CN110134006A - Bad hole track optimizing method based on improved multi-objective particle swarm algorithm - Google Patents
Bad hole track optimizing method based on improved multi-objective particle swarm algorithm Download PDFInfo
- Publication number
- CN110134006A CN110134006A CN201910410779.7A CN201910410779A CN110134006A CN 110134006 A CN110134006 A CN 110134006A CN 201910410779 A CN201910410779 A CN 201910410779A CN 110134006 A CN110134006 A CN 110134006A
- Authority
- CN
- China
- Prior art keywords
- particle
- formula
- algorithm
- optimal
- sinφ
- 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.)
- Granted
Links
- 239000002245 particle Substances 0.000 title claims abstract description 133
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 73
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000005457 optimization Methods 0.000 claims abstract description 49
- NUFBIAUZAMHTSP-UHFFFAOYSA-N 3-(n-morpholino)-2-hydroxypropanesulfonic acid Chemical compound OS(=O)(=O)CC(O)CN1CCOCC1 NUFBIAUZAMHTSP-UHFFFAOYSA-N 0.000 claims abstract description 35
- 238000005259 measurement Methods 0.000 claims abstract description 31
- 230000035772 mutation Effects 0.000 claims abstract description 20
- 230000001133 acceleration Effects 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000013461 design Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 241000208340 Araliaceae Species 0.000 claims 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 claims 1
- 235000003140 Panax quinquefolius Nutrition 0.000 claims 1
- 235000008434 ginseng Nutrition 0.000 claims 1
- 238000005553 drilling Methods 0.000 abstract description 24
- 238000011161 development Methods 0.000 description 4
- 230000007547 defect Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 241001074085 Scophthalmus aquosus Species 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 239000004744 fabric Substances 0.000 description 2
- 230000002068 genetic effect Effects 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 241000269793 Cryothenia peninsulae Species 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000001174 ascending effect Effects 0.000 description 1
- 238000002224 dissection Methods 0.000 description 1
- 230000008717 functional decline Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000004321 preservation Methods 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/0205—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric not using a model or a simulator of the controlled system
- G05B13/024—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric not using a model or a simulator of the controlled system in which a parameter or coefficient is automatically adjusted to optimise the performance
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Feedback Control In General (AREA)
Abstract
The parameter of multi-objective particle swarm algorithm MOPSO, (2) initialization population is arranged in bad hole track optimizing method based on improved multi-objective particle swarm algorithm, (1)；(3) calculating target function value, (4) update the position and speed of every generation particle；(5) mutation operation is carried out to particle；(6) target function value of each particle in population is calculated；(7) it is algorithm from the process for starting to iterate to current optimal location that more new individual is optimal, and (8) are ranked up non-dominant collection nd, and (9) carry out descending arrangement according to target function value to the noninferior solution in external archive in MOPSO；(10) method for taking truncation deletes subsequent extra individual；(11) global optimum；(12) optimal solution set of algorithm optimization is obtained, as well track actual measurement length, practical control torque reaches relatively optimal；The present invention realizes the multiple target well track parameter optimization under the conditions of actual well drilled, improves drilling success, reduces drilling cost and establishes theoretical decision basis.
Description
Technical field
The present invention relates to well track optimisation technique fields, in particular to answering based on improved multi-objective particle swarm algorithm
Miscellaneous well track optimization method.
Background technique
As more and more oil-gas explorations have turned to the areas such as deep-sea, desert from inland, in addition unconventional, deep water, depth
The quantity of the oil gas fields such as layer, polar region constantly increases, therewith adaptable drilling technology, with boring data acquisition technology, log data
There has also been significant progresses for integrated interpretation method and intelligent optimization algorithm.It is required in addition, being arranged during oil field development to well pattern
It increasingly improves, scanning and anti-collision are increasingly valued by people between well.Secondly, the production in oil field in order to further increase
Amount, the development of thin oil reservoir has come into a new stage, higher to the position accuracy demand of well track, develop to
More far-reaching target development.Thus, in drilling process, realize that the optimization of well track and accurate control just seem very heavy
It wants.Wherein, realize well track it is effective, real-time, rapidly optimization be realize well track accurately control, in raising target rate and
Reduce the premise of drilling risk.
Well track optimization is on the basis of determining technology path, arrangement and method for construction and pithead position before construction, to determine
Meet the well track optimization object function of technique requirement, considers under the constraint conditions such as drilling tool and stratum, preferably tool out
The well track parameters such as face angle, inclination angle, radius of curvature, deflecting point range, with reach improve drilling success, save drilling well at
This purpose.And the optimization of existing well track optimizes the result of three dimensional hole trajectory often mainly based on single object optimization
With actual well drilled demand there is biggish deviation, if the patent disclosure of Patent No. 201710132117.9 is " based on quick
The bad hole track optimizing method of adaptive quantum genetic algorithm ", which only optimizes well track length, can not
Really instruct actual well drilled process, therefore traditional single object optimization is no longer satisfied obtaining for well track parameter in practice
It takes, it is also necessary to consider multiple optimization aims, meet oil reservoir to more accurately bore.
Summary of the invention
In order to overcome the defects of the prior art described above, the purpose of the present invention is to provide be based on improved multi-objective particle swarm
The bad hole track optimizing method of algorithm, using improved multi-objective particle swarm algorithm (Multi-objective
Particle Swarm Optimization Algorithm), it is optimal and global optimum position selection mode to individual to change
Into, apply disturbance mutation operator to particle, thus avoid falling into local optimum, the generation of external collection, non-dominant collection selection side
Formula etc., which improves particle swarm algorithm, to optimize multi-objective problem.
To achieve the above object, the technical scheme is that being achieved:
Bad hole track optimizing method based on improved multi-objective particle swarm algorithm, comprising the following steps:
(1) parameter of multi-objective particle swarm algorithm MOPSO is set, maximum value and maximum value including dynamic inertia weight,
Accelerated factor, population scale, maximum number of iterations GEN.
(2) initialization population, it is speed, initial position including particle, the position for meeting constraint condition, external archive, a
Body is optimal and global optimum；Azimuth, inclination angle, dogleg angle, radius of curvature, each well section actual measurement length, practical vertical depth, reality
Border controls torque, casing length；Wherein the initial position of particle is randomly generated in population, and the initial position of population is assigned to accord with
The initial position of constraint condition is closed, individual is optimal and global optimum position is set as particle itself.
