CN1912072A - Non-linear continuous tank type diesel oil blending method - Google Patents

Abstract

The invention relates to a nonlinear continuous cylinder-type diesel oil mixing method, comprising: (1) establishing a nonlinear mathematical model based on real-time flows of all components of finished diesel oil and linear and nonlinear relations between the flows and linear and nonlinear relations in real-time quality attributes of all the components and the finished diesel oil; (2) inputting crude oil refining rate, real-time flow data of each siding and real-time quality data of each component into the established nonlinear mathematical model in step (1), and calculating real-time flow data of each component; (3) mixing the finished diesel oil according to the obtained real-time flow data. And it can remarkably yield of diesel oil products and benefits of refineries.

Description

Non-linear continuous tank type diesel oil blending method

Technical field

The present invention relates to a kind of diesel oil blending method, particularly a kind of non-linear continuous tank type diesel oil blending method.

Background technology

In recent years, along with the strictness day by day of environmental requirement, the product standard of diesel oil is also more and more higher.Yet, correspondingly with it be that the crude oil of refinery becomes more diverse, heaviness, poor qualityization, high sulfuration, high acidifying.Refinery is in actual production at present, and mouthful flow that distillates in real time that main employing will be produced each device of diesel oil blending component enters final tempering tank continuously, by pump circulation pot type blending method, serves as according to the blending prescription of determining the diesel oil various product with the scheduling experience.With the production that this class prescription carries out, under the new standard check, product percent of pass only is 60%～70%, and readjustment, mass excess, degradation sale all have generation, and the loss of refinery economic benefit is serious.And pipeline on-line blending technology because on-line analysis instrument price is very expensive, is seldom used in the diesel oil blending production process at present.

How to utilize existing real-time composition flow rate data, existing manual analysis quality index and up-to-date blending attribute linearity and nonlinear equation, concoct in the production process in the every day of diesel oil, calculating optimum blending prescription in real time, is the gordian technique that improves the diesel product qualification rate and improve the refinery benefit.

Summary of the invention

The object of the invention is, a kind of non-linear continuous tank type diesel oil blending method is provided, and improves diesel product qualification rate and refinery benefit with this.

The said non-linear continuous tank type diesel oil blending method of the present invention, it comprises the steps:

(1) sets up based on the linearity in the real-time qualitative attribute of the linearity between formation each component real-time traffic of finished diesel fuel and each flow and nonlinear relationship, each component and finished diesel fuel and the nonlinear mathematical model of nonlinear relationship;

(2) crude oil refinery speed, each side line real-time traffic data and the real-time qualitative data input step of each component (1) are set up in the nonlinear mathematical model, computing (finding the solution calculating as being optimized in Lingo 8.0 softwares) obtains constituting each component real-time traffic data of finished diesel fuel;

(3) carry out the blending of finished diesel fuel by each the component real-time traffic data of formation finished diesel fuel that obtain by step (2);

Wherein: said nonlinear mathematical model is as follows:

$\begin{array}{cccc}\mathrm{Max}& \underset{j=1}{\overset{n}{\mathrm{\Σ}}}{c}_{j}Q{P}_j-\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{c}_i{Q}_{i,j}& \mathrm{or}& \underset{j=1}{\overset{n}{\mathrm{\Σ}}}Q{P}_j---\left(1\right)\end{array}$

$s.t.:{\mathrm{QF}}_u=\underset{u^{\′}{\∈U}_u}{\mathrm{\Σ}}\underset{s{\∈S}_{u^{\′},u}}{\mathrm{\Σ}}{Q}_{u^{\′},s,u}---\left(2\right)$

QF _u＜QF _u-max (3)

Q _u，s＝d _u，s·QF _u，s∈S _u (4)

$Q_{u,s}=\underset{u^{\′}\∈U_{s,u}}{\mathrm{\Σ}}{Q}_{u,s,u^{\′}}---\left(5\right)$

