}_i^{*}\≤\mathrm{\γ}$

Find the solution and can treat estimation function f (x):

$f\left(x\right)=\underset{i=1}{\overset{M}{\mathrm{\Σ}}}({\mathrm{\α}}_i^{*}-{\mathrm{\α}}_i)K(x,x_i)---\left(5\right)$

Wherein, M is the number of support vector, α _iAnd α _jBe Lagrange multiplier, i=1 ..., N, j=1 ..., N; α _i ^*And α _j ^*Be support vector, i=1 ..., M, j=1 ..., M; x _iAnd x _jBe the column vector of input matrix X, i=1 ..., N, j=1 ..., N; K (x, x _i)=exp (|| x-x _i||/θ ²) be the kernel function of support vector machine, θ is a nuclear parameter, ε is insensitive coefficient, y _iBe the column vector of output variable Y, γ is a penalty coefficient;

The particle cluster algorithm module is used to find the solution following minimum problem: its formula is (6):

$\underset{\mathrm{MX}}{\min }f\left(\frac{\mathrm{KX}-\stackrel{\‾}{\mathrm{TX}}}{{\mathrm{\σ}}_x}\right)---\left(6\right)$

Wherein MX is a performance variable, is the part of KX;

Described population derivation algorithm flow process is as follows:

3.1) an initialization n random particles, each particle is expressed as MX _k ⁰, the velometer of each particle correspondence is shown v _k ⁰, order $p_k^0={\mathrm{MX}}_k^0,$ Wherein, n is a population size, p _kRepresent the historical optimal value that each particle of k oneself searches, subscript 0 expression initial value, subscript k=1 ..., n makes iteration step number K=0;

3.2) with each particle MX _k ^KThe substitution function

Calculate its fitness, and with itself and p _k ^KFitness compare, the particle that fitness is less is as new p _k ^K, relatively current iteration goes on foot the fitness of all n particle, and the particle of fitness minimum is designated as pg ^K

3.3) upgrade each particle's velocity and position according to following formula:

$v_k^{K+1}=\mathrm{\ω}{v}_k^K+c_1\mathrm{\ξ}(p_k^K-{\mathrm{MX}}_k^K)+c_2\mathrm{\η}({\mathrm{pg}}^K-{\mathrm{MX}}_k^K)---\left(7\right)$

${\mathrm{MX}}_k^{K+1}={\mathrm{MX}}_k^K+\mathrm{\γ}{v}_k^{K+1}---\left(8\right)$

Wherein, ω is an inertia weight, c ₁Be cognitive coefficient, c ₂Be coefficient of association, ξ and η are [0,1] interval interior equally distributed random numbers, and γ is a constraint factor;

3.4) judge whether to satisfy end condition, if reach maximum iteration time or pg ^KCorresponding fitness is less than predetermined threshold, termination of iterations then, otherwise, make K=K+1, return step 3.2) continue iteration;

Pg when iteration stops ^KBe the performance variable value that makes the chemical oxygen consumption (COC) minimum.

As preferred a kind of scheme: described host computer also comprises: the discrimination model update module, be used for by the sampling time interval of setting, collection site intelligence instrument signal, the actual measurement chemical oxygen consumption (COC) and function calculated value that obtains is compared, if relative error is greater than 10%, then new data is added the training sample data, renewal function f (x).

Further, described host computer also comprises: display module as a result, the optimum performance variable that is used for calculating is passed to the DCS system, and at the control station procedure for displaying state of DCS, by DCS system and fieldbus process status information is delivered to operator station simultaneously and shows.

Further again, described key variables comprise the waste liquid flow that enters incinerator, enter the air mass flow of incinerator and enter the fuel flow rate of incinerator.

The minimizing chemical oxygen demand (COD) discharge method that a kind of described minimizing chemical oxygen demand (COD) discharge of pesticide production waste liquid incinerator system realizes, described minimizing chemical oxygen demand (COD) discharge method may further comprise the steps:

1) the training sample TX of acquisition system key variables from database, the chemical oxygen consumption (COC) data Y of training sample TX correspondence is carried out standardization to training sample TX, obtains input matrix X, adopts following process to finish:

1.1) computation of mean values: $\stackrel{\‾}{\mathrm{TX}}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{TX}}_i---\left(1\right)$

1.2) basis of calculation is poor: ${\mathrm{\σ}}_x^2=\frac{1}{N-1}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}({\mathrm{TX}}_i-\stackrel{\‾}{\mathrm{TX}}),---\left(2\right)$