(3) when the initial position of population meets constraint condition, calculating target function value, constraint condition includes for practical
The value range of independent variable in bad hole track optimizing problem, casing length range, target vertical well depth nonnegativity restrictions model
It encloses, dog-leg angular region and nonnegativity restrictions in stratum, nonnegativity restrictions refers to actual measurement depth and vertical depth cannot be
It is negative, select 12 geometric parameters of well track to carry out preferably, realizing Bi-objective actual measurement depth TMD (True to be optimized
Measurement Depth, TMD), practical control torque TCT (True Control Torque, TCT) reach relatively optimal,
Geometric parameter includes kickoff point (KOP) depth, hole angle and azimuth；
Wherein, objective function may be defined as:
Obj_function=min { TMD, TCT }
Wherein: TMD=Dkop+D1+D2+D3+D4+D5+HD
TCT=T1+T2+T3+T4+T5+T6+T7
s.t.xmin≤x≤xmax (1)
TVDmin≤TVD≤TVDmin
Cimin≤Ci≤Cimax(i=1,2,3)
Ds> 0 (s=1,2,3,4,5)
In formula (1)That is solution space R12By 12 dimension decision vector X
Composition, i.e., parameter to be optimized；TMD, TCT are optimization object function, and unit is respectively ft, Nft；I is the number of segment of sleeve design,
TVDmax, TVDminThe respectively bound of the practical vertical depth of well track.
Each section of calculation formula of well track is defined as:
D2=(Dd-Dkop-D1×(sinφ1-sinφ0)/(φ1-φ0))/cos(φ1) (3)
D4=(DB-Dd-D3×(sinφ2-sinφ1)/(φ2-φ1))/cos(φ2) (5)
Total actual measurement length are as follows:
TMD=Dk+D1+D2+D3+D4+D5+HD (7)
Wherein D1, D5 are the first, second increasing hole angle section, θ1~θ2: inclination angle at two measuring points；Two measuring point prescription positions
Angle.
Increasing hole angle section actual measurement profile length increment are as follows:
In formula (8), r is radius of curvature,Formula (8) curved section exists
Incremental computations under three-dimensional coordinate may be defined as:
Each well section stress F1~F7 calculates as follows:
F7=0 (13)
F6=F7+BwHDcos φ3=BwHDcos φ3 (14)
F5=F6+BwD5(sinφ3-sinφ2)/(φ3-φ2) (15)
F4=F5+BwD4cosφ2 (16)
F3=F4+BwD3(sinφ2-sinφ1)/(φ2-φ1) (17)
F2=F3+BwD2cosφ1 (18)
F1=F2+BwD1(sinφ1-sinφ0)/(φ1-φ0) (19)
The corresponding torque T1~T7 of F1~F7 is calculated, calculating formula is as follows:
T1=μ rwDksinφ0 (20)
T3=μ rwD2sinφ1 (22)
T5=μ rwD4sinφ2 (24)
T7=μ rwHDsin φ3 (26)
Total practical control torque are as follows:
TCT=T1+T2+T3+T4+T5+T6+T7 (27)
Each section of casing length calculating formula is as follows:
C1=Dk+D1*sin(θ1)/θ1 (28)
Each section of practical vertical depth calculates as follows:
TVDdrop=D3* (sin (rad* θ2)-sin(rad*θ1))/((θ2-θ1)*rad) (33)
Total practical vertical depth are as follows:
In formula (2)~(36), the meaning and value range of each parameter are as shown in table 2.Each well section is calculated by formula (2)~(25)
Actual measurement length D1~D5, the actual loading F1~F7 of track and corresponding practical control torque T1~T7.Formula (1)~(36)
Calculating target function TMD, TCT simultaneously save the optimal solution for meeting constraint condition, and the first generation is met the optimal feasible of constraint condition
Solution, optimal objective function value are temporary, as globally optimal solution, global objective function, with the increase of the number of iterations, according to formula
(1) target fitness value TMD and TCT is calculated, contemporary optimal solution is recorded.
(4) position and speed that every generation particle is updated according to formula (37) and formula (38), if particle exceeds in the process
Boundary constraint is then taken on boundary；Shown in the speed update mode such as formula (37) of particle.
In formula (37), vpjIt (t) is the speed of t moment particle p jth dimension；psowc1And psowc2It is positive acceleration constant；
r1j(t),r2j(t) random number to be generated in section [0,1]；Yp indicates the optimal location that particle p is currently arrived；To work as
The optimal location that all particles are found in preceding population；Psow is the inertia weight introduced.
The position of more new particle by the way of matrix, shown in the speed update mode such as formula (38) of particle.
xp(t+1)=xp(t)+vp(t+1) (38)
In formula (38), xp(t) position for being t moment particle p, xpjIt (t) is in the position that t moment particle p jth is tieed up, and xi
~U (xmin, xmax)。
(5) mutation operation is carried out to particle, is acted in MOPSO using mutation operator, to guide the flight of particle, is mentioned
The ability that population jumps out local optimum is risen, part and global search dynamics are reinforced.Algorithm search for a period of time after, reduce participate in
The individual amount of variation is carried out part exploitation and is disturbed using mutation operator to particle application, prevents particle from falling into local optimum；It is right
It makes a variation in the position of particle p, particle are as follows:
xp=xp+16*varsig*(1-randr)*vp(39) in formula (39), varsig=± 1, after indicating particle variations
It is whether identical as the original direction of motion；16 be the coefficient that particle can be made to jump out local optimum position；vpFor mutation probability.
In formula (40), ite is the current algebra of algorithm, and GEN is algorithm maximum number of iterations.
For each particle in population, a random number randr in section [0,1] is generated, as randr <
When randp, mutation operation is carried out to particle, otherwise without variation；In addition, it is fixed to need will exceed when making a variation to particle
The particle on adopted domain boundary is defined on boundary.
(6) for the particle for meeting constraint condition, each particle in population is calculated according to each target of each particle
Target function value；For being unsatisfactory for the particle of constraint condition, if be unsatisfactory for constraint condition continuous four times, according to step (4),
(5) speed and the position for redistributing particle, to particle variations and boundary constraint.
(7) more new individual optimal algorithm, from starting to iterate to current optimal location, if current location x dominates its individual pole
It is worth position xp, then it is updated to current location x.