$\frac{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{a}_{i,k}{Q}_{i,j}}{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{Q}_{i,j}}=q_{j,k},j=\mathrm{1,2},\·\·\·,n;k=\mathrm{1,2},\·\·\·,l_1---\left(6\right)$

$q_{j,l_1+1}=[\ln \left(\underset{i=1}{\overset{m}{\mathrm{\Σ}}}(0.929^{a_{i,l_1+1}}\·Q_{i,j})\right)-\ln {\mathrm{QP}}_j]/\ln 0.929,j=\mathrm{1,2},...,n---\left(7\right)$

$q_{j,l_1+2}=\exp \left[\underset{i=1}{\overset{m}{\mathrm{\Σ}}}(\frac{Q_{i,j}}{Q{P}_j}\·l{\mathrm{na}}_{i,l_1+2})\right],j=\mathrm{1,2},...,n---\left(8\right)$

$q_{j,l_1+3}=9.4656T^3-57.08217T^2+129.075T-99.2741$

$T=f[(\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{Q}_{i,j}\·a_{i,l_1+3})/\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{Q}_{i,j}],j=\mathrm{1,2},...,n---\left(9\right)$

b _min，j，k≤q _j，k≤b _max，j，k j＝1，2，…，n；k＝1，2，…，l (10)

U---can produce the title of the device of diesel component oil;

N---the species number of diesel oil the finished product, as: 0# car bavin, the light bavin of 0# ,-10# diesel oil etc.;

M---participate in the quantity of the blending component of diesel oil finished product blending;

L---in diesel oil processed oil blending, need the main performance index number considered;

S _{U ', u}---the set of the oil stream of separating device u ' access to plant u;

S _u---the set of all products that device u produces;

U _{S, u}---receive the set of all device u of oil stream s;

U _u---there is product oil to flow directly into the set of all preceding continuous devices of device u;

QF _u---the feeding total flux of device u;

QF _U-max---the maximum processing capability of device u, i.e. maximum inlet total flux (ton/sky);

Q _{U, s}---the flow of certain product s that device u produces;

d _{U, s}---the output capacity (%) of certain product s that device u produces;

Q _{U, s, u '}---the flow of certain product s access to plant u ' that device u produces;

Q _{I, j}---finished product j (j=1,2 ..., n) in blending component i (i=1,2 ..., flow m);

QP _j---finished product j (j=1,2 ..., flow n);

c _j---diesel oil processed oil j (j=1,2 ..., the unit price that dispatches from the factory n);

c _i---blending component i (i=1,2 ..., m) advance factory's unit price;

a _{I, k}---blending component i (i=1,2 ..., a kind of performance index k m) (k=1,2 ..., l ₁, l ₁+ 1 ..., l), wherein preceding l ₁Individual is linear restriction, from l ₁+ 1 to l (=l ₁+ 3) individual is non-linear constrain;

q _{J, k}---finished product j (j=1,2 ..., a kind of performance index k n) (k=1,2 ..., l);

b _{Max, j, k}, b _{Min, j, k}---finished product j (j=1,2 ..., quality index k n) (k=1,2 ..., the upper and lower limit of binding occurrence l).

Formula (1) representation unit time blending gross earnings or blending ultimate production; The feeding flow of formula (2) indication device u is the summation of flow of product of the access to plant u of the middle production of all preorder device u '; The import total flux of formula (3) indication device u should be less than the maximum processing capability of device itself; The flow of certain product s of formula (4) indication device u output multiply by the output capacity (also can adopt and distillate a mouthful data on flows in real time) of corresponding product line for the feeding flow of this device; The flow of certain product s that formula (5) indication device u produces equals to flow into each branch flow summation of each device (comprise not using up and enter the storage tank part) of postorder; Formula (6) is the l of the processed oil j after concocting ₁Individual linear properties refers to target value; Formula (7) is the l of processed oil j ₁+ 1 (flash-point) non-linear index; Formula (8) is the l of processed oil j ₁The non-linear index of+2 (kinematic viscosity); Formula (9) is the l of processed oil j ₁The non-linear index of+3 (condensation points), T is by blending condensation point and the linear condensation point value that calculates of blending component condensation point in the formula, draw the corresponding with it conversion factor of tabling look-up (data are referring to Hou Xiang unicorn chief editor " Chinese oil Refining Technologies " (three), 554～556 pages in the table) and drawing after this value by linear interpolation method; (10) value of the performance index k of the processed oil j after the expression blending must satisfy the constraint that corresponding processed oil quality index requires; Except that condensation point, the non-negative DEFAULT constraint that is of all the other all parameters and variable.