1.3) standardization: $X=\frac{\mathrm{TX}-\stackrel{\‾}{\mathrm{TX}}}{{\mathrm{\σ}}_x},---\left(3\right)$

Wherein, TX is a training sample, and N is the data sample number, and TX is the average of data sample, σ _xStandard deviation for data sample;

2) with the X, the following quadratic programming problem of Y substitution that obtain:

$\underset{\mathrm{\α},{\mathrm{\α}}^{*}}{\max }\{-\frac{1}{2}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{M}{\mathrm{\Σ}}}({\mathrm{\α}}_i-{\mathrm{\α}}_i^{*})({\mathrm{\α}}_j-{\mathrm{\α}}_j^{*})K(x_i,x_j)-\mathrm{\ϵ}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}({\mathrm{\α}}_i+{\mathrm{\α}}_i^{*})+\underset{i=1}{\overset{M}{\mathrm{\Σ}}}{y}_i({\mathrm{\α}}_i-{\mathrm{\α}}_i^{*})\}---\left(4\right)$

Constraint condition: $\underset{i=1}{\overset{M}{\mathrm{\Σ}}}({\mathrm{\α}}_i-{\mathrm{\α}}_i^{*})=0$

0≤α _i≤γ

$0\≤{\mathrm{\α}}_i^{*}\≤\mathrm{\γ}$

Find the solution and can treat estimation function f (x):

$f\left(x\right)=\underset{i=1}{\overset{M}{\mathrm{\Σ}}}({\mathrm{\α}}_i^{*}-{\mathrm{\α}}_i)K(x,x_i)---\left(5\right)$

Wherein, M is the number of support vector, α _iAnd α _jBe Lagrange multiplier, i=1 ..., N, j=1 ..., N; α _i ^*(and α _j ^*(be support vector, i=1 ..., M, j=1 ..., M; x _iAnd x _jBe the column vector of input matrix X, i=1 ..., N, j=1 ..., N; K (x, x _i)=exp (|| x-x _i||/θ ²) be the kernel function of support vector machine, θ is a nuclear parameter, ε is insensitive coefficient, y _iBe the column vector of Y, γ is a penalty coefficient;

3) by the sampling time interval of setting, collection site intelligence instrument signal obtains real-time input matrix KX, finds the solution between following minimum:

$\underset{\mathrm{MX}}{\min }f\left(\frac{\mathrm{KX}-\stackrel{\‾}{\mathrm{TX}}}{{\mathrm{\σ}}_x}\right)---\left(6\right)$

Wherein MX is a performance variable, is the part of KX;

Described population derivation algorithm flow process is as follows:

3.1) an initialization n random particles, each particle is expressed as MX _k ⁰, the velometer of each particle correspondence is shown v _k ⁰, order $p_k^0={\mathrm{MX}}_k^0,$ Wherein, n is a population size, p _kRepresent the historical optimal value that each particle of k oneself searches, subscript 0 expression initial value, subscript k=1 ..., n makes iteration step number K=0;

3.2) with each particle MX _k ^KThe substitution function

Calculate its fitness, and with itself and p _k ^KFitness compare, the particle that fitness is less is as new p _k ^K, relatively current iteration goes on foot the fitness of all n particle, and the particle of fitness minimum is designated as pg ^K

3.3) upgrade each particle's velocity and position according to following formula:

$v_k^{K+1}=\mathrm{\ω}{v}_k^K+c_1\mathrm{\ξ}(p_k^K-{\mathrm{MX}}_k^K)+c_2\mathrm{\η}({\mathrm{pg}}^K-{\mathrm{MX}}_k^K)$

${\mathrm{MX}}_k^{K+1}={\mathrm{MX}}_k^K+\mathrm{\γ}{v}_k^{K+1}$

Wherein, ω is an inertia weight, c ₁Be cognitive coefficient, c ₂Be coefficient of association, ξ and η are [0,1] interval interior equally distributed random numbers, and γ is a constraint factor;

3.4) judge whether to satisfy end condition, if reach maximum iteration time or pg ^KCorresponding fitness is less than predetermined threshold, termination of iterations then, otherwise, make K=K+1, return 3.2) continue iteration;

Pg when iteration stops ^KBe the performance variable value that makes the chemical oxygen consumption (COC) minimum.

As preferred a kind of scheme: described method also comprises: 4) by the sampling time interval of setting, collection site intelligence instrument signal, the actual measurement chemical oxygen consumption (COC) and function calculated value that obtains is compared, if relative error is greater than 10%, then new data is added the training sample data, renewal function f (x).

Further, in described step 3), the optimum performance variable that calculates is passed to the DCS system, and, by DCS system and fieldbus process status information is delivered to operator station simultaneously and shows at the control station procedure for displaying state of DCS.

Further again, described key variables comprise the waste liquid flow that enters incinerator, and the air mass flow that enters incinerator enters the fuel flow rate of incinerator.

Technical conceive of the present invention is: the present invention is directed to burning process COD can't on-line measurement, sets up incinerator COD soft-sensing model.Use particle cluster algorithm solving-optimizing problem again, calculate and to make the performance variable value of COD minimum.

Beneficial effect of the present invention mainly shows: 1, set up the soft-sensing model of COD, and can on-line prediction COD value; 2, performance variable is carried out optimizing, make the COD minimum.

Description of drawings

Fig. 1 is the hardware structure diagram of system proposed by the invention;

Fig. 2 is the functional structure chart of host computer proposed by the invention;

Fig. 3 is that embodiment uses system and method for the present invention to optimize COD enforcement front and back comparison diagram.

Embodiment

Below in conjunction with accompanying drawing the present invention is further described.The embodiment of the invention is used for the present invention that explains, rather than limits the invention, and in the protection domain of spirit of the present invention and claim, any modification and change to the present invention makes all fall into protection scope of the present invention.

Embodiment 1

With reference to Fig. 1, Fig. 2, a kind of minimizing chemical oxygen demand (COD) discharge of pesticide production waste liquid incinerator system, comprise the field intelligent instrument 2, DCS system and the host computer 6 that are connected with incinerator 1, described DCS system comprises data-interface 3, control station 4 and database 5, described field intelligent instrument 2 is connected with data-interface 3, described data-interface is connected with control station 4, database 5 and host computer 6, and described host computer 6 comprises:

Standardization module 7, be used for from the data sample TX of database acquisition system key variables, the chemical oxygen consumption (COC) data Y of data sample TX correspondence, TX carries out standardization to the data sample, the average of each variable is 0, variance is 1, obtains input matrix X, adopts following process to finish: its formula is (1), (2), (3):

1.1) computation of mean values: $\stackrel{\‾}{\mathrm{TX}}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{TX}}_i---\left(1\right)$

1.2) basis of calculation is poor: ${\mathrm{\σ}}_x^2=\frac{1}{N-1}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}({\mathrm{TX}}_i-\stackrel{\‾}{\mathrm{TX}}),---\left(2\right)$

1.3) standardization: $X=\frac{\mathrm{TX}-\stackrel{\‾}{\mathrm{TX}}}{{\mathrm{\σ}}_x},---\left(3\right)$

Wherein, TX is a training sample, and N is a number of training, and TX is the average of training sample, σ _xStandard deviation for data sample;

Support vector machine (SVM) module 8 is used for soft sensor modeling, and its detailed process is as follows: its formula is: (4), (5):

$\underset{\mathrm{\α},{\mathrm{\α}}^{*}}{\max }\{-\frac{1}{2}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{M}{\mathrm{\Σ}}}({\mathrm{\α}}_i-{\mathrm{\α}}_i^{*})({\mathrm{\α}}_j-{\mathrm{\α}}_j^{*})K(x_i,x_j)-\mathrm{\ϵ}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}({\mathrm{\α}}_i+{\mathrm{\α}}_i^{*})+\underset{i=1}{\overset{M}{\mathrm{\Σ}}}{y}_i({\mathrm{\α}}_i-{\mathrm{\α}}_i^{*})\}---\left(4\right)$

Constraint condition: $\underset{i=1}{\overset{M}{\mathrm{\Σ}}}({\mathrm{\α}}_i-{\mathrm{\α}}_i^{*})=0$

0≤α _i≤γ

$0\≤{\mathrm{\α}}_i^{*}\≤\mathrm{\γ}$

Find the solution and can treat estimation function f (x):

$f\left(x\right)=\underset{i=1}{\overset{M}{\mathrm{\Σ}}}({\mathrm{\α}}_i^{*}-{\mathrm{\α}}_i)K(x,x_i)---\left(5\right)$

Wherein, M is the number of support vector, α _iAnd α _jBe Lagrange multiplier, i=1 ..., N, j=1 ..., N; α _i ^*(and α _j ^*(be support vector, i=1 ..., M, j=1 ..., M; x _iAnd x _jBe the column vector of input matrix X, i=1 ..., N, j=1 ..., N; K (x, x _i)=exp (|| x-x _i||/θ ²) be the kernel function of support vector machine, θ is a nuclear parameter, ε is insensitive coefficient, y _iBe the column vector of Y, γ is a penalty coefficient;

Particle cluster algorithm (PSO) module 9 is used to find the solution following minimum problem: its formula is (6):

$\underset{\mathrm{MX}}{\min }f\left(\frac{\mathrm{KX}-\stackrel{\‾}{\mathrm{TX}}}{{\mathrm{\σ}}_x}\right)---\left(6\right)$

Wherein MX is a performance variable, is the part of KX;

The used population derivation algorithm flow process of present embodiment is as follows:

3.1) an initialization n random particles, each particle is expressed as MX _k ⁰, the velometer of each particle correspondence is shown v _k ⁰, order $p_k^0={\mathrm{MX}}_k^0,$ Wherein, n is a population size, p _kRepresent the historical optimal value that each particle of k oneself searches, subscript 0 expression initial value, subscript k=1 ..., n makes iteration step number K=0;

3.2) with each particle MX _k ^KThe substitution function Calculate its fitness, and with itself and p _k ^KFitness compare, the particle that fitness is less is as new p _k ^K, relatively current iteration goes on foot the fitness of all n particle, and the particle of fitness minimum is designated as pg ^K

3.3) upgrade each particle's velocity and position according to following formula:

$v_k^{K+1}=\mathrm{\ω}{v}_k^K+c_1\mathrm{\ξ}(p_k^K-{\mathrm{MX}}_k^K)+c_2\mathrm{\η}({\mathrm{pg}}^K-{\mathrm{MX}}_k^K)---\left(7\right)$

${\mathrm{MX}}_k^{K+1}={\mathrm{MX}}_k^K+\mathrm{\γ}{v}_k^{K+1}---\left(8\right)$

Wherein, ω is an inertia weight, c ₁Be cognitive coefficient, c ₂Be coefficient of association, ξ and η are [0,1] interval interior equally distributed random numbers, and γ is a constraint factor;

3.4) judge whether to satisfy end condition, if reach maximum iteration time or pg ^KCorresponding fitness is less than predetermined threshold, termination of iterations then, otherwise, make K=K+1, return step 3.2) continue iteration;

Pg when iteration stops ^KBe the performance variable value that makes the COD minimum;

Described host computer 6 also comprises: signal acquisition module 11 is used for the time interval according to each sampling of setting, image data from database;

Described host computer 6 also comprises: discrimination model update module 12, by the sampling time interval of setting, collection site intelligence instrument signal, the actual measurement COD and function calculated value that obtains is compared, if relative error greater than 10%, then adds new data the training sample data, renewal function f (x).

Described system also comprises the DCS system, and described DCS system is made of data-interface 3, control station 4, database 5; Intelligence instrument 2, DCS system, host computer 6 link to each other successively by fieldbus; Host computer 6 also comprises display module 10 as a result, is used for the calculating optimal result is passed to the DCS system, and at the control station procedure for displaying state of DCS, by DCS system and fieldbus process status information is delivered to operator station simultaneously and shows.

The hardware components of described host computer 6 comprises: the I/O element is used for the collection of data and the transmission of information; Data-carrier store, data sample that storage running is required and operational factor etc.; Program storage, storage realizes the software program of functional module; Arithmetical unit, executive routine, the function of realization appointment; Display module shows the parameter and the operation result that are provided with.

When the liquid waste incineration process has been furnished with the DCS system, the real-time and historical data base of the detection of sample real-time dynamic data, memory by using DCS system, the computation optimization function is mainly finished on host computer.

When the liquid waste incineration process is not equipped with the DCS system, adopted data memory substitutes the data storage function of the real-time and historical data base of DCS system, and the computation optimization system is manufactured an independently complete SOC (system on a chip) of the DCS system that do not rely on that comprises I/O element, data-carrier store, program storage, arithmetical unit, several big members of display module, whether be equipped with under the situation of DCS regardless of burning process, can both independently use, more be of value to and promoting the use of.

Embodiment 2

With reference to Fig. 1, Fig. 2 and Fig. 3, a kind of minimizing chemical oxygen demand (COD) discharge of pesticide production waste liquid incinerator method, described method may further comprise the steps:

1) the training sample TX of acquisition system key variables from historical data base, chemical oxygen consumption (COC) (COD) data Y of training sample TX correspondence is carried out standardization to training sample TX, obtains input matrix X, adopts following process to finish:

1.1) computation of mean values: $\stackrel{\‾}{\mathrm{TX}}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{TX}}_i---\left(1\right)$

1.2) basis of calculation is poor: ${\mathrm{\σ}}_x^2=\frac{1}{N-1}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}({\mathrm{TX}}_i-\stackrel{\‾}{\mathrm{TX}}),---\left(2\right)$

1.3) standardization: $X=\frac{\mathrm{TX}-\stackrel{\‾}{\mathrm{TX}}}{{\mathrm{\σ}}_x},---\left(3\right)$

Wherein, TX is a training sample, and N is the data sample number, and TX is the average of data sample, σ _xStandard deviation for data sample;

2) with the X, the following quadratic programming problem of Y substitution that obtain:

$\underset{\mathrm{\α},{\mathrm{\α}}^{*}}{\max }\{-\frac{1}{2}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{M}{\mathrm{\Σ}}}({\mathrm{\α}}_i-{\mathrm{\α}}_i^{*})({\mathrm{\α}}_j-{\mathrm{\α}}_j^{*})K(x_i,x_j)-\mathrm{\ϵ}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}({\mathrm{\α}}_i+{\mathrm{\α}}_i^{*})+\underset{i=1}{\overset{M}{\mathrm{\Σ}}}{y}_i({\mathrm{\α}}_i-{\mathrm{\α}}_i^{*})\}---\left(4\right)$

Constraint condition: $\underset{i=1}{\overset{M}{\mathrm{\Σ}}}({\mathrm{\α}}_i-{\mathrm{\α}}_i^{*})=0$

0≤α _i≤γ

$0\≤{\mathrm{\α}}_i^{*}\≤\mathrm{\γ}$

Find the solution and can treat estimation function f (x):

$f\left(x\right)=\underset{i=1}{\overset{M}{\mathrm{\Σ}}}({\mathrm{\α}}_i^{*}-{\mathrm{\α}}_i)K(x,x_i)---\left(5\right)$

Wherein, M is the number of support vector, α _iAnd α _jBe Lagrange multiplier, i=1 ..., N, j=1 ..., N; α _i ^*And α _j ^*Be support vector, i=1 ..., M, j=1 ..., M; x _iAnd x _jBe the column vector of input matrix X, i=1 ..., N, j=1 ..., N; K (x, x _i)=exp (|| x-x _i||/θ ²) be the kernel function of support vector machine, θ is a nuclear parameter, ε is insensitive coefficient, y _iBe the column vector of Y, γ is a penalty coefficient;

3) by the sampling time interval of setting, collection site intelligence instrument signal obtains real-time input matrix KX, finds the solution following minimum problem:

$\underset{\mathrm{MX}}{\min }f\left(\frac{\mathrm{KX}-\stackrel{\‾}{\mathrm{TX}}}{{\mathrm{\σ}}_x}\right)---\left(6\right)$

Wherein MX is a performance variable, is the part of KX;

The used population derivation algorithm flow process of the present invention is as follows:

3.1) an initialization n random particles, each particle is expressed as MX _k ⁰, the velometer of each particle correspondence is shown v _k ⁰, order $p_k^0={\mathrm{MX}}_k^0,$ Wherein, n is a population size, p _kRepresent the historical optimal value that each particle of k oneself searches, subscript 0 expression initial value, subscript k=1 ..., n makes iteration step number K=0;

3.2) with each particle MX _k ^KThe substitution function

Calculate its fitness, and with itself and p _k ^KFitness compare, the particle that fitness is less is as new p _k ^K, relatively current iteration goes on foot the fitness of all n particle, and the particle of fitness minimum is designated as pg ^K

3.3) upgrade each particle's velocity and position according to following formula:

$v_k^{K+1}=\mathrm{\ω}{v}_k^K+c_1\mathrm{\ξ}(p_k^K-{\mathrm{MX}}_k^K)+c_2\mathrm{\η}({\mathrm{pg}}^K-{\mathrm{MX}}_k^K)---\left(7\right)$

${\mathrm{MX}}_k^{K+1}={\mathrm{MX}}_k^K+\mathrm{\γ}{v}_k^{K+1}---\left(8\right)$

Wherein, ω is an inertia weight, c ₁Be cognitive coefficient, c ₂Be coefficient of association, ξ and η are [0,1] interval interior equally distributed random numbers, and γ is a constraint factor;

3.4) judge whether to satisfy end condition, if reach maximum iteration time or pg ^KCorresponding fitness is less than predetermined threshold, termination of iterations then, otherwise, make K=K+1, return step 3.2) continue iteration;

Pg when iteration stops ^KBe the performance variable value that makes the COD minimum;

Described method also comprises: 4), by the sampling time interval of setting, and collection site intelligence instrument signal, with the actual measurement COD and function calculated value that obtains relatively, if relative error greater than 10%, then adds new data the training sample data, renewal function f (x).

In described step 3), the optimum performance variable that calculates is passed to the DCS system, and, by DCS system and fieldbus process status information is delivered to operator station simultaneously and shows at the control station procedure for displaying state of DCS.

Described key variables comprise the waste liquid flow that enters incinerator, and the air mass flow that enters incinerator enters the fuel flow rate of incinerator.

Described data storage device 5 is the historical data base of DCS system, and described DCS system is made of data-interface 3, control station 4 and historical data base 5, and intelligence instrument 2, DCS system, soft measurement intelligent processor 6 link to each other successively by fieldbus; In described step 3), calculate optimum operating value, the result is passed to the DCS system, show, and be delivered to operator station by DCS system and fieldbus and show at the control station of DCS.

Fig. 3 is that embodiment uses comparison diagram before and after the system and method computation optimization COD of the present invention, as can be seen from the figure, optimize COD value after implementing be significantly less than optimize enforcement before, reach emission request substantially, it is remarkable that hazardous emission minimizes effect.

Claims (6)

1. a minimizing chemical oxygen demand (COD) discharge of pesticide production waste liquid incinerator system comprises the field intelligent instrument, DCS system and the host computer that are connected with pesticide production waste liquid incinerator, and described DCS system comprises control station and database; Described field intelligent instrument is connected with the DCS system, and described DCS system is connected with host computer, and it is characterized in that: described host computer comprises:

The standardization module, be used for from the training sample TX of database acquisition system key variables, described key variables comprise the waste liquid flow that enters incinerator, enter the air mass flow of incinerator and enter the fuel flow rate of incinerator, the chemical oxygen consumption (COC) data Y of training sample TX correspondence, training sample TX is carried out standardization, and the average of each variable is 0, and variance is 1, obtain input matrix X, adopt following process to finish: its formula is (1), (2), (3):

1.1) computation of mean values: $\stackrel{\‾}{\mathrm{TX}}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{TX}}_i,---\left(1\right)$

1.2) basis of calculation is poor: ${\mathrm{\σ}}_x^2=\frac{1}{N-1}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}({\mathrm{TX}}_i-\stackrel{\‾}{\mathrm{TX}}),---\left(2\right)$

1.3) standardization: $X=\frac{\mathrm{TX}-\stackrel{\‾}{\mathrm{TX}}}{{\mathrm{\σ}}_x},---\left(3\right)$

Wherein, TX is a training sample, and N is a number of training,

Be the average of training sample, σ _xStandard deviation for data sample;

The support vector machine module is used for soft sensor modeling, and its detailed process is as follows: its formula is: (4), (5):

$\underset{\mathrm{\α},{\mathrm{\α}}^{*}}{\max }\{-\frac{1}{2}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{M}{\mathrm{\Σ}}}({\mathrm{\α}}_i-{\mathrm{\α}}_i^{*})({\mathrm{\α}}_j-{\mathrm{\α}}_j^{*})K(x_i,x_j)-\mathrm{\ϵ}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}({\mathrm{\α}}_i+{\mathrm{\α}}_i^{*})+\underset{i=1}{\overset{M}{\mathrm{\Σ}}}{y}_i({\mathrm{\α}}_i-{\mathrm{\α}}_i^{*})\}---\left(4\right)$

Constraint condition: $\underset{i=1}{\overset{M}{\mathrm{\Σ}}}({\mathrm{\α}}_i-{\mathrm{\α}}_i^{*})=0$

0≤α _i≤γ

$0\≤{\mathrm{\α}}_i^{*}\≤\mathrm{\γ}$

Find the solution and can treat estimation function f (x):

$f\left(x\right)=\underset{i=1}{\overset{M}{\mathrm{\Σ}}}({\mathrm{\α}}_i^{*}-{\mathrm{\α}}_i)K(x,x_i)---\left(5\right)$

Wherein, M is the number of support vector, α _iAnd α _jBe Lagrange multiplier, i=1 ..., N, j=1 ..., N;

With

Be support vector, i=1 ..., M, j=1 ..., M; x _iAnd x _jBe the column vector of input matrix X, i=1 ..., N, j=1 ..., N; K (x, x _i)=exp (|| x-x _i||/θ ²) be the kernel function of support vector machine, θ is a nuclear parameter, ε is insensitive coefficient, y _iBe the column vector of output variable Y, γ is a penalty coefficient;

The particle cluster algorithm module is used to find the solution following minimum problem: its formula is (6):

$\underset{\mathrm{MX}}{\min }f\left(\frac{\mathrm{KX}-\stackrel{\‾}{\mathrm{TX}}}{{\mathrm{\σ}}_x}\right)---\left(6\right)$

Wherein MX is a performance variable, is the part of KX;

Described particle cluster algorithm flow process is as follows:

3.1) an initialization n random particles, each particle is expressed as

The velometer of each particle correspondence is shown

Order

Wherein, n is a population size, p _kRepresent the historical optimal value that each particle of k oneself searches, subscript 0 expression initial value, subscript k=1 ..., n makes iteration step number K=0;

3.2) with each particle The substitution function

Calculate its fitness, and with its with Fitness compare, the particle that fitness is less is as new

Relatively current iteration goes on foot the fitness of all n particle, and the particle of fitness minimum is designated as pg ^K

3.3) upgrade each particle's velocity and position according to following formula:

$v_k^{K+1}=\mathrm{\ω}{v}_k^K+c_1\mathrm{\ξ}(p_k^K-M{X}_k^K)+c_2\mathrm{\η}(p{g}^K-M{X}_k^K)---\left(7\right)$

${\mathrm{MX}}_k^{K+1}={\mathrm{MX}}_k^K+\mathrm{\γ}{v}_k^{K+1}---\left(8\right)$

Wherein, ω is an inertia weight, c ₁Be cognitive coefficient, c ₂Be coefficient of association, ξ and η are [0,1] interval interior equally distributed random numbers, and γ is a constraint factor;

3.4) judge whether to satisfy end condition, if reach maximum iteration time or pg ^KCorresponding fitness is less than predetermined threshold, termination of iterations then, otherwise, make K=K+1, return step 3.2) continue iteration;

Pg when iteration stops ^KBe the performance variable value that makes the chemical oxygen consumption (COC) minimum.

2. minimizing chemical oxygen demand (COD) discharge of pesticide production waste liquid incinerator as claimed in claim 1 system, it is characterized in that: described host computer also comprises:

The discrimination model update module is used for by the sampling time interval of setting, collection site intelligence instrument signal, the actual measurement chemical oxygen consumption (COC) and function calculated value that obtains is compared, if relative error greater than 10%, then adds new data the training sample data, renewal function f (x).

3. minimizing chemical oxygen demand (COD) discharge of pesticide production waste liquid incinerator as claimed in claim 1 or 2 system, it is characterized in that: described host computer also comprises:

Display module is used for the optimum performance variable that calculates is passed to the DCS system as a result, and at the control station procedure for displaying state of DCS, by DCS system and fieldbus process status information is delivered to operator station simultaneously and shows.

4. minimizing chemical oxygen demand (COD) discharge method that minimizing chemical oxygen demand (COD) discharge of pesticide production waste liquid incinerator as claimed in claim 1 system realizes, it is characterized in that: described minimizing chemical oxygen demand (COD) discharge method may further comprise the steps:

1) the training sample TX of acquisition system key variables from database, described key variables comprise the waste liquid flow that enters incinerator, enter the air mass flow of incinerator and enter the fuel flow rate of incinerator, the chemical oxygen consumption (COC) data Y of training sample TX correspondence, training sample TX is carried out standardization, obtain input matrix X, adopt following process to finish:

1.1) computation of mean values: $\stackrel{\‾}{\mathrm{TX}}=\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{TX}}_i,---\left(1\right)$

1.2) basis of calculation is poor: ${\mathrm{\σ}}_x^2=\frac{1}{N-1}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}({\mathrm{TX}}_i-\stackrel{\‾}{\mathrm{TX}}),---\left(2\right)$

1.3) standardization: $X=\frac{\mathrm{TX}-\stackrel{\‾}{\mathrm{TX}}}{{\mathrm{\σ}}_x},---\left(3\right)$

Wherein, TX is a training sample, and N is the data sample number,

Be the average of data sample, σ _xStandard deviation for data sample;

2) with the X, the following quadratic programming problem of Y substitution that obtain:

$\underset{\mathrm{\α},{\mathrm{\α}}^{*}}{\max }\{-\frac{1}{2}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{M}{\mathrm{\Σ}}}({\mathrm{\α}}_i-{\mathrm{\α}}_i^{*})({\mathrm{\α}}_j-{\mathrm{\α}}_j^{*})K(x_i,x_j)-\mathrm{\ϵ}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}({\mathrm{\α}}_i+{\mathrm{\α}}_i^{*})+\underset{i=1}{\overset{M}{\mathrm{\Σ}}}{y}_i({\mathrm{\α}}_i-{\mathrm{\α}}_i^{*})\}---\left(4\right)$

Constraint condition: $\underset{i=1}{\overset{M}{\mathrm{\Σ}}}({\mathrm{\α}}_i-{\mathrm{\α}}_i^{*})=0$

0≤α _i≤γ

$0\≤{\mathrm{\α}}_i^{*}\≤\mathrm{\γ}$

Find the solution and can treat estimation function f (x):

$f\left(x\right)=\underset{i=1}{\overset{M}{\mathrm{\Σ}}}({\mathrm{\α}}_i^{*}-{\mathrm{\α}}_i)K(x,x_i)---\left(5\right)$

Wherein, M is the number of support vector, α _iAnd α _jBe Lagrange multiplier, i=1 ..., N, j=1 ..., N;

With

Be support vector, i=1 ..., M, j=1 ..., M; x _iAnd x _jBe the column vector of input matrix X, i=1 ..., N, j=1 ..., N; K (x, x _i)=exp (|| x-x _i||/θ ²) be the kernel function of support vector machine, θ is a nuclear parameter, ε is insensitive coefficient, y _iBe the column vector of Y, γ is a penalty coefficient;

3) by the sampling time interval of setting, collection site intelligence instrument signal obtains real-time input matrix KX, finds the solution following minimum problem:

$\underset{\mathrm{MX}}{\min }f\left(\frac{\mathrm{KX}-\stackrel{\‾}{\mathrm{TX}}}{{\mathrm{\σ}}_x}\right)---\left(6\right)$

Wherein MX is a performance variable, is the part of KX;

Described particle cluster algorithm flow process is as follows:

3.1) an initialization n random particles, each particle is expressed as

The velometer of each particle correspondence is shown

Order

Wherein, n is a population size, p _kRepresent the historical optimal value that each particle of k oneself searches, subscript 0 expression initial value, subscript k=1 ..., n makes iteration step number K=0;

3.2) with each particle

The substitution function

Calculate its fitness, and with its with Fitness compare, the particle that fitness is less is as new

Relatively current iteration goes on foot the fitness of all n particle, and the particle of fitness minimum is designated as pg ^K

3.3) upgrade each particle's velocity and position according to following formula:

$v_k^{K+1}=\mathrm{\ω}{v}_k^K+c_1\mathrm{\ξ}(p_k^K-M{X}_k^K)+c_2\mathrm{\η}(p{g}^K-M{X}_k^K)$

${\mathrm{MX}}_k^{K+1}={\mathrm{MX}}_k^K+\mathrm{\γ}{v}_k^{K+1}$

Wherein, ω is an inertia weight, c ₁Be cognitive coefficient, c ₂Be coefficient of association, ξ and η are [0,1] interval interior equally distributed random numbers, and γ is a constraint factor;

3.4) judge whether to satisfy end condition, if reach maximum iteration time or pg ^KCorresponding fitness is less than predetermined threshold, termination of iterations then, otherwise, make K=K+1, return step 3.2) continue iteration;

Pg when iteration stops ^KBe the performance variable value that makes the chemical oxygen consumption (COC) minimum.

5. minimizing chemical oxygen demand (COD) discharge method as claimed in claim 4, it is characterized in that: described method also comprises: 4) by the sampling time interval of setting, collection site intelligence instrument signal, the actual measurement chemical oxygen consumption (COC) and function calculated value that obtains is compared, if relative error is greater than 10%, then new data is added the training sample data, renewal function f (x).

6. as claim 4 or 5 described minimizing chemical oxygen demand (COD) discharge of pesticide production waste liquid incinerator methods, it is characterized in that: in described step 3), the optimum performance variable that calculates is passed to the DCS system, and, by DCS system and fieldbus process status information is delivered to operator station simultaneously and shows at the control station procedure for displaying state of DCS.

Images