(8) non-dominant collection nd is ranked up, after carrying out local optimum update to each particle in population, saves algorithm
Noninferior solution in iteration searches for optimum individual using non-dominant set algorithm, rather than the selection of dominant set is dominated using multiple target and closed
It is quicksort.
(9) with iterations going on, then right compared with every group of noninferior solution being carried out one by one with the solution of current non-dominant collection
Noninferior solution in MOPSO in external archive carries out descending arrangement according to target function value, reduces the computation complexity of algorithm.
(10) dynamic control is carried out to external archive scale using dynamic crowding distance method, collection ex, which exceeds, when outside sets
Range when, the method for taking truncation deletes subsequent extra individual.
(11) global optimum selects a particle as the overall situation after the foremost portion in the external archive after descending sort
Optimal, guidance population, which is constantly searched for, more preferably to be solved, and guarantees the distributivity of algorithm.
(12) ite=ite+1 goes to step (3) and continues if the number of iterations ite < GEN；Otherwise output is external
The particle of concentration obtains the optimal solution set of algorithm optimization, and as well track actual measurement length, practical control torque reaches phase
To optimal.
The present invention constructs external collection by using dominance relation, and it is non-dominant from starting to find till now to save population
Solution, bootstrap algorithm are quickly approached to the front end Pareto, maintain the distributivity of feasible solution.And mutation operator is introduced to grain
Son applies disturbance, prevents algorithm from falling into local optimum, improves individual optimal and global optimum position selection mode, devises more
MOPSO is used for actual measurement depth TMD, the reality control torque of complex three-dimensional well track by intended particle group algorithm MOPSO
Multiple well track optimized parameter kickoff point (KOP) depth, hole angle and azimuthal preferred, the practical brill of realization are completed in the optimization of TCT
Multiple target well track parameter optimization under the conditions of well, improves drilling success, reduces drilling cost.
Using MOPSO realize bad hole track optimizing problem solving the experimental results showed that, TMD, TCT result of optimization
Traditional well track single object optimization is overcome in boring in fact and actual well drilled demand haves the defects that relatively large deviation, is met
The acquisition of the well track parameter under multiple goal conditions under the conditions of actual well drilled, by this method application wisdom drilling process
Well track optimization, has substantially met the demand of well track actual parameter optimization, has been more conducive to drilling well staff and is boring
The decision of well bore track process, to more accurately reach oil reservoir.It is further by the research to well track MOPSO
It realizes in accurate control, correction, anti-collision and the raising for boring interactive well track multiobjective Dynamic Optimization and well track
Target rate realizes the multiple target well track parameter optimization under the conditions of actual well drilled, improves drilling success, reduces drilling cost and establish
Theorem opinion decision basis.
Detailed description of the invention
Fig. 1 is the vertical cross-section of bad hole track.
Fig. 2 is the schematic three dimensional views of D1, D5 increasing hole angle section.
Fig. 3 is corresponding torque T3, T5 schematic diagram of D2, D4 steady tilted section.
Fig. 4 is corresponding torque T2, T6 schematic diagram of D1 and D5 increasing hole angle section.
Fig. 5 is the corresponding torque T4 schematic diagram of D3 drop angle section.
Fig. 6 is the optimum results figure that three dimensional hole trajectory is realized using MOPSO.
Subordinate list explanation
Table 4 be the present invention implement after preferably well track parameter compared with other intelligent algorithm single object optimization results.
Specific embodiment
The embodiment that the invention will now be described in detail with reference to the accompanying drawings.
Bad hole track optimizing method based on improved multi-objective particle swarm algorithm, comprising the following steps:
(1) parameter of multi-objective particle swarm algorithm MOPSO is set, maximum value and maximum value including dynamic inertia weight,
Accelerated factor, population scale, maximum number of iterations GEN.
Constraint condition and independent variable restrained boundary condition are as shown in table 2.
The parameter setting of 1 MOPSO of table
In table 1, POP is population scale, and exPOP is external archive scale, generally takes POP=exPOP；GEN changes for maximum
Generation number；In MOPSO, psoc1 and psoc2 are accelerated factor；psowmax、psowminFor the maximum value and minimum of Inertia Weight
Value.
2 well track variable bound boundary of table and constraint condition
(2) initialization population, it is speed, initial position including particle, the position for meeting constraint condition, external archive, a
Body is optimal and global optimum；Azimuth, inclination angle, dogleg angle, radius of curvature, each well section actual measurement length, practical vertical depth, reality
Border controls torque, casing length；Wherein the initial position of particle is randomly generated in population, and the initial position of population is assigned to accord with
The initial position of constraint condition is closed, individual is optimal and global optimum position is set as particle itself.
(3) when the initial position of population meets constraint condition, calculating target function value, constraint condition includes for practical
The value range of independent variable in well track optimization problem, casing length range, target vertical well depth nonnegativity restrictions range,
Dog-leg angular region and nonnegativity restrictions in layer, nonnegativity restrictions refer to that actual measurement depth and vertical depth cannot be negative, and select
12 geometric parameters of well track are selected to carry out preferably, realizing Bi-objective actual measurement depth TMD (True to be optimized
Measurement Depth, TMD), practical control torque TCT (True Control Torque, TCT) reach relatively optimal,
Geometric parameter includes kickoff point (KOP) depth, hole angle and azimuth；
Wherein, the vertical cross-section of bad hole track to be optimized is as shown in Figure 1.
Wherein, objective function may be defined as:
In formula (1)That is solution space R12It is made of 12 dimension decision vector X,
Parameter i.e. to be optimized；TMD, TCT are optimization object function, and unit is respectively ft, N*ft, and i is the number of segment of sleeve design.Wherein,
D1:First build-up section first segment inclination section；The positive dissection D3:Drop-off drop angle of D2:Tangent section
Section；D4:Hold section steady tilted section；D5:Second build-up section second segment increasing hole angle section；HD:Horizontal
Section horizontal segment well bore length, ft；TVDmax, TVDminThe respectively bound of the practical vertical depth of well track.
In Fig. 1, each section of calculation formula of well track is defined as:
D2=(Dd-Dkop-D1×(sinφ1-sinφ0)/(φ1-φ0))/cos(φ1) (43)
D4=(DB-Dd-D3×(sinφ2-sinφ1)/(φ2-φ1))/cos(φ2) (45)
Total actual measurement length are as follows:
TMD=Dk+D1+D2+D3+D4+D5+HD (47)
Wherein D1, D5 are the first, second increasing hole angle sections, and schematic three dimensional views are as shown in Figure 2.θ1~θ2: inclination angle at two measuring points；Two measuring point azimuthals.
In Fig. 2, increasing hole angle section actual measurement profile length increment are as follows:
In formula (48), r is radius of curvature,
Incremental computations of formula (48) curved section under three-dimensional coordinate may be defined as:
In Fig. 3~Fig. 5, P1, P2 are two observation points；E1, e2 are the unit vector on pit shaft direction；F1 is directional well bottom
The axial force in portion, N；F2 is the axial force at the top of directional well, N；β is total angle change；B is buoyancy coefficient, B=0.7；μ is
Coefficient of friction, μ=0.3；W is unit drilling tool weight w=0.3kN/ft；R is drilling rod radius r=10mm.
Each well section stress F1~F7 calculates as follows:
F7=0 (53)
F6=F7+BwHDcos φ3=BwHDcos φ3 (54)
F5=F6+BwD5(sinφ3-sinφ2)/(φ3-φ2) (55)
F4=F5+BwD4cosφ2 (56)
F3=F4+BwD3(sinφ2-sinφ1)/(φ2-φ1) (57)
F2=F3+BwD2cosφ1 (58)
F1=F2+BwD1(sinφ1-sinφ0)/(φ1-φ0) (59)
The corresponding torque T1~T7 of F1~F7 is calculated, calculating formula is as follows:
T1=μ rwDksinφ0 (60)
T3=μ rwD2sinφ1 (62)
T5=μ rwD4sinφ2 (64)
T7=μ rwHDsin φ3 (66)
Total practical control torque are as follows:
TCT=T1+T2+T3+T4+T5+T6+T7 (67)
Each section of casing length calculating formula is as follows:
C1=Dk+D1*sin(θ1)/θ1 (68)
Each section of practical vertical depth calculates as follows:
TVDdrop=D3* (sin (rad* θ2)-sin(rad*θ1))/((θ2-θ1)*rad) (73)
Total practical vertical depth are as follows:
In formula (42)~(76), the meaning and value range of each parameter are as shown in table 2.Each well is calculated by formula (42)~(75)
Actual measurement length D1~D5, the actual loading F1~F7 of section track and corresponding practical control torque T1~T7.Formula (1)~
(36) calculating target function TMD, TCT and preservation meet the optimal solution of constraint condition, and the first generation is met the optimal of constraint condition
Feasible solution, optimal objective function value are temporary, as globally optimal solution, global objective function, with the increase of the number of iterations, according to
Formula (1) calculates target fitness value TMD and TCT, recording individual optimal solution xp, its global optimum xg, individual optimal objective function
Yp, global optimum objective function yg；If meeting constraint condition, globally optimal solution xg and the overall situation are saved most according to dominance relation
Excellent target function value yg；Otherwise, it remains unchanged.Enable ite=ite+1；
(4) position and speed that every generation particle is updated according to formula (77) and formula (78), if particle exceeds in the process
Boundary constraint is then taken on boundary；According to the application study to improved MOPSO, the part that the size of speed affects algorithm is sought
Excellent ability, biggish speed corresponds to stronger global optimizing ability, and lesser speed is then conducive to particle local optimal searching.?
In the present invention, MOPSO keeps diversity and search capability using the inertia weight psow that dynamic changes convenient for population, increases algorithm
The ability of later development, shown in the speed update mode such as formula (77) of particle.
In formula (77), vpjIt (t) is the speed of t moment particle p jth dimension；Psowc1 and psowc2 is positive acceleration constant；
r1j(t),r2j(t) random number to be generated in section [0,1]；Yp indicates the optimal location that particle p is currently arrived；To work as
The optimal location that all particles are found in preceding population；Psow is the inertia weight introduced；
The position of more new particle is used for by the way of matrix, shown in the speed update mode such as formula (78) of particle.
xp(t+1)=xp(t)+vp(t+1) (78)
In formula (78), xp(t) position for being t moment particle p, xpjIt (t) is in the position that t moment particle p jth is tieed up, and xi
~U (xmin, xmax)。
(5) mutation operation is carried out to particle.Particle swarm algorithm and genetic algorithm are all to apply more extensive population algorithm,
But GA is based on intersecting and variation generates new individual, and particle swarm algorithm is by personal best particle and global optimum position
Study generates.The advantage and disadvantage of both algorithms are directed to, although such as GA possesses stronger ability of searching optimum, but its convergence speed
Degree is larger compared with slow and computation complexity；Particle swarm algorithm possesses faster convergence rate, but when encountering multi-peak the problem of, calculates
Method easily falls into locally optimal solution, and the hybrid algorithm for then occurring combining the two is to improve respective disadvantage.
It is successfully applied in single goal particle swarm algorithm in view of mutation operation, the present invention is acted on using mutation operator
In MOPSO, to guide the flight of population, the ability that population jumps out local optimum is promoted, part and global search dynamics are reinforced.
In addition, algorithm search for a period of time after, it is desirable to reduce participate in variation individual amount, carry out part exploitation use mutation operator
Particle is applied and is disturbed, prevents particle from falling into local optimum；For particle p, the position of particle makes a variation are as follows:
xp=xp+16*va.rsig(-randr)*vp (79)
In formula (79), varsig=± 1, indicate particle variations after it is whether identical as the original direction of motion；16 be that can make
Particle jumps out a coefficient of local optimum position；vpFor mutation probability.
In formula (80), ite is the current algebra of algorithm, and GEN is algorithm maximum number of iterations.
For each particle in population, a random number randr in section [0,1] is generated, as randr <
When randp, mutation operation is carried out to particle, otherwise without variation；In addition, it is fixed to need will exceed when making a variation to particle
The particle on adopted domain boundary is defined on boundary.
(6) it for the particle for meeting constraint condition, needs to calculate each grain in population according to each target of each particle
The target function value of son；For being unsatisfactory for the particle of constraint condition, if be unsatisfactory for constraint condition continuous four times, need according to step
Suddenly speed and the position of particle are redistributed in (4), (5), to particle variations and boundary constraint.
(7) it is algorithm from the process for starting to iterate to current optimal location that more new individual is optimal, if current location x dominates it
Individual extreme value place xp, then it is updated to current location x.
(8) non-dominant collection nd is ranked up, after carrying out local optimum update to each particle in population, needs to save
Noninferior solution in algorithm iteration.In order to improve the efficiency of MOPSO, the computation complexity of MOPSO is reduced, this algorithm is using non-dominant
Set algorithm searches for optimum individual, rather than the selection of dominant set uses multiple target dominance relation quicksort.
The specific method is as follows:
1. the individual apb in selected population ap usually chooses an individual, apb is deleted from ap；
2. comparing other individuals and individual apb in population.Population is divided into two classes at this time, and one piece is to be dominated by individual apb
Individual composition, another part be dominate individual apb or with individual apb is incoherent individual forms；
3. if individual apb is not dominated by individuals other in population, that is to say, that individual apb is non-domination solution, by individual apb
It is put into non-dominant collection nd；Otherwise it is not put into；
4. first kind individual is deleted from ap, if 1. ap non-empty, goes to；
5. nd is exactly required non-dominant collection when ap is empty.
This non-dominant collection of method construct, every time circulation start when population be all the domination individual apb obtained last time or
What person and the incoherent individual of individual apb formed, rather than entire population, the range shorter when carrying out individual relatively improve
The speed of service.
(9) external collection ex is updated.With iterations going on, the solution of every group of noninferior solution and current non-dominant collection is carried out one by one
Compare, descending arrangement then is carried out according to target function value to the noninferior solution in external archive in MOPSO, reduces the calculating of algorithm
Complexity.
Update method are as follows:
1. the individual in non-dominant collection nd is put into outer when outside collection is empty, that is, when algorithm just brings into operation
Portion collects in ex；
2. choose some individual p in non-dominant collection nd when outside collection is not empty, individual in relatively external collection ex and
This individual p cannot be put it into external collection ex if individual p is dominated by the individual in external collection ex；Otherwise by individual p
It is put into external collection ex, and deletes by those of individual p domination individual.
3. circulation finishes until comparing.
(10) dynamic control is carried out to external archive scale using dynamic crowding distance method.In order to maintain point of external archive
Cloth, improves the distributivity and algorithm ability of searching optimum of noninferior solution, and the efficiency for improving algorithm needs to keep external archive ex dilute
The particle at place is dredged, the particle intensively located is reduced, needs to recalculate crowding distance, controls the external scale collected.When outside collects ex
When beyond the range set, the method for taking truncation deletes subsequent extra individual.
Crowding distance calculation method is as follows:
1. initializing set exPOP individual crowding distance dis=0；
2. ex is carried out ascending sort according to the value of the yd objective function；
3. setting infinitely great for two individual crowding distances of first and last after sequence；
4. the crowding distance of other individuals p is calculated by formula (80).
In formula (80), ex (p, j) is j-th of target function value of p-th of particle in ex.
5. changing the value of yd, go to 2., until traversing all target dimensions.
(11) global optimum selects a particle as the overall situation after the foremost portion in the external archive after descending sort
It is optimal, it guides population and constantly searches for and more preferably solve, ensure that the distributivity of algorithm.MOPSO when solving multi-objective problem,
Each iteration generates one group of noninferior solution.The individual that global optimum is set as from 10% before external collection is chosen.It is this complete
The selection mode of office's optimal location has different global optimum positions for particle each in population, ensure that point of algorithm
Cloth.
(12) ite=ite+1 goes to step (3) and continues if the number of iterations ite < GEN；Otherwise output is external
The particle of concentration obtains the optimal solution set of algorithm optimization, and as well track actual measurement length, practical control torque reaches phase
To optimal.
External archive ex is exactly the best result that algorithm obtains, and exports the result.Export global optimum position xgWith it is optimal
Objective function yg.Exportyg=(TMD, TCT).It is realized using MOPSO complicated
The result of well track optimization is compared with other several intelligent optimization algorithm single object optimization results, as shown in table 3.Using
The preferred result that MOPSO optimizes the simulation result of TMD, TCT is as shown in Figure 6.
Three groups of optimal solutions that table 3 is chosen
Table 3 is the three groups of global optimum's objective function solutions chosen.By three groups of globally optimal solutions of comparison it can be found that the 2nd
Actual measurement length variation in group solution is little, but but than the 1st group solution is much higher for practical control torque, the reality in the 3rd group of solution
Border measurement length becomes smaller, but but than the 2nd group solution is high more for practical control torque, it follows that well-drilling borehole track optimizing
It influences each other between parameters in the process, mutually restrict.So working as actual measurement in actual drilling process
When length meets actual requirement in a certain range, selects the 1st group of solution as optimal solution, that is, meet wanting for actual measurement length
It asks, and practical control torque can be effectively reduced.
In order to which further description improves the optimization performance of MOPSO algorithm, the present invention is improved to the simulation result of MOPSO
With classical with GA (Shokir et al.al.2004), NPSO (Amin Atashnezhad2014) and PSO (Shokiret
Al.2004 it) compares and analyzes, specific data are shown in Table 4.
Table 4 is using MOPSO optimization well track optimum results compared with other several algorithm single object optimization results
As shown in Table 4, in the optimization of complex three-dimensional well track, the optimization of multiple target TMD and TCT are realized using MOPSO,
Its optimum results is preferable, remain substantially single object optimization as a result, meeting the demand of multiple-objection optimization in practice.
The optimization of three dimensional hole trajectory is realized using MOPSO, the optimum results of TMD, TCT are as shown in Figure 6.Fig. 6 is used
MOPSO realizes the optimization of complex three-dimensional well track TMD, TCT, and abscissa is optimal well track actual measurement length TMD, indulges
Coordinate is the practical control torque of optimal well track.It will be appreciated from fig. 6 that in two objective optimisation problems of well track parameter
The forward position Pareto, it is shown that the value of an objective function declines with the rising of the value of another objective function.The side Pareto
Boundary is not meant to that minimum torque corresponds to longest well track or shortest well track corresponds to maximum torque.Two
Be between a objective function it is nonlinear, optimal solution is on a declining curve.
The present invention constructs external collection by using dominance relation, and it is non-dominant from starting to find till now to save population
Solution, bootstrap algorithm are quickly approached to the front end Pareto, maintain the distributivity of feasible solution.And mutation operator is introduced to grain
Son applies disturbance, prevents algorithm from falling into local optimum, improves individual optimal and global optimum position selection mode, devises more
MOPSO is used for actual measurement depth TMD, the reality control torque of complex three-dimensional well track by intended particle group algorithm MOPSO
Multiple well track optimized parameter kickoff point (KOP) depth, hole angle and azimuthal preferred, the practical brill of realization are completed in the optimization of TCT
Multiple target well track parameter optimization under the conditions of well, improves drilling success, reduces drilling cost.
Using MOPSO realize bad hole track optimizing problem solving the experimental results showed that, TMD, TCT result of optimization
Traditional well track single object optimization is overcome in boring in fact and actual well drilled demand haves the defects that relatively large deviation, is met
The acquisition of the well track parameter under multiple goal conditions under the conditions of actual well drilled, by this method application wisdom drilling process
Well track optimization, has substantially met the demand of well track actual parameter optimization, has been more conducive to drilling well staff and is boring
The decision of well bore track process, to more accurately reach oil reservoir.It is further by the research to well track MOPSO
It realizes in accurate control, correction, anti-collision and the raising for boring interactive well track multiobjective Dynamic Optimization and well track
Target rate realizes the multiple target well track parameter optimization under the conditions of actual well drilled, improves drilling success, reduces drilling cost and establish
Theorem opinion decision basis.
Claims (2)
1. the bad hole track optimizing method based on improved multi-objective particle swarm algorithm, which is characterized in that including following step
It is rapid:
(1) parameter of multi-objective particle swarm algorithm MOPSO, maximum value and maximum value including dynamic inertia weight, acceleration are set
The factor, population scale, maximum number of iterations GEN；
(2) initialization population, speed, initial position including particle, the position for meeting constraint condition, external archive, individual are most
Excellent and global optimum；Azimuth, inclination angle, dogleg angle, radius of curvature, each well section actual measurement length, practical vertical depth, practical control
Torque processed, casing length；Wherein the initial position of particle is randomly generated in population, and the initial position of population is assigned to meet about
The initial position of beam condition, individual is optimal and global optimum position is set as particle itself；
(3) when the initial position of population meets constraint condition, calculating target function value, constraint condition includes being directed to practical wellbore
The value range of independent variable in track optimizing problem, casing length range, target vertical well depth nonnegativity restrictions range, in stratum
Dog-leg angular region and nonnegativity restrictions, nonnegativity restrictions refers to that actual measurement depth and vertical depth cannot be negative, and selects well
12 geometric parameters in eye track carry out preferably, realizing that Bi-objective actual measurement depth TMD to be optimized, practical control torque TCT reach
To relatively optimal, geometric parameter includes kickoff point (KOP) depth, hole angle and azimuth；
(4) position and speed that every generation particle is updated according to formula (37) and formula (38), if particle exceeds boundary in the process,
Then take boundary constraint；Shown in the speed update mode such as formula (37) of particle,
In formula (37), vpjIt (t) is the speed of t moment particle p jth dimension；psowc1And psowc2It is positive acceleration constant；r1j
(t), r2j(t) random number to be generated in section [0,1]；ypIndicate the optimal location that particle p is currently arrived；It is current
The optimal location that all particles are found in population；Psow is the inertia weight introduced；
The position of more new particle by the way of matrix, shown in the speed update mode such as formula (38) of particle；
xp(t+1)=xp(t)+vp(t+1) (38)
In formula (38), xp(t) position for being t moment particle p, xpjIt (t) is in the position that t moment particle p jth is tieed up, and xi~U
(xmin, xmax)；
(5) mutation operation is carried out to particle, is acted in MOPSO using mutation operator, to guide the flight of particle, promote kind
Group jumps out the ability of local optimum, reinforces part and global search dynamics；In addition, algorithm search for a period of time after, reduce ginseng
With the individual amount of variation, carries out part exploitation and particle application is disturbed using mutation operator, prevent particle from falling into local optimum；
For particle p, the position of particle makes a variation are as follows:
xp=xp+16*varsig*(1-randr)*vp (39)
In formula (39), varsig=± 1, indicate particle variations after it is whether identical as the original direction of motion；16 be that can make particle
Jump out a coefficient of local optimum position；vpFor mutation probability；
In formula (40), ite is the current algebra of algorithm, and GEN is algorithm maximum number of iterations；
For each particle in population, a random number randr in section [0,1] is generated, as randr < randp
When, mutation operation is carried out to particle, otherwise without variation；In addition, will exceed domain boundary when making a variation to particle
Particle is defined on boundary；
(6) for the particle for meeting constraint condition, the target of each particle in population is calculated according to each target of each particle
Functional value；For being unsatisfactory for the particle of constraint condition, if be unsatisfactory for constraint condition continuous four times, according to step (4), step
(5) speed and the position for redistributing particle, to particle variations and boundary constraint；
(7) more new individual optimal algorithm is from the process for starting to iterate to current optimal location, if current location x dominates its individual pole
It is worth position xp, then it is updated to current location x；
(8) non-dominant collection nd is ranked up, after carrying out local optimum update to each particle in population, saves algorithm iteration
In noninferior solution, using non-dominant set algorithm search for optimum individual, rather than the selection of dominant set using multiple target dominance relation it is fast
Speed sequence；
(9) with iterations going on, compared with every group of noninferior solution being carried out one by one with the solution of current non-dominant collection, then to MOPSO
Noninferior solution in middle external archive carries out descending arrangement according to target function value, reduces the computation complexity of algorithm；
(10) dynamic control is carried out to external archive scale using dynamic crowding distance method, when external collection ex exceeds the model set
When enclosing, the method for taking truncation deletes subsequent extra individual；
(11) it selects a particle as global optimum after the foremost portion in the external archive after descending sort, guides particle
Group's constantly search more preferably solves, and guarantees the distributivity of algorithm；
(12) ite=ite+1 goes to step (3) and continues if the number of iterations ite < GEN；Otherwise the external collection of output
In particle, obtain the optimal solution set of algorithm optimization, as well track actual measurement length, practical control torque reaches opposite
It is optimal.
2. the bad hole track optimizing method according to claim 1 based on improved multi-objective particle swarm algorithm,
It is characterized in that,
In step (3), objective function be may be defined as:
In formula (1)That is solution space R12It is made of 12 dimension decision vector X,
Parameter i.e. to be optimized；TMD, TCT are optimization object function, and unit is respectively ft, Nft；I is the number of segment of sleeve design,
TVDmax, TVDminThe respectively bound of the practical vertical depth of well track；
Each section of calculation formula of well track is defined as:
D2=(Dd-Dkop-D1×(sinφ1-sinφ0)/(φ1-φ0))/cos(φ1) (3)
D4=(DB-Dd-D3×(sinφ2-sinφ1)/(φ2-φ1))/cos(φ2) (5)
Total actual measurement length are as follows:
TMD=Dk+D1+D2+D3+D4+D5+HD (7)
Wherein D1, D5 are the first, second increasing hole angle section, θ1~θ2: inclination angle at two measuring points；Two measuring point azimuthals；
Increasing hole angle section actual measurement profile length increment are as follows:
In formula (8), r is radius of curvature,
Incremental computations of formula (8) curved section under three-dimensional coordinate may be defined as:
Each well section stress F1~F7 calculates as follows:
F7=0 (13)
F6=F7+BwHDcos φ3=BwHDcos φ3 (14)
F5=F6+BwD5(sinφ3-sinφ2)/(φ3-φ2) (15)
F4=F5+BwD4cosφ2 (16)
F3=F4+BwD3(sinφ2-sinφ1)/(φ2-φ1) (17)
F2=F3+BwD2cosφ1 (18)
F1=F2+BwD1(sinφ1-sinφ0)/(φ1-φ0) (19)
The corresponding torque T1~T7 of F1~F7 is calculated, calculating formula is as follows:
T1=μ rwDksinφ0 (20)
T3=μ rwD2sinφ1 (22)
T5=μ rwD4sinφ2 (24)
T7=μ rwHDsin φ3 (26)
Total practical control torque are as follows:
TCT=T1+T2+T3+T4+T5+T6+T7 (27)
Each section of casing length calculating formula is as follows:
C1=Dk+D1*sin(θ1)/θ1 (28)
Each section of practical vertical depth calculates as follows:
TVDdrop=D3* (sin (rad* θ2)-sin(rad*θ1))/((θ2-θ1)*rad) (33)
Total practical vertical depth are as follows:
In formula (2)~(36), the meaning and value range of each parameter are as shown in the table:
Well track variable bound boundary and constraint condition
Actual measurement length D1~D5, actual loading F1~F7 and corresponding reality by formula (2)~(25) each well section track of calculating
Border controls torque T1~T7；Formula (1)~(36) calculating target function TMD, TCT simultaneously saves the optimal solution for meeting constraint condition, will
The first generation meets the optimal feasible solution of constraint condition, optimal objective function value is kept in, as globally optimal solution, global object letter
Number calculates target fitness value TMD and TCT according to formula (1) with the increase of the number of iterations, records contemporary optimal solution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910410779.7A CN110134006B (en) | 2019-05-16 | 2019-05-16 | Complex borehole trajectory optimization method based on improved multi-target particle swarm optimization |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910410779.7A CN110134006B (en) | 2019-05-16 | 2019-05-16 | Complex borehole trajectory optimization method based on improved multi-target particle swarm optimization |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110134006A true CN110134006A (en) | 2019-08-16 |
| CN110134006B CN110134006B (en) | 2022-04-29 |
Family
ID=67574737
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910410779.7A Expired - Fee Related CN110134006B (en) | 2019-05-16 | 2019-05-16 | Complex borehole trajectory optimization method based on improved multi-target particle swarm optimization |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110134006B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111382543A (en) * | 2020-03-13 | 2020-07-07 | 中国海洋石油集团有限公司 | Offshore cluster well drilling sequence optimization method based on improved ant colony algorithm |
| CN112101672A (en) * | 2020-09-22 | 2020-12-18 | 郑州铁路职业技术学院 | Gas field group collaborative development optimization method based on particle swarm optimization algorithm |
| CN112163763A (en) * | 2020-09-25 | 2021-01-01 | 大连大学 | Weapon target allocation solving method based on improved multi-target HQPsOGA algorithm |
| CN113051775A (en) * | 2021-04-13 | 2021-06-29 | 中南大学 | Horizontal directional drilling track optimization method based on improved radial movement algorithm |
| CN113205171A (en) * | 2021-05-07 | 2021-08-03 | 柳州华世通汽车部件股份有限公司 | Multi-objective optimization method for discrete binary particle swarm motor |
| CN113610314A (en) * | 2021-08-17 | 2021-11-05 | 中海石油(中国)有限公司 | Drilling parameter optimization method, system, equipment and medium based on particle swarm optimization |
| CN113688486A (en) * | 2021-07-29 | 2021-11-23 | 国网吉林省电力有限公司四平供电公司 | Ultrasonic detection method for inner lead of transformer bushing based on echo model parameter estimation |
| CN113761730A (en) * | 2021-08-27 | 2021-12-07 | 浙江理工大学 | Tea garden four-rod intertillage mechanism multi-target-point track comprehensive method based on improved difference |
| CN113935132A (en) * | 2021-10-15 | 2022-01-14 | 福州大学 | Parameter optimization method for viscous damper based on improved multi-target particle swarm algorithm |
| CN116954239A (en) * | 2023-06-12 | 2023-10-27 | 成都丰千达科技有限公司 | Unmanned aerial vehicle track planning method and system based on improved particle swarm optimization |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106408111A (en) * | 2016-08-29 | 2017-02-15 | 方洋旺 | Novel multi-target particle swarm optimization method |
| CN106940742A (en) * | 2017-03-07 | 2017-07-11 | 西安石油大学 | Bad hole track optimizing method based on quick self-adapted quantum genetic algorithm |
| CN107992051A (en) * | 2017-12-26 | 2018-05-04 | 江南大学 | Unmanned vehicle paths planning method based on improved multi-objective particle swarm algorithm |
-
2019
- 2019-05-16 CN CN201910410779.7A patent/CN110134006B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106408111A (en) * | 2016-08-29 | 2017-02-15 | 方洋旺 | Novel multi-target particle swarm optimization method |
| CN106940742A (en) * | 2017-03-07 | 2017-07-11 | 西安石油大学 | Bad hole track optimizing method based on quick self-adapted quantum genetic algorithm |
| CN107992051A (en) * | 2017-12-26 | 2018-05-04 | 江南大学 | Unmanned vehicle paths planning method based on improved multi-objective particle swarm algorithm |
Non-Patent Citations (5)
| Title |
|---|
| AMINATASHNEZHAD等: "Designing and optimizing deviated wellbore trajectories using novel particle swarm algorithms", 《JOURNAL OF NATURAL GAS SCIENCE AND ENGINEERING》 * |
| JUN ZHENG等: "Multi-objective cellular particle swarm optimization for wellbore trajectory design", 《APPLIED SOFT COMPUTING》 * |
| 张栋栋: "基于粒子群算法的钻进参数多目标优化研究", 《中国优秀博硕士学位论文全文数据库(硕士)工程科技Ⅰ辑》 * |
| 沙林秀等: "基于NPSO的三维复杂井眼轨迹控制转矩的优选", 《石油机械》 * |
| 爨莹等: "基于多目标粒子群算法与TOPSIS的射孔方案优选研究", 《科学技术与工程》 * |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111382543A (en) * | 2020-03-13 | 2020-07-07 | 中国海洋石油集团有限公司 | Offshore cluster well drilling sequence optimization method based on improved ant colony algorithm |
| CN112101672A (en) * | 2020-09-22 | 2020-12-18 | 郑州铁路职业技术学院 | Gas field group collaborative development optimization method based on particle swarm optimization algorithm |
| CN112163763B (en) * | 2020-09-25 | 2023-09-01 | 大连大学 | Weapon target distribution solving method based on improved multi-target HQPSOGA algorithm |
| CN112163763A (en) * | 2020-09-25 | 2021-01-01 | 大连大学 | Weapon target allocation solving method based on improved multi-target HQPsOGA algorithm |
| CN113051775A (en) * | 2021-04-13 | 2021-06-29 | 中南大学 | Horizontal directional drilling track optimization method based on improved radial movement algorithm |
| CN113205171A (en) * | 2021-05-07 | 2021-08-03 | 柳州华世通汽车部件股份有限公司 | Multi-objective optimization method for discrete binary particle swarm motor |
| CN113205171B (en) * | 2021-05-07 | 2024-05-14 | 柳州华世通汽车部件股份有限公司 | Multi-objective optimization method for discrete binary particle swarm motor |
| CN113688486A (en) * | 2021-07-29 | 2021-11-23 | 国网吉林省电力有限公司四平供电公司 | Ultrasonic detection method for inner lead of transformer bushing based on echo model parameter estimation |
| CN113688486B (en) * | 2021-07-29 | 2024-05-28 | 国网吉林省电力有限公司四平供电公司 | Ultrasonic detection method for lead in transformer bushing based on echo model parameter estimation |
| CN113610314A (en) * | 2021-08-17 | 2021-11-05 | 中海石油(中国)有限公司 | Drilling parameter optimization method, system, equipment and medium based on particle swarm optimization |
| CN113761730A (en) * | 2021-08-27 | 2021-12-07 | 浙江理工大学 | Tea garden four-rod intertillage mechanism multi-target-point track comprehensive method based on improved difference |
| CN113761730B (en) * | 2021-08-27 | 2023-09-29 | 浙江理工大学 | Tea garden four-bar intertillage mechanism multi-target point track comprehensive method based on improved difference |
| CN113935132A (en) * | 2021-10-15 | 2022-01-14 | 福州大学 | Parameter optimization method for viscous damper based on improved multi-target particle swarm algorithm |
| CN116954239A (en) * | 2023-06-12 | 2023-10-27 | 成都丰千达科技有限公司 | Unmanned aerial vehicle track planning method and system based on improved particle swarm optimization |
| CN116954239B (en) * | 2023-06-12 | 2024-03-19 | 成都丰千达科技有限公司 | Unmanned aerial vehicle track planning method and system based on improved particle swarm optimization |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110134006B (en) | 2022-04-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110134006A (en) | Bad hole track optimizing method based on improved multi-objective particle swarm algorithm | |
| WO2022179266A1 (en) | Deviation rectification control method and apparatus for shield tunneling attitude | |
| CN109884900B (en) | Design method of harvester path tracking controller based on adaptive model predictive control | |
| CN109033730A (en) | A kind of tool wear prediction technique based on improved particle swarm optimization algorithm | |
| CN107992648A (en) | A kind of Adaptive radial basis function neural network carries out aeroengine thrust algorithm for estimating | |
| CN106875050A (en) | A kind of Engineering constraint parameter optimization method based on improvement chaos ant colony algorithm | |
| CN113051775B (en) | Horizontal directional drilling track optimization method based on improved radial movement algorithm | |
| CN109885061B (en) | Improved NSGA-II-based dynamic positioning multi-objective optimization method | |
| CN110836670B (en) | Mixed firework particle swarm cooperation method for solving unmanned aerial vehicle constrained route planning | |
| CN116341390B (en) | Global search rapid convergence multi-constraint trajectory optimization method | |
| CN113435025B (en) | Robot high-performance track automatic generation method combined with multistage optimization model | |
| CN110134062B (en) | Multi-axis numerical control machine tool machining path optimization method based on reinforcement learning | |
| CN106940742A (en) | Bad hole track optimizing method based on quick self-adapted quantum genetic algorithm | |
| CN112859855A (en) | Robot multi-target path planning based on locust optimization algorithm | |
| CN104252132A (en) | Adaptive genetic algorithm-based interplanetary orbit control optimization method | |
| CN115903816A (en) | Low-energy-consumption mobile robot path planning method | |
| CN118244687B (en) | Intelligent equipment adjusting control method and system based on industrial control | |
| CN112330064A (en) | New drilling workload prediction method based on ensemble learning | |
| CN117608200B (en) | Ocean aircraft path planning method | |
| CN116728404A (en) | Mechanical arm fuzzy PID control method based on quantum pigeon swarm optimization | |
| CN114253215B (en) | Civil cabin door automatic drilling and riveting path planning method based on improved ant colony algorithm | |
| CN114912828A (en) | Optimization method based on numerical control machine tool technological parameters | |
| CN111997583A (en) | Rotating guide motor speed regulation method based on fuzzy neural network | |
| CN114610024B (en) | Multi-agent collaborative searching energy-saving method for mountain land | |
| CN115600633A (en) | Method for predicting abrasive belt abrasion height of curved surface of abrasive belt grinding screw |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20220429 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |