CN106801114B - A kind of blast furnace material distribution process burden distribution matrix optimization method and system - Google Patents
A kind of blast furnace material distribution process burden distribution matrix optimization method and system Download PDFInfo
- Publication number
- CN106801114B CN106801114B CN201710003901.XA CN201710003901A CN106801114B CN 106801114 B CN106801114 B CN 106801114B CN 201710003901 A CN201710003901 A CN 201710003901A CN 106801114 B CN106801114 B CN 106801114B
- Authority
- CN
- China
- Prior art keywords
- charge
- shape
- charge level
- particle
- distribution matrix
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- C—CHEMISTRY; METALLURGY
- C21—METALLURGY OF IRON
- C21B—MANUFACTURE OF IRON OR STEEL
- C21B5/00—Making pig-iron in the blast furnace
- C21B5/006—Automatically controlling the process
-
- C—CHEMISTRY; METALLURGY
- C21—METALLURGY OF IRON
- C21B—MANUFACTURE OF IRON OR STEEL
- C21B7/00—Blast furnaces
- C21B7/18—Bell-and-hopper arrangements
- C21B7/20—Bell-and-hopper arrangements with appliances for distributing the burden
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- C—CHEMISTRY; METALLURGY
- C21—METALLURGY OF IRON
- C21B—MANUFACTURE OF IRON OR STEEL
- C21B2300/00—Process aspects
- C21B2300/04—Modeling of the process, e.g. for control purposes; CII
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Manufacturing & Machinery (AREA)
- Materials Engineering (AREA)
- Metallurgy (AREA)
- Organic Chemistry (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Blast Furnaces (AREA)
Abstract
The present invention provides a kind of blast furnace material distribution process burden distribution matrix optimization method and system, this method comprises: setting ideal shape of charge level;To minimize the shape of charge level of next group furnace charge and the deviation of ideal shape of charge level as target, optimize burden distribution matrix;Optimize burden distribution matrix of the obtained optimal burden distribution matrix as next group furnace charge.The system includes: setup unit, optimization unit, output unit.The present invention provides foundation for the adjusting of blast furnace material distribution matrix, the burden distribution matrix that the method and system that operator can provide according to the present invention optimize, realize the adjusting of burden distribution matrix in blast furnace material distribution operation, the present invention optimizes the shape of charge level under gained burden distribution matrix close to ideal shape of charge level.Make gas fluid in blast furnace reasonable layout, improves raw material availability, achieve the effect that blast furnace is highly efficient and productive.
Description
Technical field
The invention belongs to blast furnace process automatic control technology field, in particular to a kind of blast furnace material distribution process burden distribution matrix is excellent
Change method and system.
Background technique
Cloth operation is one of big basic operation system of blast furnace ironmaking four, decides furnace charge inside the shape and blast furnace of charge level
Layered distribution.Directly affect gas fluid distrbution, the gas utilization rate of upper blast furnace.Charge level shape is adjusted by burden distribution system
Shape makes Gas Flow reasonable layout inside blast furnace, increases the utilization rate of coal gas, achieve the effect that high-yield and high-efficiency, to the stabilization of blast furnace
Direct motion has profound significance.However, blast furnace is one along with high temperature, high pressure, closed large-scale reactor, due to measuring instrument
The limitation of device is difficult to directly observe blast furnace inside in normal production now.Thus, it is operated by cloth and adjusts burden distribution matrix
Obtained shape of charge level can not also be predicted.Current practices personnel mainly pass through according to stock rod height and cross temperature temperature in conjunction with itself
Estimation shape of charge level indirectly is tested, there is extremely strong subjectivity and contingency, lead to that reasonable cloth operation can not be carried out often.Cause
This, it is necessary to accurate mathematical model is established, the shape of charge level under different burden distribution matrixes is obtained, instructs operator to carry out reasonable
Cloth operation.
Patent publication No. CN106011350A proposes a kind of estimation method of the shape of charge level of blast furnace material distribution, it is already possible to
The shape of charge level under different burden distribution matrixes is obtained, realizes the visualization of shape of charge level to a certain extent.However, this method is also
It cannot achieve and operator is instructed to be adjusted burden distribution matrix, the shape of charge level under burden distribution matrix adjusted is made to reach ideal
Shape of charge level.
Summary of the invention
In view of the above-mentioned problems, the present invention proposes a kind of blast furnace material distribution process burden distribution matrix optimization method and system, make to optimize
Shape of charge level under gained burden distribution matrix finally makes blast furnace reach stabilization, direct motion, highly efficient and productive close to ideal shape of charge level
Effect.
The technical scheme is that
A kind of blast furnace material distribution process burden distribution matrix optimization method, comprising:
Set ideal shape of charge level;
To minimize the shape of charge level of next group furnace charge and the deviation of ideal shape of charge level as target, optimize burden distribution matrix;
Optimize burden distribution matrix of the obtained optimal burden distribution matrix as next group furnace charge.
Burden distribution matrix: initial bit of the burden distribution matrix being randomly generated as particle in population is optimized using particle swarm algorithm
It sets, burden distribution matrix corresponding to particle current location is formed by square of the deviation between shape of charge level and ideal shape of charge level
With as adaptive value, iteration optimization particle position, the population optimal location acquired the i.e. optimal burden distribution matrix of next group furnace charge.