The present invention is taking all factors into consideration the real-time traffic data of producing each device of diesel oil blending component, in real time on the basis of qualitative data, can calculate the real-time concoction optimization prescription of diesel oil that satisfies quality test index and mass balance requirement fast, thereby provide real-time diesel oil blending ratio, for the control of diesel product quality provides favourable foundation; Because adopt the mathematical programming model mode to carry out modeling, versatility is very strong, has improved the scope of application of model greatly.Adopt mathematical programming popular software Lingo to find the solution, fast operation can carry out by real-time online calculating.It is diversified day by day that this refines the crude oil kind at present refinery, even under the also very big condition of production of the crude oil ingredient daily variation in an oil field, for the optimization production of diesel product provides sound assurance.

Description of drawings

Fig. 1 is diesel oil continuous tank type blending typical production schema

Wherein: CDU1-atmospheric and vacuum distillation unit 1; CDU2-atmospheric and vacuum distillation unit 2; HT1-gasoline and diesel hydrogenation device 1; HT2-gasoline and diesel hydrogenation device 2; LHT-lube oil hydrogenation device; Do not use up remaining ingredient before the DP-; DP1-diesel oil finished product tempering tank 1; DP2-diesel oil finished product tempering tank 2.

QF _CDU1The feeding total flux of-device CDU1;

QF _CDU2The feeding total flux of-device CDU2;

QF _HT1The feeding total flux of-device HT1;

QF _HT2The feeding total flux of-device HT2;

QF _LHTThe feeding total flux of-device LHT;

QF _DP1The feeding total flux of-diesel oil the finished product tempering tank DP1;

QF _DP2The feeding total flux of-diesel oil the finished product tempering tank DP2;

Q _F1, Q _F2, Q _F3-by the flow of other three kinds of oil products of proportioning access to plant HT1;

Q _F4, Q _F5, Q _F6-by the flow of other three kinds of oil products of proportioning access to plant HT2;

Q _{CDU1, s1}, Q _{CDU2, s1}The normal separately line rate of discharge of-device CDU1, CDU2;

Q _{CDU1, s2}, Q _{CDU2, s2}The normal separately two wires rate of discharge of-device CDU1, CDU2;

Q _{CDU1, s3}, Q _{CDU2, s3}The rate of discharge of atmosphere 3rd side cut separately of-device CDU1, CDU2;

Q _{CDU1, s4}, Q _{CDU2, s4}The normal separately four line rates of discharge of-device CDU1, CDU2;

Q _{CDU1, s5}, Q _{CDU2, s5}-device CDU1, CDU2 subtract the top line rate of discharge separately;

Q _{CDU1, s6}, Q _{CDU2, s6}The rate of discharge of vacuum 1st side cut separately of-device CDU1, CDU2;

Q _{HT1, S}, Q _{HT2, S}-device HT1, HT2 be the diesel component side line rate of discharge of output separately;

Q _{LHT, S}The diesel component side line rate of discharge of-device LHT output;

Q _{CDU1, S1, DP1}, Q _{CDU1, S1, DP2}The normal line (S1) of-device CDU1 is the flow of access to plant DP1, DP2 respectively;

Q _{CDU1, S2, DP1}, Q _{CDU1, S2, DP2}The normal two wires (S2) of-device CDU1 is the flow of access to plant DP1, DP2 respectively;

Q _{CDU1, S3, DP1}, Q _{CDU1, S3, DP2}The atmosphere 3rd side cut (S3) of-device CDU1 is the flow of access to plant DP1, DP2 respectively;

Q _{CDU1, S4, HT1}, Q _{CDU1, S4, HT2}Normal four lines (S4) of-device CDU1 are the flow of access to plant HT1, HT2 respectively;

Q _{CDU1, S5, HT1}, Q _{CDU1, S5, HT2}The flow that subtracts top line (S5) difference access to plant HT1, HT2 of-device CDU1;

Q _{CDU1, S6, HT1}, Q _{CDU1, S6, HT2}The vacuum 1st side cut (S6) of-device CDU1 is the flow of access to plant HT1, HT2 respectively;

Q _{HT1, S, DP1}, Q _{HT1, S, DP2}The diesel component of-device HT1 output is the flow of access to plant DP1, DP2 respectively;

Q _{HT2, S, DP1}, Q _{HT2, S, DP2}The diesel component of-device HT2 output is the flow of access to plant DP1, DP2 respectively;

Q _{LTS, S, DP1}, Q _{LTS, S, DP2}The diesel component of-device LHT output is the flow of access to plant DP1, DP2 respectively;

Q _1,1(=Q _{CDU1, S1, DP1}The blending component 1 of)-diesel oil finished product 1;

Q _2,1(=Q _{CDU1, S2, DP1}The blending component 2 of)-diesel oil finished product 1;

Q _3,1(=Q _{CDU1, S3, DP1}The blending component 3 of)-diesel oil finished product 1;

Q _4,1(=Q _{CDU2, S1, DP1}The blending component 4 of)-diesel oil finished product 1;

Q _5,1(=Q _{CDU2, S2, DP1}The blending component 5 of)-diesel oil finished product 1;

Q _6,1(=Q _{CDU2, S3, DP1}The blending component 6 of)-diesel oil finished product 1;

Q _7,1(=Q _{HT1, S, DP1}The blending component 7 of)-diesel oil finished product 1;

Q _8,1(=Q _{HT2, S, DP1}The blending component 8 of)-diesel oil finished product 1;

Q _9,1(=Q _{LHT, S, DP1}The blending component 9 of)-diesel oil finished product 1;

Q _10,1The blending component 10 of-diesel oil finished product 1 (mixing the component that do not use up before of refining, fixed) by particular case;

Q _11,1(=Q _{CDU2, S6, DP1}The blending component 11 of)-diesel oil finished product 1;

Q _1,2(=Q _{CDU1, S1, DP2}The blending component 1 of)-diesel oil finished product 2;

Q _2,2(=Q _{CDU1, S2, DP2}The blending component 2 of)-diesel oil finished product 2;

Q _3,2(=Q _{CDU1, S3, DP2}The blending component 3 of)-diesel oil finished product 2;

Q _4,2(=Q _{CDU2, S1, DP2}The blending component 4 of)-diesel oil finished product 2;

Q _5,2(=Q _{CDU2, S2, DP2}The blending component 5 of)-diesel oil finished product 2;

Q _6,2(=Q _{CDU2, S3, DP2}The blending component 6 of)-diesel oil finished product 2;

Q _7,2(=Q _{HT1, S, DP2}The blending component 7 of)-diesel oil finished product 2;

Q _8,2(=Q _{HT2, S, DP2}The blending component 8 of)-diesel oil finished product 2;

Q _9,2(=Q _{LHT, S, DP2}The blending component 9 of)-diesel oil finished product 2;

Q _10,2The blending component 10 of-diesel oil finished product 2 (mixing the component that do not use up before of refining, fixed) by practical situation;

Q _11,2(=Q _{CDU2, S6, DP2}The blending component 11 of)-diesel oil finished product 2;

Embodiment

The said non-linear continuous tank type diesel oil blending method of the present invention, it comprises the steps:

(1) sets up based on the linearity in the real-time qualitative attribute of the linearity between formation each component real-time traffic of finished diesel fuel and each flow and nonlinear relationship, each component and finished diesel fuel and the nonlinear mathematical model of nonlinear relationship;

A) selection of variable in the diesel oil nonlinear programming blending model

Because diesel product quality final inspection index has 18, wherein %10 steam total insoluble substance after excess carbon residue, ash content, mechanical impurity, moisture, copper corrosion, colourity, the oxidation, cold filter clogging temperature totally 8 indexs do not have the blending equation, therefore do not add in the calculating variable of concocting model.And density (20 ℃), cetane value, boiling range (50%, 90%, 95%), viscosity (20 ℃), sulphur content, flash-point, acidity, condensation point totally 10 indexs strict linear or non-linear blending equation is arranged, therefore choosing these 10 indexs is the qualitative attribute variable index of setting up diesel oil nonlinear programming blending model; Real-time inlet flow rate and the rate of discharge that can produce each device of diesel component and the finished product are the flow attribution variable index.

B) determine the nonlinear mathematical model that diesel oil blending is filled a prescription

Because in the oil refining production process, import fluid of each device generally all is the mixture of the various fluid that come out of preorder device, and the fluid of each device outlet all is diverted in a plurality of different postorder devices, therefore, there is a mixing tank front that we are located at each device that can produce the diesel oil blending component oil, and handle effusive each strand fluid place, back through each device a separator is arranged all.

In the time of will concocting gross earnings or ultimate production and be target to the maximum, set up the described formula of preamble (1)～formula (10) nonlinear mathematical model.

(2) crude oil refinery speed, each side line real-time traffic data and the real-time qualitative data of each component (can obtain in the computer network in refinery) input step (1) are set up in the nonlinear mathematical model, in Lingo 8.0 softwares, be optimized and find the solution calculating, obtain constituting each component real-time traffic data of finished diesel fuel on the same day;

(3) carry out the blending of finished diesel fuel by each the component real-time traffic data of formation finished diesel fuel that obtain by step (2).

Blending production with certain factory's 0# one day derv fuel oil and the light bavin of 0# is example below, and the invention will be further described, and its purpose only is better to understand content of the present invention, and unrestricted protection scope of the present invention:

The production process example calculation of certain factory's 0# one day derv fuel oil and the light bavin of 0#.

Normal one, normal two, the atmosphere 3rd side cut product of two in example cover atmospheric and vacuum distillation unit (CDU1, CDU2) are participated in final blending directly, and the vacuum 1st side cut portioned product of device CDU2 participates in final blending; The Chang Si of device CDU1 and CDU2, subtract top, vacuum 1st side cut product and other oil product proportioning and enter two and overlap hydrogenation units, the diesel component after the hydrotreatment enters final blending; The diesel component of one cover lube oil hydrogenation device production enters the finished product blending.

Ultimate production with two kinds of product oil of diesel oil is target to the maximum, and with diesel oil end product quality attribute equation, the mass balance equation at mixing tank and separator place is the constraint that must satisfy, is optimized in Lingo 8.0 softwares and finds the solution calculating.

Metrical instrument is not had extra demand, only need existing production configuration to get final product.

Concocting preceding 2 hours, the input data of flow and each constituent mass data are obtained in the computer network in factory, is the input data of formula calculation with these data, calculates in real time and concocts required prescription the same day.

1, online data of getting have:

1) data on flows (T/day)

QF _CDU1＝7367.04；

QF _CDU2＝19843.25；

QF _LHT＝141.23；

Q _CDU1，s1＝667.4538；Q _CDU2，s1＝0；

Q _CDU1，s2＝425.8149；Q _CDU2，s2＝2057.745；

Q _CDU1，s3＝681.4512；Q _CDU2，s3＝1728.347；

Q _CDU1，s4＝215.8543；Q _CDU2，s4＝583.3915；

Q _CDU1，s5＝39.78202；Q _CDU2，s5＝57.54542；

Q _CDU1，s6＝456.7565；Q _CDU2，s6＝1002.084；

Mix the front and do not use up 737.0932 tons of the vacuum 1st side cut residual contentes of device CDU2 one day.

2) qualitative data

2, with the calculation result that obtains after the above-mentioned data substitution mathematical model be:

0# derv fuel oil blending on same day prescription	The light bavin of 0# blending on same day prescription
		Each composition flow rate (ton/day) Q_11 (the normal line of CDU1) 667.4538 Q_21 (the normal two wires of CDU1) 425.8149 Q_31 (CDU1 atmosphere 3rd side cut), 681.4512 Q_41 (the normal line of CDU2) 0.000000 Q_51 (the normal two wires of CDU2) 0.000000 Q_61 (CDU2 atmosphere 3rd side cut), 0.000000 Q_71 (HT1 hydrogenation), 0.000000 Q_81 (HT2 hydrogenation), 1816.885 Q_91 (LHT hydrogenation) 0.000000 Q_101 (mixing the refining residue) 39.92103 Q_111 (CDU2 vacuum 1st side cut on the same day) 0.000000	Each composition flow rate (ton/day) Q_12 (the normal line of CDU1) 0.000000 Q_22 (the normal two wires of CDU1) 0.000000 Q_32 (CDU1 atmosphere 3rd side cut), 0.000000 Q_42 (the normal line of CDU2) 0.000000 Q_52 (the normal two wires of CDU2) 2057.745 Q_62 (CDU2 atmosphere 3rd side cut), 1728.347 Q_72 (HT1 hydrogenation), 2730.000 Q_82 (HT2 hydrogenation), 343.1145 Q_92 (LHT hydrogenation) 141.2300 Q_102 (mixing the refining residue) 697.1713 Q_112 (CDU2 vacuum 1st side cut on the same day) 1002.084
0# derv fuel oil quality product on same day data	The light bavin of 0# quality product on same day data
		50% boiling range (℃) 270.1347 90% boiling ranges (℃) 307.2590 95% boiling ranges (℃) 256.0928 sulphur (W%), 0.049 acidity (mgKOH/100ml), 3.578941 flash-points (℃) 57.05042 condensation points (℃)-13.43157	50% boiling range (℃) 287.2203 90% boiling ranges (℃) 326.8638 95% boiling ranges (℃) 338.5817 sulphur (W%), 0.1800992 acidity (mgKOH/100ml), 6.990000 flash-points (℃) 71.30422 condensation points (℃)-7.768467
The 0# derv fuel oil is when day output (T)	The light bavin of 0# is when day output (T)
		3631.526	8699.692

Each required time that calculates can satisfy the needs in line computation fully less than 30 seconds.

By the calculating that the diesel oil blending production that certain factory is carried out month every day is carried out, two kinds of diesel products of the light bavin of 0# derv fuel oil and 0# are on all qualified basis of all quality index, and the ultimate production error whole month of calculated value and actual production value only is+0.32%.

The qualitative data of each component that the required data on flows of above-mentioned diesel oil blending and manual detection and analysis obtain (carrying out obtaining in preceding 2 hours in blending) all can satisfy in great majority " fuel-lubricating oil " type refinery, is existing main production process.Therefore this invention has universality.

Claims (2)

1, a kind of non-linear continuous tank type diesel oil blending method, it comprises the steps:

(1) sets up based on the linearity in the real-time qualitative attribute of each the component real-time traffic that constitutes finished diesel fuel and the dependent linearity between each flow and nonlinear relationship, each component and finished diesel fuel and the nonlinear mathematical model of nonlinear relationship;

(2) crude oil refinery speed, each side line real-time traffic data and the real-time qualitative data input step of each component (1) are set up in the nonlinear mathematical model, computing obtains constituting each component real-time traffic data of finished diesel fuel;

(3) carry out the blending of finished diesel fuel by each the component real-time traffic data of formation finished diesel fuel that obtain by step (2).

2, non-linear continuous tank type diesel oil blending method as claimed in claim 1 is characterized in that, nonlinear mathematical model is as follows described in the step (1):

$\mathrm{Max}\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{c}_j{\mathrm{QP}}_j-\underset{i=1}{\overset{m}{\mathrm{\Σ}}}\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{c}_i{Q}_{i,j}\mathrm{or}\underset{j=1}{\overset{n}{\mathrm{\Σ}}}{\mathrm{QP}}_j...\left(1\right)$

$s.t.:{\mathrm{QF}}_u=\underset{u^{\′}\∈U_u}{\mathrm{\Σ}}\underset{s\∈S_{u^{\′},u}}{\mathrm{\Σ}}{Q}_{u^{\′},s,u}...\left(2\right)$

QF _u＜QF _u-max (3)

Q _u，s＝d _u，s·QF _u，s∈S _u (4)

$Q_{u,s}=\underset{u^{\′}\∈U_{s,u}}{\mathrm{\Σ}}{Q}_{u,s,u^{\′}}...\left(5\right)$

$\frac{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{a}_{i,k}{Q}_{i,j}}{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{Q}_{i,j}}{q}_{j,k}j=\mathrm{1,2},...,n;k=\mathrm{1,2},...,l_1...\left(6\right)$

$q_{j,l_1+1}=[\ln \left(\underset{i=1}{\overset{m}{\mathrm{\Σ}}}({0.929}^{a_{i,l_1+1}}\·Q_{i,j})\right)-{\ln \mathrm{QP}}_j]/\ln 0.929,j=\mathrm{1,2},...,n...\left(7\right)$

$q_{j,l_1+2}=\exp \left[\underset{i=1}{\overset{m}{\mathrm{\Σ}}}(\frac{Q_{i,j}}{{\mathrm{QP}}_j}\·\ln a_{i,l_1+2})\right],j=\mathrm{1,2},...,n...\left(8\right)$

$q_{j,l_1+3}={9.4656T}^3-{57.08217T}^2+129.075T-99.2741$

$T=f[(\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{Q}_{i,j}\·a_{i,l_1+3})/\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{Q}_{i,j}],j=\mathrm{1,2},...,n...\left(9\right)$

b _min，j，k≤q _j，k≤b _max，j，k j＝1，2，…，n；k＝1，2，…，l (10)

Wherein:

U---can produce the title of the device of diesel component oil;

N---the species number of diesel oil the finished product, as: 0# car bavin, the light bavin of 0# ,-10# diesel oil etc.;

M---participate in the quantity of the blending component of diesel oil finished product blending;

L---in diesel oil processed oil blending, need the main performance index number considered;

S _{U ', u}---the set of the oil stream of separating device u ' access to plant u;

S _u---the set of all products that device u produces;

U _{S, u}---receive the set of all device u of oil stream s;

U _u---there is product oil to flow directly into the set of all preceding continuous devices of device u;

QF _u---the feeding total flux of device u;

QF _U-max---the maximum processing capability of device u, i.e. maximum inlet total flux (ton/sky);

Q _{U, s}---the flow of certain product s that device u produces;

d _{U, s}---the output capacity (%) of certain product s that device u produces;

Q _{U, s, u '}---the flow of certain product s access to plant u ' that device u produces;

Q _{I, j}---finished product j (j=1,2 ..., n) in blending component i (i=1,2 ..., flow m);

QP _j---finished product j (j=1,2 ..., flow n);

c _j---diesel oil processed oil j (j=1,2 ..., the unit price that dispatches from the factory n);

c _i---blending component i (i=1,2 ..., m) advance factory's unit price;

α _{I, k}---blending component i (i=1,2 ..., a kind of performance index k m) (k=1,2 ..., l ₁, l ₁+ 1 ..., l), wherein preceding l ₁Individual is linear restriction, from l ₁+ 1 to l (=l ₁+ 3) individual is non-linear constrain;

q _{J, k}---finished product j (j=1,2 ..., a kind of performance index k n) (k=1,2 ..., l);

b _{Max, j, k}, b _{Min, j, k}---finished product j (j=1,2 ..., quality index k n) (k=1,2 ..., the upper and lower limit of binding occurrence l).