It is described that burden distribution matrix is optimized using particle swarm algorithm, comprising:
Initial position of the burden distribution matrix being randomly generated as particle in population;
Obtain burden distribution matrix corresponding to the cloth process parameter, furnace charge parameter and particle current location of next group furnace charge;
It calculates next group furnace charge burden distribution matrix according to corresponding to particle current location and is formed by shape of charge level;
It is formed by with burden distribution matrix corresponding to particle current location inclined between shape of charge level and ideal shape of charge level
The quadratic sum of difference is as adaptive value, iteration optimization particle position, the population optimal location acquired the i.e. optimal cloth of next group furnace charge
Expect matrix.
Calculating next group furnace charge burden distribution matrix according to corresponding to particle current location is formed by shape of charge level, packet
It includes:
The initial velocity that furnace charge leaves throttle valve is calculated according to the cloth process parameter of next group furnace charge;
Furnace charge is calculated in the speed of central throat tube end according to the initial velocity that furnace charge leaves throttle valve;
Speed according to furnace charge in central throat tube end calculates speed when furnace charge reaches chute;
Speed when reaching chute according to furnace charge calculates furnace charge in the speed of chute end;
Speed according to furnace charge in chute end calculates each chute in burden distribution matrix corresponding to particle current location and inclines
The formed material heap heap cusp abscissa of furnace charge under the gear of angle;
Calculate the shape of charge level under the last one chute inclination angle corresponding to particle current location in burden distribution matrix, i.e., under
Shape of charge level under a collection of furnace charge burden distribution matrix corresponding to particle current location.
The present invention also provides a kind of blast furnace material distribution process burden distribution matrix optimization systems, comprising:
Setup unit: ideal shape of charge level is set;
Optimization unit: to minimize the shape of charge level of next group furnace charge and the deviation of ideal shape of charge level as target, optimize
Burden distribution matrix;
Output unit: the optimal burden distribution matrix optimized is exported as the burden distribution matrix of next group furnace charge.
The optimization unit, optimize burden distribution matrix using particle swarm algorithm: the burden distribution matrix being randomly generated is as in population
The initial position of particle, burden distribution matrix corresponding to particle current location are formed by between shape of charge level and ideal shape of charge level
Deviation quadratic sum as adaptive value, iteration optimization particle position, the population optimal location acquired i.e. next group furnace charge is most
Excellent burden distribution matrix.
The optimization unit, comprising:
Initialization module: initial position of the burden distribution matrix being randomly generated as particle in population;
It obtains module: obtaining corresponding to the cloth process parameter, furnace charge parameter and particle current location of next group furnace charge
Burden distribution matrix;
Shape of charge level computing module: it calculates next group furnace charge burden distribution matrix according to corresponding to particle current location and is formed
Shape of charge level;
Iteration optimization module: shape of charge level and ideal charge level are formed by with burden distribution matrix corresponding to particle current location
The quadratic sum of deviation between shape is as adaptive value, iteration optimization particle position, the population optimal location i.e. next group acquired
The optimal burden distribution matrix of furnace charge.
The shape of charge level computing module, comprising:
First computing module: the initial speed that furnace charge leaves throttle valve is calculated according to the cloth process parameter of next group furnace charge
Degree;
Second computing module: furnace charge is calculated in the speed of central throat tube end according to the initial velocity that furnace charge leaves throttle valve
Degree;
Third computing module: the speed according to furnace charge in central throat tube end calculates speed when furnace charge reaches chute;
4th computing module: speed when reaching chute according to furnace charge calculates furnace charge in the speed of chute end;
5th computing module: the speed according to furnace charge in chute end calculates burden distribution matrix corresponding to particle current location
In each chute inclination angle gear under the formed material heap heap cusp abscissa of furnace charge;
6th computing module: it calculates under the last one chute inclination angle in burden distribution matrix corresponding to particle current location
Shape of charge level, i.e. shape of charge level under next group furnace charge burden distribution matrix corresponding to particle current location.
The utility model has the advantages that
The present invention provides foundation for the adjusting of blast furnace material distribution matrix, method that operator can provide according to the present invention and
The burden distribution matrix that system optimizes, realizes the adjusting of burden distribution matrix in blast furnace material distribution operation, and the present invention optimizes gained cloth
Shape of charge level under matrix is close to ideal shape of charge level.Make gas fluid in blast furnace reasonable layout, improve raw material availability, reaches high
The highly efficient and productive effect of furnace.
Detailed description of the invention
Fig. 1 is ideal shape of charge level figure in the specific embodiment of the invention;
Fig. 2 is basic shape of charge level sectional view in the specific embodiment of the invention;
Fig. 3 is charge level shaped cross figure under first chute inclination angle in the specific embodiment of the invention;
Fig. 4 is the Growth Mechanism sectional view of polycyclic distributing charge level in the specific embodiment of the invention;
Fig. 5 is current shape of charge level figure compared with ideal shape of charge level in the specific embodiment of the invention;
Fig. 6 is optimization process iteration diagram in the specific embodiment of the invention;
Fig. 7 is effect of optimization figure in the specific embodiment of the invention;
Fig. 8 is blast furnace material distribution process burden distribution matrix optimization method flow chart in the specific embodiment of the invention;
Fig. 9 is blast furnace material distribution process burden distribution matrix optimization method step 2 flow chart in the specific embodiment of the invention;
Figure 10 is blast furnace material distribution process burden distribution matrix optimization method step 2-3 flow chart in the specific embodiment of the invention;
Figure 11 is blast furnace material distribution process burden distribution matrix optimization system block diagram in the specific embodiment of the invention;
Figure 12 is to optimize unit block diagram in the specific embodiment of the invention;
Figure 13 is shape of charge level computing module block diagram in the specific embodiment of the invention.
Specific embodiment
Specific embodiments of the present invention will be described in detail with reference to the accompanying drawing.
Implement method and system of the invention without clock-type blast furnace for string tank in present embodiment, the communicated band of furnace charge is successively
Reach material feeding jar, blanking tank, central throat tube and swivel chute, final furnace charge cloth is to inside furnace throat.The blast furnace material distribution process cloth
Matrix optimizing method, as shown in Figure 8, comprising:
Step 1 sets ideal shape of charge level in conjunction with feed information and the current working of a furnace;
Furnace throat position shape of charge level of the tank without clock-type blast furnace of going here and there is that can guarantee good coal gas flow point under ideal charge level
Cloth and raw material availability.Usually ideal shape of charge level center journey " funnel " type, both sides are " platform " type, as shown in Figure 1, setting
Radial ideal shape of charge level γb(x) expression formula are as follows:
Wherein,For the interior angle of rest (repose) of furnace charge,For the outer angle of rest (repose) of furnace charge, x is the distance to blast furnace center, and R is furnace throat half
Diameter;
The deviation of step 2, the shape of charge level to minimize next group furnace charge and ideal shape of charge level optimizes cloth as target
Matrix;
Set objective function are as follows:
Constraint condition are as follows:
Wherein, ψ (x, αi, mi) it is shape of charge level function, αiFor i-th of chute inclination angle, miFor i-th chute inclination angle is right
The rotating cycle answered;I=1,2 ... ... n, wherein n is the cloth gear number set in cloth process, and n usually takes 4~7.
Optimize burden distribution matrix using particle swarm algorithm in present embodiment: the burden distribution matrix being randomly generated is as grain in population
The initial position of son, burden distribution matrix corresponding to particle current location are formed by between shape of charge level and ideal shape of charge level
The quadratic sum of deviation is as adaptive value, iteration optimization particle position, the i.e. next group furnace charge of the population optimal location acquired it is optimal
Burden distribution matrix.
It is described that burden distribution matrix is optimized using particle swarm algorithm, as shown in Figure 9, comprising:
Step 2-1, initial position of the burden distribution matrix being randomly generated as particle in population;
Parameter setting: Studying factors c is carried out according to Liu Gang 2# blast furnace actual conditions1=c2=2, inertia weight maximum value
ωmax=0.9, minimum value ωmin=0.4, maximum number of iterations is 100 times, population number N=100, the dimension D=10 of particle,
The position x of i-th of particlei=(αi1, αi2..., αi5, mi1, mi2..., mi5), wherein αi1, αi2..., αi5Chute inclination angle is represented,
[10 °, 45 °] of value range, and be integer, mi1, mi2..., mi5Represent the corresponding rotating cycle in each chute inclination angle, value
Range is [Isosorbide-5-Nitrae], and is integer, the velocity interval [v of particlemin, vmax] it is [0,2].Particle position value range and
The position and speed of particle is initialized in the value range of speed.
Before in the initial position of particleEach chute inclination angle is necessary for the random integers in [10 °, 45 °] range in dimension,
AfterwardsThe rotating cycle under corresponding chute inclination angle is tieed up, value should be the random integers in [Isosorbide-5-Nitrae] range.
Step 2-2, cloth corresponding to the cloth process parameter, furnace charge parameter and particle current location of next group furnace charge is obtained
Expect matrix;
The cloth process parameter of next group furnace charge, comprising: furnace charge batch weight M, central throat tube length H, throttle valve opening κ, group
At the radius R of the hemispherical material valve of throttle valve1, furnace throat radius R, chute length L, chute revolving speed w, chute fascinate and rub away from b, chute
Wipe coefficient μ, stockline depth h0Pass through the time T of throttle valve with furnace charge.
The furnace charge parameter of next group furnace charge includes: the interior angle of rest (repose) of the averag density ρ of furnace charge, furnace chargeWith outer angle of rest (repose)
Burden distribution matrix corresponding to particle current location, comprising: chute inclination alpha=[α1, α2..., αi... αn] and
Rotating cycle m=[m corresponding to each chute inclination angle1, m2..., mi... mn], 10 °≤αi≤ 45 °, 1≤mi≤4。
Step 2-3, it calculates next group furnace charge burden distribution matrix according to corresponding to particle current location and is formed by charge level shape
Shape;
Specific steps are as shown in Figure 10, comprising:
Step 2-3-1, the initial velocity that furnace charge leaves throttle valve is calculated according to the cloth process parameter of next group furnace charge;
According to throttle valve opening κ, the radius R for the hemispherical material valve for forming throttle valve1, the batch weight M of furnace charge, furnace charge density
ρ, furnace charge calculate the initial velocity that furnace charge leaves throttle valve by time T of throttle valveWherein S0For throttle valve cross
Sectional area.
Step 2-3-2, the initial velocity v of throttle valve is left according to furnace charge0Furnace charge is calculated in the speed of central throat tube end
v1;
Furnace charge by central throat tube can approximation regard furnace charge as and do the movement of falling object, only by gravity, height of drop
For center length of throat H, meet following relationship:G=9.8m/s2For acceleration of gravity.
Step 2-3-3, the speed according to furnace charge in central throat tube end calculates speed v when furnace charge reaches chute2;
Furnace charge speed during hitting with chute would detract from losing, and direction will change, and meet following relationship: v2
=kcos α v1, k is furnace charge impact velocity attenuation coefficient, k=0.8.
Step 2-3-4, speed when reaching chute according to furnace charge calculates furnace charge in the speed v of chute end3;
Furnace charge when being moved on chute, respectively by self gravity, the holding power of chute, frictional force, coriolis force and by
Centrifugal force caused by chute rotates, these power are decomposed, and furnace charge will be accelerated along chute direction, meet such as ShiShimonoseki
System:
Step 2-3-5, the speed v according to furnace charge in chute end3Calculate burden distribution matrix corresponding to particle current location
In each chute inclination angle gear under the formed material heap heap cusp abscissa of furnace charge;
Furnace charge is acted on after leaving chute by the drag of gravity and rising coal gas, inclined throw movements is done in dead zone, due to coal gas
Drag smaller negligible do not remember.Therefore furnace charge can be approximately considered only by gravity.It is obtained under n chute inclination angle by kinematical equation
The formed material heap heap cusp abscissa x of furnace chargenAre as follows:
Step 2-3-6, the material under the last one chute inclination angle in burden distribution matrix corresponding to particle current location is calculated
Face shape, i.e. shape of charge level under next group furnace charge burden distribution matrix corresponding to particle current location.
Step 2-3-6-1, basic shape of charge level is constructed;
As shown in Fig. 2, the cross section profile of shape of charge level of the furnace charge in furnace throat can approximation regard as and be made of straight line CB, CA,
The inside and outside angle of rest (repose) of furnace charge is the angle of this two straight lines and horizontal plane.
Construct the function of basic shape of charge level are as follows:
In formula, m=1 is the rotating cycle of chute, and line segment BCA is basic shape of charge level, (xn, yh) material formed by furnace charge
Heap heap cusp C coordinate, rLFor the abscissa of straight line CB and horizontal line OL intersection point, rRFor the abscissa of straight line CA and furnace wall intersection point.
Step 2-3-6-2, relationship, that is, shape of charge level function between the shape of charge level and basic shape of charge level of building output,
Shape of charge level described in the shape of charge level function of output is the superposition of basic shape of charge level;
Shape of charge level is mainly determined that final shape of charge level is the superposition of basic shape of charge level, building material by burden distribution matrix
Face shape function is as follows:
In formula, γ (x) is shape of charge level function, α=[α1, α2..., αi... αn] it is chute inclination angle, m=[m1,
m2..., mi... mn] it is the corresponding rotating cycle in chute inclination angle.
Step 2-3-6-3, the cloth Material in burden distribution matrix corresponding to particle current location under each chute inclination angle is constructed
Constraint condition is accumulated, i.e., the cloth volume under each chute inclination angle is equal under the chute inclination angle under shape of charge level and upper chute inclination angle
The volume integral at shape of charge level interval;
Since volume does not change furnace charge during the motion, therefore the shape of charge level and initial shape of charge level (water exported
Plane) between furnace charge volume it is equal with the volume that furnace charge enters material feeding jar, furnace charge volume meets following constraint:
V=M/ ρ (8)
In formula, V is furnace charge volume.
The cloth volume constraint condition under each chute inclination angle is constructed according to formula (8), (9), i.e., under each chute inclination angle
Cloth volume be equal to the volume integral at shape of charge level interval under shape of charge level and upper chute inclination angle under the chute inclination angle:
Cloth volume constraint condition is as follows:
In formula, miFor the rotating cycle under i-th of chute inclination angle, γi(x) the charge level shape to be exported under i-th of chute inclination angle
Shape.
Step 2-3-6-4, according to the cloth volume constraint condition under first chute inclination angle, particle current location institute is calculated
The formed material heap heap cusp ordinate of furnace charge under first chute inclination angle in corresponding burden distribution matrix;
The volume constraint that furnace charge meets under first chute inclination angle is as follows:
With the progress of cloth, material stack height yh1It gradually increases.When cloth is completed, pass through iteration furnace charge heap cusp height yh1,
The height y of lower the formed material heap heap point in first chute inclination angle can be acquiredh1That is the formed material heap of furnace charge under first chute inclination angle
Heap cusp ordinate, then heap cusp C1 C1(xn1, yh1) can uniquely determine.Again since interior angle of rest (repose), the outer angle of rest (repose) of furnace charge are constant, therefore
Straight line C1B1With straight line C1A1Slope it is constant, it is known that on straight line a little and its slope can uniquely determine straight line.It then can be unique
Acquire the intersection points B of left and right charge level Yu initial charge level1、A1.Its abscissa r can be obtainedL1、rR1。
Step 2-3-6-5, furnace charge institute under first chute inclination angle in the burden distribution matrix according to corresponding to particle current location
Material heap heap cusp coordinate and shape of charge level function are formed, first chute in burden distribution matrix corresponding to particle current location is solved
Shape of charge level under inclination angle.
As shown in figure 3, setting blast furnace center as coordinate origin, the horizontal plane where straight line OL is initial charge level, then basis
Formula (6), (7) can obtain the shape of charge level in burden distribution matrix corresponding to particle current location under first chute inclination angle are as follows:
In formula, (xn1, yh1) by forming material heap heap cusp C under first chute inclination angle1Coordinate, rL1For straight line C1B1With
The abscissa of initial charge level intersection point, rR1For straight line C1A1With the abscissa of initial charge level intersection point.
Step 2-3-7, the body that furnace charge meets under each chute inclination angle in the burden distribution matrix according to corresponding to particle current location
The shape of charge level of constraint, the shape of charge level under previous chute inclination angle and basic charge level is accumulated, successively iteration finds out second chute
Shape of charge level under inclination angle to complete n-th of cloth each chute inclination angle, the shape of charge level under the last one the chute inclination angle acquired, i.e. grain
Burden distribution matrix corresponding to sub- current location is formed by shape of charge level.
As shown in figure 4, with the progress of cloth under second chute inclination angle, material stack height yh2It gradually increases.By continuous
Change furnace charge heap cusp height yh2, so that it is met formula (11), the height y of furnace charge heap point under second chute inclination angle can be acquiredh2I.e.
The formed material heap heap cusp ordinate of furnace charge under second chute inclination angle, then heap cusp C2(xn2, yh2) can uniquely determine, and due to
The inside and outside angle of rest (repose) of furnace charge is constant, therefore straight line C2B2With straight line C2A2Slope it is constant, then can uniquely acquire under second chute inclination angle
The intersection points B of charge level under the left and right charge level of furnace charge formation and first chute inclination angle2、A2.R can be obtainedL2, rR2.Therefore second can be obtained
Shape of charge level γ under a chute inclination angle2(x).Similarly, the shape of charge level under third to n-th of chute inclination angle can successively be acquired.
Polycyclic distributing is the superposition of institute's cloth furnace charge shape of charge level under each chute inclination angle, and current shape of charge level will be used as down
One encloses the initial shape of charge level of cloth, i.e. shape of charge level under first chute inclination angle will be as the initial charge level of the second circle cloth
Shape, the following are the shape of charge level behind cloth complete second chute inclination angle to cloth complete n-th of chute inclination angle.
Shape of charge level under second and n-th of chute inclination angle is respectively as follows:
In formula, (xni, yni) by forming material heap heap cusp coordinate, γ under i-th of chute inclination anglei(x) incline for i-th of chute
Shape of charge level under angle.rLi、rRiFor institute's cloth furnace charge under i-th of chute inclination angle and upper charge level γi-1(x) in heap cusp (xni,
yni) left and right sides intersection points Bi、AiAbscissa.
Shape of charge level γ under the shape of charge level γ (x) of final output and the last one chute inclination anglen(x) identical.
Step 2-4, shape of charge level and ideal shape of charge level are formed by with burden distribution matrix corresponding to particle current location
Between deviation quadratic sum as adaptive value, iteration optimization particle position, the population optimal location acquired i.e. next group furnace charge
Optimal burden distribution matrix.
Step 2-4-1, adaptive value is evaluated:
Adaptation value function in present embodiment is the cloth optimized in ideal shape of charge level function and particle group optimizing
Expect the quadratic sum of the deviation of the formed shape of charge level of matrix, i.e.,Wherein, initial charge level is current cloth
Expect that the shape of charge level under matrix, Fig. 5 are the comparison of current shape of charge level and ideal shape of charge level.For each particle, worked as
The adaptive value of front position is compared with the adaptive value of its excellent position of history experienced, if the adaptive value of its current location is excellent
In the adaptive value of its history optimal location, then its history optimal value is replaced, replaces history optimal with current position
Place value.The adaptive value of the adaptive value of the individual history optimal location of each particle and the history optimal location of group is compared
Compared with, when the adaptive value of personal best particle is better than the adaptive value of history optimal location, then group's optimal location is replaced,
Group's optimal location is replaced with individual history optimal location.
Step 2-4-2, particle state updates:
The state of particle includes the position of particle and the speed of particle, utilizes formula (15) to the speed of particle in present embodiment
Degree is updated, and is updated using formula (16) to the position of particle.
In formula,For the position of k+1 moment particle,For the position of k moment particle,For the speed of k+1 moment particle
Degree,For the speed of k moment particle, ξ1With ξ2It is the random number in [0,1] range,It is optimal in k moment individual for particle
Position,It is particle in k moment population optimal location;
Step 2-4-3, stop iteration judgement: iterative process as shown in fig. 6, as can be seen from the figure when 20 step of iteration,
Preferable effect is had reached, when the number of iterations arrival maximum number of iterations 100 times, then stops the iteration output optimal position of current group
It sets, i.e., optimal burden distribution matrix is precedingA is chute inclination angle, afterBefore number is followed successively byIt is corresponding under a chute inclination angle to slip
Slot rotating cycle, since chute is to turn to lesser inclination angle by biggish inclination angle in cloth process, it is therefore desirable to chute inclination angle
Angle it is descending successively sort, while corresponding rotating cycle under the chute inclination angle is followed into the carry out sequence change of chute inclination angle
It changes, obtains the burden distribution matrix of final optimization pass.Otherwise, go to step 2-4-1, continues iteration.Final optimization pass is obtained
Burden distribution matrix calls shape of charge level calculation method, obtains the shape of charge level and ideal charge level shape under the burden distribution matrix as shown in Figure 7
The comparison diagram of shape.
Burden distribution matrix of the optimal burden distribution matrix that step 3, optimization obtain as next group furnace charge.
The present invention also provides a kind of blast furnace material distribution process burden distribution matrix optimization systems, as shown in figure 11, comprising:
Setup unit: ideal shape of charge level is set;
Optimization unit: to minimize the shape of charge level of next group furnace charge and the deviation of ideal shape of charge level as target, optimize
Burden distribution matrix;
Output unit: the optimal burden distribution matrix optimized is exported as the burden distribution matrix of next group furnace charge.
The optimization unit, optimize burden distribution matrix using particle swarm algorithm: the burden distribution matrix being randomly generated is as in population
The initial position of particle, burden distribution matrix corresponding to particle current location are formed by between shape of charge level and ideal shape of charge level
Deviation quadratic sum as adaptive value, iteration optimization particle position, the population optimal location acquired i.e. next group furnace charge is most
Excellent burden distribution matrix.
The optimization unit, as shown in figure 12, comprising:
Initialization module: initial position of the burden distribution matrix being randomly generated as particle in population;
It obtains module: obtaining corresponding to the cloth process parameter, furnace charge parameter and particle current location of next group furnace charge
Burden distribution matrix;
Shape of charge level computing module: it calculates next group furnace charge burden distribution matrix according to corresponding to particle current location and is formed
Shape of charge level;
Iteration optimization module: shape of charge level and ideal charge level are formed by with burden distribution matrix corresponding to particle current location
The quadratic sum of deviation between shape is as adaptive value, iteration optimization particle position, the population optimal location i.e. next group acquired
The optimal burden distribution matrix of furnace charge.
The shape of charge level computing module, as shown in figure 13, comprising:
First computing module: the initial speed that furnace charge leaves throttle valve is calculated according to the cloth process parameter of next group furnace charge
Degree;
Second computing module: furnace charge is calculated in the speed of central throat tube end according to the initial velocity that furnace charge leaves throttle valve
Degree;
Third computing module: the speed according to furnace charge in central throat tube end calculates speed when furnace charge reaches chute;
4th computing module: speed when reaching chute according to furnace charge calculates furnace charge in the speed of chute end;
5th computing module: the speed according to furnace charge in chute end calculates burden distribution matrix corresponding to particle current location
In each chute inclination angle gear under the formed material heap heap cusp abscissa of furnace charge;
6th computing module: it calculates under the last one chute inclination angle in burden distribution matrix corresponding to particle current location
Shape of charge level, i.e. shape of charge level under next group furnace charge burden distribution matrix corresponding to particle current location.
Blast furnace material distribution process burden distribution matrix optimization method provided by the invention and system, can optimize burden distribution matrix,
Make the shape of charge level under optimization gained burden distribution matrix close to ideal shape of charge level, to the tune of burden distribution matrix in blast furnace material distribution operation
It is whole that there is important directive significance.
Claims (2)
1. a kind of blast furnace material distribution process burden distribution matrix optimization method, comprising:
Set ideal shape of charge level;
To minimize the shape of charge level of next group furnace charge and the deviation of ideal shape of charge level as target, optimize burden distribution matrix;
Optimize burden distribution matrix of the obtained optimal burden distribution matrix as next group furnace charge;
Burden distribution matrix: initial position of the burden distribution matrix being randomly generated as particle in population, grain is optimized using particle swarm algorithm
The quadratic sum that burden distribution matrix corresponding to sub- current location is formed by the deviation between shape of charge level and ideal shape of charge level is made
For adaptive value, iteration optimization particle position, the population optimal location acquired the i.e. optimal burden distribution matrix of next group furnace charge;
It is described that burden distribution matrix is optimized using particle swarm algorithm, comprising:
Initial position of the burden distribution matrix being randomly generated as particle in population;
Obtain burden distribution matrix corresponding to the cloth process parameter, furnace charge parameter and particle current location of next group furnace charge;
It calculates next group furnace charge burden distribution matrix according to corresponding to particle current location and is formed by shape of charge level;
The deviation between shape of charge level and ideal shape of charge level is formed by with burden distribution matrix corresponding to particle current location
Quadratic sum is as adaptive value, iteration optimization particle position, the optimal cloth square of the i.e. next group furnace charge of the population optimal location acquired
Battle array;
Calculating next group furnace charge burden distribution matrix according to corresponding to particle current location is formed by shape of charge level, comprising:
The initial velocity that furnace charge leaves throttle valve is calculated according to the cloth process parameter of next group furnace charge;
Furnace charge is calculated in the speed of central throat tube end according to the initial velocity that furnace charge leaves throttle valve;
Speed according to furnace charge in central throat tube end calculates speed when furnace charge reaches chute;
Speed when reaching chute according to furnace charge calculates furnace charge in the speed of chute end;
Speed according to furnace charge in chute end calculates each chute inclination angle shelves in burden distribution matrix corresponding to particle current location
Lower the formed material heap heap cusp abscissa of furnace charge in position;
Calculate the shape of charge level under the last one chute inclination angle in burden distribution matrix corresponding to particle current location, i.e. next group
Shape of charge level under furnace charge burden distribution matrix corresponding to particle current location;
It is characterized in that, the blast furnace is string tank without clock-type blast furnace;
To minimize the shape of charge level of next group furnace charge and the deviation of ideal shape of charge level as target, optimize burden distribution matrix, specifically
The objective function of setting are as follows:
Constraint condition are as follows:
Wherein, ψ (x, αi,mi) it is shape of charge level function, αiFor i-th of chute inclination angle, miFor rotation corresponding to i-th of chute inclination angle
It turn-takes number;I=1,2 ... ... n, wherein n is the cloth gear number set in cloth process;
Calculate the formula of the shape of charge level under the last one chute inclination angle in burden distribution matrix corresponding to particle current location such as
Under:
Shape of charge level under second and n-th of chute inclination angle is respectively as follows:
In formula, chute inclination alpha=[α1,α2,...,αi,...αn], rotating cycle m=[m corresponding to each chute inclination angle1,
m2,...,mi,...mn], αiFor i-th of chute inclination angle, miFor rotating cycle corresponding to i-th of chute inclination angle;For furnace charge
Interior angle of rest (repose), R be furnace throat radius, rL2、rR2The left and right charge level formed for furnace charge under second chute inclination angle inclines with first chute
The intersection point of charge level under angle;(xn2,yh2) material heap heap cusp coordinate, (x formed by furnace charge under second chute inclination angleni,yni) it is the
Lower the formed material heap heap cusp coordinate in i chute inclination angle, γiIt (x) is the shape of charge level under i-th of chute inclination angle;rLi、rRiIt is
Institute's cloth furnace charge and upper charge level γ under i chute inclination anglei-1(x) in heap cusp (xni,yni) left and right sides intersection point abscissa;
Shape of charge level γ under the shape of charge level γ (x) of final output and the last one chute inclination anglen(x) identical;
When iteration optimization particle position, the speed of particle is updated using formula (15), using formula (16) to the position of particle
It is updated:
In formula,For the position of k+1 moment particle,For the position of k moment particle,For the speed of k+1 moment particle,For the speed of k moment particle, ξ1With ξ2It is the random number in [0,1] range,It is particle in the optimal position of k moment individual
It sets,It is particle in k moment population optimal location.
2. a kind of blast furnace material distribution process burden distribution matrix optimization system, comprising:
Setup unit: ideal shape of charge level is set;
Optimization unit: to minimize the shape of charge level of next group furnace charge and the deviation of ideal shape of charge level as target, optimize cloth
Matrix;
Output unit: the optimal burden distribution matrix optimized is exported as the burden distribution matrix of next group furnace charge;
The optimization unit, optimize burden distribution matrix using particle swarm algorithm: the burden distribution matrix being randomly generated is as particle in population
Initial position, burden distribution matrix corresponding to particle current location is formed by inclined between shape of charge level and ideal shape of charge level
The quadratic sum of difference is as adaptive value, iteration optimization particle position, the population optimal location acquired the i.e. optimal cloth of next group furnace charge
Expect matrix;
The optimization unit, comprising:
Initialization module: initial position of the burden distribution matrix being randomly generated as particle in population;
It obtains module: obtaining cloth corresponding to the cloth process parameter, furnace charge parameter and particle current location of next group furnace charge
Matrix;
Shape of charge level computing module: it calculates next group furnace charge burden distribution matrix according to corresponding to particle current location and is formed by material
Face shape;
Iteration optimization module: shape of charge level and ideal shape of charge level are formed by with burden distribution matrix corresponding to particle current location
Between deviation quadratic sum as adaptive value, iteration optimization particle position, the population optimal location acquired i.e. next group furnace charge
Optimal burden distribution matrix;
The shape of charge level computing module, comprising:
First computing module: the initial velocity that furnace charge leaves throttle valve is calculated according to the cloth process parameter of next group furnace charge;
Second computing module: furnace charge is calculated in the speed of central throat tube end according to the initial velocity that furnace charge leaves throttle valve;
Third computing module: the speed according to furnace charge in central throat tube end calculates speed when furnace charge reaches chute;
4th computing module: speed when reaching chute according to furnace charge calculates furnace charge in the speed of chute end;
5th computing module: the speed according to furnace charge in chute end calculates in burden distribution matrix corresponding to particle current location
The formed material heap heap cusp abscissa of furnace charge under the gear of each chute inclination angle;
6th computing module: the charge level under the last one chute inclination angle in burden distribution matrix corresponding to particle current location is calculated
Shape, i.e. shape of charge level under next group furnace charge burden distribution matrix corresponding to particle current location;
It is characterized in that, the deviation of shape of charge level and ideal shape of charge level to minimize next group furnace charge optimizes cloth as target
Expect matrix, the objective function specifically set are as follows:
Constraint condition are as follows:
Wherein, ψ (x, αi,mi) it is shape of charge level function, αiFor i-th of chute inclination angle, miFor rotation corresponding to i-th of chute inclination angle
It turn-takes number;I=1,2 ... ... n, wherein n is the cloth gear number set in cloth process;
In 6th computing module, the material under the last one chute inclination angle in burden distribution matrix corresponding to particle current location is calculated
The formula of face shape is as follows:
Shape of charge level under second and n-th of chute inclination angle is respectively as follows:
In formula, chute inclination alpha=[α1,α2,...,αi,...αn], rotating cycle m=[m corresponding to each chute inclination angle1,
m2,...,mi,...mn], αiFor i-th of chute inclination angle, miFor rotating cycle corresponding to i-th of chute inclination angle;I=1,
2 ... ... n, wherein n is the cloth gear number set in cloth process,For the interior angle of rest (repose) of furnace charge, R is furnace throat radius, rL2、rR2
The intersection point of charge level under the left and right charge level formed for furnace charge under second chute inclination angle and first chute inclination angle;(xn2,yh2) it is the
The formed material heap heap cusp coordinate of furnace charge, (x under two chute inclination anglesni,yni) by forming material heap heap under i-th of chute inclination angle
Cusp coordinate, γiIt (x) is the shape of charge level under i-th of chute inclination angle;rLi、rRiFor institute's cloth furnace charge under i-th of chute inclination angle with it is upper
One charge level γi-1(x) in heap cusp (xni,yni) left and right sides intersection point abscissa;
Shape of charge level γ under the shape of charge level γ (x) of final output and the last one chute inclination anglen(x) identical;
When the iteration optimization module iteration optimization particle position, the speed of particle is updated using formula (15), utilizes formula
(16) position of particle is updated:
In formula,For the position of k+1 moment particle,For the position of k moment particle,For the speed of k+1 moment particle,For the speed of k moment particle, ξ1With ξ2It is the random number in [0,1] range,It is particle in the optimal position of k moment individual
It sets,It is particle in k moment population optimal location.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710003901.XA CN106801114B (en) | 2017-01-04 | 2017-01-04 | A kind of blast furnace material distribution process burden distribution matrix optimization method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710003901.XA CN106801114B (en) | 2017-01-04 | 2017-01-04 | A kind of blast furnace material distribution process burden distribution matrix optimization method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106801114A CN106801114A (en) | 2017-06-06 |
CN106801114B true CN106801114B (en) | 2019-02-26 |
Family
ID=58984287
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710003901.XA Active CN106801114B (en) | 2017-01-04 | 2017-01-04 | A kind of blast furnace material distribution process burden distribution matrix optimization method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106801114B (en) |
Families Citing this family (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107119159B (en) * | 2017-06-14 | 2019-02-22 | 内蒙古科技大学 | It is a kind of that there are the optimization methods of the blast furnace material distribution process burden distribution matrix of integer programming problem |
CN107245540B (en) * | 2017-06-14 | 2019-03-12 | 内蒙古科技大学 | A kind of control strategy of blast furnace material distribution process radial direction thickness of feed layer distribution |
CN107992974B (en) * | 2017-12-13 | 2021-03-05 | 沈阳东博热工科技有限公司 | Optimization method of vacuum heat treatment distribution matrix |
CN108277314B (en) * | 2018-02-27 | 2020-11-20 | 首钢京唐钢铁联合有限责任公司 | Material distribution method and device |
CN111304388B (en) * | 2020-03-27 | 2021-12-17 | 武汉钢铁有限公司 | Method for optimizing blast furnace gas flow distribution by using scanning radar |
CN112226557B (en) * | 2020-09-27 | 2022-08-19 | 北京科技大学 | Blast furnace burden distribution process evaluation method and system, readable storage medium and application thereof |
CN113637819B (en) * | 2021-08-16 | 2022-10-18 | 中冶东方工程技术有限公司 | Blast furnace material distribution method and system based on deep reinforcement learning |
CN114807468B (en) * | 2022-04-02 | 2023-10-24 | 武汉钢铁有限公司 | Blast furnace charge level monitoring-based method for improving blast furnace gas utilization rate |
CN114800828B (en) * | 2022-04-09 | 2024-04-05 | 沈阳建筑大学 | Intelligent optimization method for load distribution of prefabricated part concrete distribution robot |
CN115584370B (en) * | 2022-09-01 | 2023-10-13 | 武安市裕华钢铁有限公司 | Method for improving air permeability of soft melting belt in large-slag-quantity smelting of blast furnace |
CN116694837B (en) * | 2023-08-09 | 2023-10-13 | 山西建龙实业有限公司 | Distribution method for high-proportion bituminous coal injection |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102732659A (en) * | 2011-04-11 | 2012-10-17 | 宝山钢铁股份有限公司 | Burden surface profile control method of blast furnace burden distribution and control system |
CN106011350A (en) * | 2016-07-05 | 2016-10-12 | 东北大学 | Method and system for estimating charge surface shapes in blast furnace charge distribution process |
-
2017
- 2017-01-04 CN CN201710003901.XA patent/CN106801114B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102732659A (en) * | 2011-04-11 | 2012-10-17 | 宝山钢铁股份有限公司 | Burden surface profile control method of blast furnace burden distribution and control system |
CN106011350A (en) * | 2016-07-05 | 2016-10-12 | 东北大学 | Method and system for estimating charge surface shapes in blast furnace charge distribution process |
Non-Patent Citations (1)
Title |
---|
无钟高炉环形布料优化及布料精度;马财生等;《钢铁研究学报》;20161231;第28卷(第12期);第15-20页 |
Also Published As
Publication number | Publication date |
---|---|
CN106801114A (en) | 2017-06-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106801114B (en) | A kind of blast furnace material distribution process burden distribution matrix optimization method and system | |
CN106011350B (en) | A kind of blast furnace material distribution process shape of charge level method of estimation and system | |
CN104531924B (en) | A kind of blast furnace burden distribution Real-time Forecasting System and method | |
CN107723399B (en) | A kind of blast furnace material distribution intelligent monitor system and method for adjustment | |
CN104133945B (en) | A kind of control method of blast furnace material distribution process radial direction ore coke ratio | |
CN104694680B (en) | A kind of radially-arranged control method of blast furnace bed structure | |
CN105399095B (en) | A kind of device and method that active carbon with high specific surface area is prepared for alkali activation method | |
CN103966373B (en) | A kind of blast furnace stable smooth operation without clock distributing process | |
WO2022198914A1 (en) | Blast furnace throat temperature estimation method based on multilayer ore-coke ratio distribution model | |
CN110427715B (en) | Method for predicting furnace hearth thermal state trend based on time sequence and multiple dimensions of blast furnace | |
CN102653801A (en) | Bell-less blast furnace top burden distribution control method established on basis of genetic algorithm | |
CN107622279A (en) | The sorting technique of blast furnace internal state | |
CN107245540B (en) | A kind of control strategy of blast furnace material distribution process radial direction thickness of feed layer distribution | |
Zhou et al. | Grey correlation analysis of factors influencing maldistribution in feeding device of copper flash smelting | |
Li et al. | Analysis on seeding process of pneumatic cylinder-type centralized rapeseed precision metering device | |
CN116694837B (en) | Distribution method for high-proportion bituminous coal injection | |
CN107119159B (en) | It is a kind of that there are the optimization methods of the blast furnace material distribution process burden distribution matrix of integer programming problem | |
CN113656966A (en) | Blast furnace bell-less top on-line material distribution model simulation method | |
JP7077842B2 (en) | Blast furnace charge distribution prediction method, program and computer storage medium | |
Bogus | Investigation of the kinematics of seed movement in the centrifugal distributor of a pneumatic seeder | |
CN112553390A (en) | Coal injection optimization control method based on Gaussian process regression | |
CN110083936A (en) | Consider form parameter is crushed gravel particle generation method | |
Zhao et al. | Study on positive-negative pressure seed metering device for wide-seedling-strip-seeding | |
CN116680927A (en) | Burden accumulation modeling and burden distribution matrix optimization method and system in blast furnace burden distribution process | |
CN117408051A (en) | Method for determining initial air flow distribution state of blast furnace |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |