CN107025501A - A kind of multi fuel economic load dispatching optimization method based on gene editing difference algorithm - Google Patents

A kind of multi fuel economic load dispatching optimization method based on gene editing difference algorithm Download PDF

Info

Publication number
CN107025501A
CN107025501A CN201710237099.0A CN201710237099A CN107025501A CN 107025501 A CN107025501 A CN 107025501A CN 201710237099 A CN201710237099 A CN 201710237099A CN 107025501 A CN107025501 A CN 107025501A
Authority
CN
China
Prior art keywords
population
individual
formula
gene editing
economic load
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.)
Pending
Application number
CN201710237099.0A
Other languages
Chinese (zh)
Inventor
林艺城
孟安波
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangdong University of Technology
Original Assignee
Guangdong University of Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Guangdong University of Technology filed Critical Guangdong University of Technology
Priority to CN201710237099.0A priority Critical patent/CN107025501A/en
Publication of CN107025501A publication Critical patent/CN107025501A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/12—Computing arrangements based on biological models using genetic models
    • G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06312—Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Electricity, gas or water supply

Abstract

A kind of multi fuel economic load dispatching optimization method based on gene editing difference algorithm provided in an embodiment of the present invention, by setting up the economic load dispatching model of meter and valve point effect and multi fuel and according to the economic load dispatching model initialization population, and then row variation is entered to population, intersect, selection operation, and perform gene editing technical operation and update the big population of fitness, population corresponding to economic load dispatching model has carried out gene editing operation and differential evolution algorithm, gene editing technical operation has been merged on the Research foundation of DE algorithms, by controlling all chromosomes in current nucleus to participate in the gene scale edited, part dimension is assisted to avoid damage to normal dimension while breaking away from dimension local optimum, it is effectively prevented from algorithm and crosses Premature Convergence, in the economic load dispatching model in face of meter and valve point effect and multi fuel, it can solve the problem that the existing economic load dispatching optimization method late convergence based on differential evolution algorithm is slow, it is easily trapped into the technical problem of local optimum.

Description

A kind of multi fuel economic load dispatching optimization method based on gene editing difference algorithm
Technical field
Adjusted the present invention relates to field of power, more particularly to a kind of multi fuel economy based on gene editing difference algorithm Spend optimization method.
Background technology
In Operation of Electric Systems, economic load dispatching (ED) is an important optimization task, and its main target is to meet Exert oneself-the inequality constraints condition such as the equality constraint of balancing the load and generator output on the premise of so that total power production cost is most It is small, have great importance to the safety and economic operation of power system.
In traditional ED optimization problems, generating set intake valve suddenly open produced by hot candied phenomenon --- valve point effect is past Disregard toward being ignored, it reduce the solving precision of model.Meanwhile, be only capable of using single-fuel conventional electric power generation unit not Economy and the demand of environment can be met, therefore the generating set of pluralities of fuel can be used to turn into the main flow of current thermal power generation.Meter And valve point effect adds substantial amounts of local best points to the optimizing space of ED problems, and consider multi fuel and then make it that ED solution is empty Between it is discontinuous, therefore the problem shows a series of higher-dimensions, non-convex, non-linear, discontinuous feature, further increases problem Solution difficulty.As what the system engineering theory was studied reaches its maturity with the present computer technology in meter and valve point effect and many combustions Expect continuing to develop and applying for economic load dispatching field, various new methods and new technology also emerge in an endless stream.It is common to have traditional mathematics Optimization method：Such as linear programming, Non-Linear Programming, Dynamic Programming, traditional optimization method depend on mathematical modeling unduly, and Mathematical modeling need to be simplified during solution, and more sensitive to initial solution, therefore handle such meter and valve point effect and many Local best points are easily trapped into during the complicated DE optimization problems of fuel.In recent years, intelligent optimization algorithm has obtained high speed development, because This, uses heuristic value in large quantities on multi fuel DE is solved the problems, such as, such as original Heuristic Method mainly has Genetic algorithm (GA), particle cluster algorithm (PSO), difference algorithm (DE), gravity searching algorithm (GSO), auction distributed algorithm (AA), cuckoo searching algorithm (CSA), tabu search algorithm (TSA), Biological optimization algorithm (BBO), seeker optimization algorithm (SOA) etc..Relative to traditional mathematicses optimization method, these original Heuristic algorithms are not special to the mathematical modeling of required problem Limitation, with powerful adaptability.However, primal algorithm, which is still suffered from, precocious phenomenon easily occurred, convergence precision is not high to be lacked Fall into.Therefore, more scholars are directed to studying its improved method, such as from the improvement on algorithm structure, mainly there is improved genetic algorithms Method, particle cluster algorithm is improved, cuckoo algorithm is improved, improves the algorithm that leapfrogs, disturbance difference algorithm (SDE), global optimum's harmony Searching algorithm (GHS) etc., or combine the mixing of two and algorithm above, such as blending heredity multiplier more new algorithm, mixed distribution formula Population and tabu search algorithm etc., these are improved or hybrid algorithm is more stronger than primal algorithm adaptability, and optimum results are also more Tool advantage, but these algorithms are facing many combustions of non-differentiability, discontinuous, non-convex, nonlinear extensive meter and valve point effect When expecting economic load dispatching optimization problem, equally there is also some defects, such as population diversity is not enough, is easily trapped into local optimum, receives Speed is held back slower etc., it is, therefore, desirable to provide a kind of more efficient method solves the optimization problem.
The basic thought of differential evolution (Differential Evolution, DE) algorithm is to copy biological evolution mechanism, New individual is produced using the difference between individual in population, is then intersected, selection operation realizes the evolution of population, passed through Successive ignition search finally gives globally optimal solution.DE algorithms not only possess simple in construction, and control parameter is few, it is easy to accomplish, have The advantages of stronger robustness, and its intrinsic concurrency contributes to Algorithm for Solving large scale system model.Especially low Tieing up has the advantages such as obvious fast convergence rate, low optimization accuracy height in unimodal function optimization problem.But facing higher-dimension, it is many During the complex optimization functions such as peak, non-linear and non-differentiability, there is late convergence slowly, be easily trapped into the defects such as local optimum.
For this problem, the present invention provides a kind of multi fuel economic load dispatching optimization side based on gene editing difference algorithm Method, and gene editing difference algorithm (GEDE) is proposed first, the algorithm is on the basis of standard difference algorithm, by incorporating me The operation of newly developed gene editing, the operation can effectively alleviate the DE algorithm later stages and restrain population diversity deficiency and cause to be absorbed in The phenomenon of local optimum, and after convergence in phase population individual is strict meet equality constraint in the case of, each gene editing behaviour Work is to be searched in feasible zone, and this greatly reduces whole search space, and algorithm search efficiency is significantly increased.
The content of the invention
The embodiments of the invention provide a kind of multi fuel economic load dispatching optimization method based on gene editing difference algorithm, use Higher-dimension, multimodal, non-linear and non-differentiability are being faced in solving the existing economic load dispatching optimization method based on differential evolution algorithm Deng complex optimization function when the late convergence that exists is slow, be easily trapped into the technical problem of local optimum.
A kind of multi fuel economic load dispatching optimization method based on gene editing difference algorithm provided in an embodiment of the present invention, bag Include：
S1：Set up the economic load dispatching model of meter and valve point effect and multi fuel；
S2：Population Size M, maximum iteration max gen, crossover probability p are setDE, editor Probability pc, according to the warp Ji scheduling model generates initial population and calculates initial fitness value according to the initial population at random in solution space；
S3：Mutation operation, crossover operation and selection operation, and Population Regeneration are performed to all individuals in population；
S4：Gene editing technical operation is performed to the population after renewal, Population Regeneration and iterations is recorded again and plus one；
S5：Step S3 to S4 is repeated, until iterations reaches default maximum iteration max gen.
Preferably, the step S1 is specifically included：
Setting up traditional economic load dispatching object function is：
Wherein, FjFor unit j fuel cost；N is generating set number；aj、bj、cjFor generating set j fuel cost system Number PjFor exerting oneself for generating set j；
Traditional economic load dispatching object function is modified according to valve point effect, revised economic load dispatching target is obtained Function is：
Wherein ej、fjFor generating set j valve point effect coefficient；Minimum technology for generating set j is exerted oneself；
Segment processing is carried out to revised economic load dispatching object function according to valve point effect and multi fuel, obtained finally The economic load dispatching model of meter and valve point effect and multi fuel, the economic load dispatching model of the meter and valve point effect and multi fuel can use Formula is expressed as：
Wherein, ajK、bjK、cjK、ejK、fjKFor the cost coefficient of unit j K type fuel, the piecewise function has K Section；For unit j EIAJ；
Also, the constraints of the economic load dispatching model of the meter and valve point effect and multi fuel includes account load balancing constraints With unit output constraint；
The account load balancing constraints are：
Wherein, PDFor system load demand；
The unit output is constrained to：
Preferably, the step S2 is specifically included：
Initial population, institute are generated by initialization of population formula at random according to the economic load dispatching model in solution space Stating initialization of population formula is：
Wherein, i ∈ (1, M), j ∈ (1, N)；
Each individual represents a solution in population, is passed through for the multi fuel of the meter comprising D platform generating sets and valve point effect Helped Problems of Optimal Dispatch, and its i-th of individual is represented by：
X (i)=[Xi,1,Xi,2,...Xi,N]=[Pi,1,Pi,2,...Pi,N]；
Wherein, X (i) is i-th of individual, X in populationi,1For first control variable, P in individual ii,1For individual i the 1st Generating set is exerted oneself；
Enter the out-of-limit processing of row constraint to the population after the initialization according to out-of-limit processing formula is constrained, the constraint is out-of-limit Handling formula is：
Fitness value, the fitness are calculated to entering the population after the out-of-limit processing of row constraint by fitness value calculation formula Value calculation formula is：
Wherein, λ is penalty coefficient,For the actual overall-fuel cost of optimization,For for system power injustice The penalty term of weighing apparatus, it can thus be seen that when this tends to 0,I.e. fitness value is overall-fuel cost value.
Preferably, the step S3 is specifically included：
Mutation operation is carried out to all individuals in population by mutation operation formula, the mutation operation formula is：
Wherein, g is current algebraically, r1、r2、r3∈ N (1, M) ∩ r1≠r2≠r3≠ i,Be g generation variation individual i,Be g for the different individual selected at random in population, G ∈ [0,2] are the scaling factor, for controlling difference ResoluteTo individualInfluence；
By crossover operation formula according to individual in current populationWith the variation individual produced by mutation operationImplement Crossover operation, produces experimental subjectsThe crossover operation formula is：
Wherein, rand be [0,1] between obey equally distributed random number, pDE∈ [0,1] is crossover probability, and N is variable Dimension；
By selection operation formula to parent individualityWith experiment individualSelection operation is performed, son is preferentially retained as In generation, is individual, and the selection operation formula is：
Wherein,For i-th of offspring individual；
It regard the corresponding population of the population at individual after selection operation as the population after renewal.
Preferably, the step S4 is specifically included：
S401：It is the nucleus in gene editing technical operation to mark population, and it is gene editing to mark the individual in population Individual dimension in chromosome in technical operation, mark population is all of the homologue in gene editing technical operation All genes are carried out two and neither repeat to match, obtain pairing gene by gene；
S402：If obeying equally distributed random number rand between pairing gene corresponding [0,1] more than default editor Probability pc, then pairing gene editing operation is performed to pairing gene；
S403：To the population for having performed the population at individual of gene editing technical operation Yu being not carried out gene editing technical operation Individual carries out fitness value calculation and compares the size of fitness value, and selection retains the big population at individual of fitness value, abandons fitting Answer the population at individual that angle value is small；
S404：Record iterations adds one；
Pairing gene editing operation in the step S402 is specifically included：
A1：D in chromosome i is set1、d2Wiki is because of respectively x (i, d1) and x (i, d2), and pass through genetic donor formula Fixed point shearing and donor restructuring are carried out to gene, the genetic donor extracted is obtained, the genetic donor formula is：
Wherein, L (d1)、L(d2) it is gene conserved sequence, gd is the genetic donor that is extracted；
A2：Random shearing is carried out to the genetic donor of the extraction by genetic donor fragment formula, after being sheared Genetic donor fragment, the genetic donor fragment formula is：
Wherein, gs1, gs2 are respectively the genetic donor fragment after shearing；
A3：Fragment replacement is carried out to the genetic donor fragment after the shearing by fragment replacement formula, the fragment is replaced It is changed to：
Wherein, gr (i, d1)、gr(i,d2) be respectively chromosome i repair after d1、d2Wiki because；
A4：Judge whether gene editing succeeds by genetic test formula, if so, then performing step to gene to next assemble Rapid S402, until all pairing gene editings are finished, step A1 is performed if it is not, then returning；
The genetic test formula is：
gr(i,d1)≤U(d1) and gr (i, d2)≤U(d2)；
Wherein, U (d1)、U(d2) it is the gene order upper limit.
A kind of multi fuel economic load dispatching optimization device based on gene editing difference algorithm provided in an embodiment of the present invention, bag Include：
Economic load dispatching model building module, by set up based on and valve point effect and multi fuel economic load dispatching model；
Initialization of population module, for setting Population Size M, maximum iteration max gen, crossover probability pDE, editor Probability pc, initial population is generated and according to the initial population meter according to the economic load dispatching model at random in solution space Calculate initial fitness value；
Make a variation cross selection module, is grasped for performing mutation operation, crossover operation and selection to all individuals in population Make, and Population Regeneration；
Gene editing module, for performing gene editing technical operation to the population after renewal, Population Regeneration and remembers again Record iterations adds one；
Loop module, for repeating variation cross selection module and gene editing module, until iterations reaches Default maximum iteration max gen.
Preferably, the economic load dispatching model building module is specifically included：
Unit is set up in traditional economy scheduling, and the economic load dispatching object function traditional for setting up is：
Wherein, FjFor unit j fuel cost；N is generating set number；aj、bj、cjFor generating set j fuel cost system Number PjFor exerting oneself for generating set j；
Valve point effect amending unit, for being modified according to valve point effect to traditional economic load dispatching object function, is obtained Obtaining revised economic load dispatching object function is：
Wherein ej、fjFor generating set j valve point effect coefficient；Minimum technology for generating set j is exerted oneself；
Valve point effect and multi fuel amending unit, for according to valve point effect and multi fuel to revised economic load dispatching mesh Scalar functions carry out the economic load dispatching model of segment processing, the final meter of acquisition and valve point effect and multi fuel, the meter and valve point The economic load dispatching model of effect and multi fuel can be formulated as：
Wherein, ajK、bjK、cjK、ejK、fjKFor the cost coefficient of unit j K type fuel, the piecewise function has K Section；For unit j EIAJ；
Also, the constraints of the economic load dispatching model of the meter and valve point effect and multi fuel includes account load balancing constraints With unit output constraint；
The account load balancing constraints are：
Wherein, PDFor system load demand；
The unit output is constrained to：
Preferably, the initialization of population module is specifically included：
Initialization of population unit, for by initialization of population formula according to the economic load dispatching model in solution space with Machine generates initial population, and the initialization of population formula is：
Wherein, i ∈ (1, M), j ∈ (1, N)；
Each individual represents a solution in population, is passed through for the multi fuel of the meter comprising D platform generating sets and valve point effect Helped Problems of Optimal Dispatch, and its i-th of individual is represented by：
X (i)=[Xi,1,Xi,2,...Xi,N]=[Pi,1,Pi,2,...Pi,N]；
Wherein, X (i) is i-th of individual, X in populationi,1For first control variable, P in individual ii,1For individual i the 1st Generating set is exerted oneself；
Out-of-limit processing unit is constrained, for entering row constraint to the population after the initialization according to the out-of-limit processing formula of constraint Out-of-limit processing, it is described constraint it is out-of-limit processing formula be：
Fitness value calculation unit, by by fitness value calculation formula to entering based on the population after the out-of-limit processing of row constraint Fitness value is calculated, the fitness value calculation formula is：
Wherein, λ is penalty coefficient,For the actual overall-fuel cost of optimization,For for system power injustice The penalty term of weighing apparatus, it can thus be seen that when this tends to 0,I.e. fitness value is overall-fuel cost value.
Preferably, the variation cross selection module is specifically included：
Mutation operation unit, it is described for carrying out mutation operation to all individuals in population by mutation operation formula Mutation operation formula is：
Wherein, g is current algebraically, r1、r2、r3∈ N (1, M) ∩ r1≠r2≠r3≠ i,Be g generation variation individual i,Be g for the different individual selected at random in population, G ∈ [0,2] are the scaling factor, for controlling difference ResoluteTo individualInfluence；
Crossover operation unit, for by crossover operation formula according to individual in current populationWith produced by mutation operation Variation individualImplement crossover operation, produce experimental subjectsThe crossover operation formula is：
Wherein, rand be [0,1] between obey equally distributed random number, pDE∈ [0,1] is crossover probability, and N is variable Dimension；
Selection operation unit, for by selection operation formula to parent individualityWith experiment individualPerform selection behaviour Make, be preferentially retained as offspring individual, the selection operation formula is：
Wherein,For i-th of offspring individual；
First population updating block, for using the corresponding population of the population at individual after selection operation as after renewal Population.
Preferably, the gene editing module is specifically included：
Mark and pairing unit, for marking population to be the nucleus in gene editing technical operation, are marked in population Individual is that the individual dimension in the chromosome in gene editing technical operation, mark population is pair in gene editing technical operation All genes of chromosome are answered, all genes are carried out with two and neither repeats to match, pairing gene is obtained；
Gene editing operating unit is matched, if equally distributed random for being obeyed between pairing gene corresponding [0,1] Number rand is more than default editor's Probability pc, then pairing gene editing operation is performed to pairing gene；
Second population updating block, for being compiled to having performed the population at individual of gene editing technical operation and being not carried out gene The population at individual of volume technical operation carries out fitness value calculation and compares the size of fitness value, and it is big that selection retains fitness value Population at individual, abandons the small population at individual of fitness value；
Iterations summing elements, plus one for recording iterations；
The unit that pairing gene editing operation is carried out in the pairing gene editing operating unit is specifically included：
Genetic donor extracts subelement, for setting the d in chromosome i1、d2Wiki is because of respectively x (i, d1) and x (i, d2), and fixed point shearing and donor restructuring are carried out to gene by genetic donor formula, obtain the genetic donor extracted, the base Because donor formula is：
Wherein, L (d1)、L(d2) it is gene conserved sequence, gd is the genetic donor that is extracted；
Random shearing subelement, for being cut at random to the genetic donor of the extraction by genetic donor fragment formula Cut, the genetic donor fragment after being sheared, the genetic donor fragment formula is：
Wherein, gs1, gs2 are respectively the genetic donor fragment after shearing；
Fragment replaces subelement, for carrying out fragment to the genetic donor fragment after the shearing by fragment replacement formula Replace, the fragment is replaced with：
Wherein, gr (i, d1)、gr(i,d2) be respectively chromosome i repair after d1、d2Wiki because；
Genetic test subelement, for judging whether gene editing succeeds by genetic test formula, if so, then to next Assemble and pairing gene editing operating unit is performed to gene, until all pairing gene editings are finished, base is performed if it is not, then returning Because donor extracts subelement；
The genetic test formula is：
gr(i,d1)≤U(d1) and gr (i, d2)≤U(d2)；
Wherein, U (d1)、U(d2) it is the gene order upper limit.
As can be seen from the above technical solutions, the embodiment of the present invention has advantages below：
A kind of multi fuel economic load dispatching optimization method based on gene editing difference algorithm provided in an embodiment of the present invention, leads to Cross set up meter and valve point effect and multi fuel economic load dispatching model and according to the economic load dispatching model initialization population, and then Enter row variation, intersection, selection operation to population, and perform gene editing technical operation and update the big population of fitness, to warp The corresponding population of Ji scheduling model has carried out gene editing operation and differential evolution algorithm, is merged on the Research foundation of DE algorithms Gene editing technical operation, by controlling all chromosomes in current nucleus to participate in the gene scale of editor, assists part Dimension avoids damage to normal dimension while breaking away from dimension local optimum, be effectively prevented from algorithm and cross Premature Convergence, in face of meter and valve point During the economic load dispatching model of effect and multi fuel, the existing economic load dispatching optimization method based on differential evolution algorithm can solve the problem that The late convergence that exists when facing the complex optimization functions such as higher-dimension, multimodal, non-linear and non-differentiability is slow, be easily trapped into office The optimal technical problem in portion.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing There is the accompanying drawing used required in technology description to be briefly described, it should be apparent that, drawings in the following description are only this Some embodiments of invention, for those of ordinary skill in the art, without having to pay creative labor, may be used also To obtain other accompanying drawings according to these accompanying drawings.
Fig. 1 is a kind of multi fuel economic load dispatching optimization side based on gene editing difference algorithm provided in an embodiment of the present invention One embodiment of method；
Fig. 2 is a kind of multi fuel economic load dispatching optimization side based on gene editing difference algorithm provided in an embodiment of the present invention Another embodiment of method；
Fig. 3 is a kind of multi fuel economic load dispatching optimization side based on gene editing difference algorithm provided in an embodiment of the present invention The consumption performance diagram of valve point effect is considered in another embodiment of method；
Fig. 4 is a kind of multi fuel economic load dispatching optimization side based on gene editing difference algorithm provided in an embodiment of the present invention The consumption performance diagram of valve point effect and multi fuel is considered in another embodiment of method；
Fig. 5 is a kind of multi fuel economic load dispatching optimization side based on gene editing difference algorithm provided in an embodiment of the present invention The schematic diagram of gene editing operation is matched in another embodiment of method；
Fig. 6 is a kind of multi fuel economic load dispatching optimization side based on gene editing difference algorithm provided in an embodiment of the present invention Pass through the search schematic diagram of gene editing technical operation in another embodiment of method；
Fig. 7 is a kind of multi fuel economic load dispatching optimization side based on gene editing difference algorithm provided in an embodiment of the present invention 40 unit DE algorithms and GEDE algorithmic statement curve comparison figures in another embodiment of method.
Embodiment
The embodiments of the invention provide a kind of multi fuel economic load dispatching optimization method based on gene editing difference algorithm, use Higher-dimension, multimodal, non-linear and non-differentiability are being faced in solving the existing economic load dispatching optimization method based on differential evolution algorithm Deng complex optimization function when the late convergence that exists is slow, be easily trapped into the technical problem of local optimum.
To enable goal of the invention, feature, the advantage of the present invention more obvious and understandable, below in conjunction with the present invention Accompanying drawing in embodiment, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that disclosed below Embodiment be only a part of embodiment of the invention, and not all embodiment.Based on the embodiment in the present invention, this area All other embodiment that those of ordinary skill is obtained under the premise of creative work is not made, belongs to protection of the present invention Scope.
Referring to Fig. 1, a kind of multi fuel economic load dispatching based on gene editing difference algorithm provided in an embodiment of the present invention One embodiment of optimization method, including：
101：Set up the economic load dispatching model of meter and valve point effect and multi fuel；
102：Population Size M, maximum iteration max gen, crossover probability p are setDE, editor Probability pc, according to described Economic load dispatching model generates initial population and calculates initial fitness value according to the initial population at random in solution space；
103：Mutation operation, crossover operation and selection operation, and Population Regeneration are performed to all individuals in population；
104：Gene editing technical operation is performed to the population after renewal, Population Regeneration and iterations is recorded again and adds One；
105：Step 103 is repeated to 104, until iterations reaches default maximum iteration max gen.
A kind of multi fuel economic load dispatching optimization method based on gene editing difference algorithm provided in an embodiment of the present invention, leads to Cross set up meter and valve point effect and multi fuel economic load dispatching model and according to the economic load dispatching model initialization population, and then Enter row variation, intersection, selection operation to population, and perform gene editing technical operation and update the big population of fitness, to warp The corresponding population of Ji scheduling model has carried out gene editing operation and differential evolution algorithm, is merged on the Research foundation of DE algorithms Gene editing technical operation, by controlling all chromosomes in current nucleus to participate in the gene scale of editor, assists part Dimension avoids damage to normal dimension while breaking away from dimension local optimum, be effectively prevented from algorithm and cross Premature Convergence, in face of meter and valve point During the economic load dispatching model of effect and multi fuel, the existing economic load dispatching optimization method based on differential evolution algorithm can solve the problem that The late convergence that exists when facing the complex optimization functions such as higher-dimension, multimodal, non-linear and non-differentiability is slow, be easily trapped into office The optimal technical problem in portion.
Above is excellent to a kind of multi fuel economic load dispatching based on gene editing difference algorithm provided in an embodiment of the present invention One embodiment of change method is described in detail, below will be to provided in an embodiment of the present invention a kind of poor based on gene editing Another embodiment of the multi fuel economic load dispatching optimization method of algorithm is divided to be described in detail.
Referring to Fig. 2, a kind of multi fuel economic load dispatching based on gene editing difference algorithm provided in an embodiment of the present invention Another embodiment of optimization method, including：
The first step：Set up the economic load dispatching model of meter and valve point effect and multi fuel；
Second step：Population Size M, maximum iteration max gen, crossover probability p are setDE, editor Probability pc, according to institute Economic load dispatching model is stated to generate initial population at random in solution space and calculate initial fitness according to the initial population Value；
3rd step：Mutation operation, crossover operation and selection operation, and Population Regeneration are performed to all individuals in population；
4th step：Gene editing technical operation is performed to the population after renewal, Population Regeneration and iterations is recorded again Plus one；
5th step：The step of step the three is repeated to the 4th step, until iterations reaches default maximum iteration max gen。
It should be noted that after iterations reaches default maximum iteration max gen, exportable fitness value Best solution is as final optimization pass result, and the Population Regeneration again in the 4th step includes the selection more preferable population of fitness value.
The first step is specifically included：
Setting up traditional economic load dispatching object function is：
Wherein, FjFor unit j fuel cost；N is generating set number；aj、bj、cjFor generating set j fuel cost system Number PjFor exerting oneself for generating set j；
It should be noted that the main target of tradition ED problems is to meet the various constraints of safe operation of power system Under, by optimizing each generating set active power output so that all generator overall-fuel costs are minimum.
Traditional economic load dispatching object function is modified according to valve point effect, revised economic load dispatching target is obtained Function is：
Wherein ej、fjFor generating set j valve point effect coefficient；Minimum technology for generating set j is exerted oneself；
It should be noted that in actual economic load dispatching, generally also needing to consider that steam turbine intake valve is opened out suddenly Existing hot candied phenomenon, the phenomenon can be superimposed a pulsation effect --- valve point effect, such as Fig. 3 on the consumption characteristic curve of unit It is shown.
Segment processing is carried out to revised economic load dispatching object function according to valve point effect and multi fuel, obtained finally The economic load dispatching model of meter and valve point effect and multi fuel, the economic load dispatching model of the meter and valve point effect and multi fuel can use Formula is expressed as：
Wherein, ajK、bjK、cjK、ejK、fjKFor the cost coefficient of unit j K type fuel, the piecewise function has K Section；For unit j EIAJ；
Also, the constraints of the economic load dispatching model of the meter and valve point effect and multi fuel includes account load balancing constraints With unit output constraint；
The account load balancing constraints are：
Wherein, PDFor system load demand；
The unit output is constrained to：
It should be noted that when generating set considers valve point effect and multi fuel simultaneously, the consumption characteristic curve of unit It will appear from segmentation pulsation effect as shown in Figure 4.
Second step is specifically included：
Initial population, institute are generated by initialization of population formula at random according to the economic load dispatching model in solution space Stating initialization of population formula is：
Wherein, i ∈ (1, M), j ∈ (1, N)；
Each individual represents a solution in population, is passed through for the multi fuel of the meter comprising D platform generating sets and valve point effect Helped Problems of Optimal Dispatch, and its i-th of individual is represented by：
X (i)=[Xi,1,Xi,2,...Xi,N]=[Pi,1,Pi,2,...Pi,N]；
Wherein, X (i) is i-th of individual, X in populationi,1For first control variable, P in individual ii,1For individual i the 1st Generating set is exerted oneself；
Enter the out-of-limit processing of row constraint to the population after the initialization according to out-of-limit processing formula is constrained, the constraint is out-of-limit Handling formula is：
Fitness value, the fitness are calculated to entering the population after the out-of-limit processing of row constraint by fitness value calculation formula Value calculation formula is：
Wherein, λ is penalty coefficient,For the actual overall-fuel cost of optimization,For for system power injustice The penalty term of weighing apparatus, it can thus be seen that when this tends to 0,I.e. fitness value is overall-fuel cost value.
Similar to genetic algorithm, DE algorithms also comprising intersecting, make a variation and selection operation, unlike, DE is in random selection Parent individuality between generation variation individual on the basis of differential vector；Secondly, by certain crossover probability to parent individuality and change Different individual performs crossover operation, generation experiment individual；Finally retained using greedy strategy and fitted between parent individuality and experiment individual Preferable individual should be worth.3rd step is specifically included：
Mutation operation is carried out to all individuals in population by mutation operation formula, the mutation operation formula is：
Wherein, g is current algebraically, r1、r2、r3∈ N (1, M) ∩ r1≠r2≠r3≠ i,Be g generation variation individual i,Be g for the different individual selected at random in population, G ∈ [0,2] are the scaling factor, for controlling difference ResoluteTo individualInfluence；
It should be noted that DE algorithms are to perform mutation operation on the basis of differential vector between parent individuality, often Individual differential vector includes two Different Individuals of parent (such as g generations).DE has a variety of combined schemes in practical application, and it makes a variation individual The generating mode of body is also each variant.
By crossover operation formula according to individual in current populationWith the variation individual produced by mutation operationImplement Crossover operation, produces experimental subjectsThe crossover operation formula is：
Wherein, rand be [0,1] between obey equally distributed random number, pDE∈ [0,1] is crossover probability, and N is variable Dimension；
It should be noted that DE algorithms by crossover operation to improve the diversity of population.The process is according to current population Middle individualWith the variation individual produced by mutation operationImplement crossover operation, produce experimental subjects
By selection operation formula to parent individualityWith experiment individualSelection operation is performed, filial generation is preferentially retained as Individual, the selection operation formula is：
Wherein,For i-th of offspring individual；
It regard the corresponding population of the population at individual after selection operation as the population after renewal.
It should be noted that DE algorithms produce offspring individual using competitive strategy, pass through parent individualityWith experiment individualSelection operation is performed, offspring individual is preferentially retained as.
The performance of standard DE algorithms depends primarily on its global search and localized detection ability, and in a sense, this takes The setting of the control parameters such as Population Size, the scaling factor, crossover probability certainly in the algorithm, it is heuristic compared to others For algorithm, it is few that DE algorithms possess control parameter, workable, collective search, with the individual optimal and global optimum of memory Guarantor's excellent function the advantages of.However, most heuristic with other genetic algorithms, ant group algorithm, particle cluster algorithm etc. Algorithm is similar, and DE algorithms cause the algorithm due to itself intrinsic way of search for carrying out mutation generation new individual using difference vector When iterations is up to when proceeding to certain number of times, the drastically decline of population diversity forms " aggregation " phenomenon, causes receipts too early The problem of holding back.
To strengthen the global convergence ability of DE algorithms, make algorithm when handling higher-dimension multimodal complicated optimum problem, reduction is calculated There is the possibility stagnated in a certain local best points in method, it is to avoid population diversity premature loss, assists it to jump out local optimum. The present invention on the basis of above-mentioned standard DE algorithms by incorporating our newly developed gene editing technical operations.4th step has Body includes：
S401：It is the nucleus in gene editing technical operation to mark population, and it is gene editing to mark the individual in population Individual dimension in chromosome in technical operation, mark population is all of the homologue in gene editing technical operation All genes are carried out two and neither repeat to match, obtain pairing gene by gene；
S402：If obeying equally distributed random number rand between pairing gene corresponding [0,1] more than default editor Probability pc, then pairing gene editing operation is performed to pairing gene；
S403：To the population for having performed the population at individual of gene editing technical operation Yu being not carried out gene editing technical operation Individual carries out fitness value calculation and compares the size of fitness value, and selection retains the big population at individual of fitness value, abandons fitting Answer the population at individual that angle value is small；
S404：Record iterations adds one；
Referring to Fig. 5, the pairing gene editing operation in step S402 is specifically included：
A1：D in chromosome i is set1、d2Wiki is because of respectively x (i, d1) and x (i, d2), and pass through genetic donor formula Fixed point shearing and donor restructuring are carried out to gene, the genetic donor extracted is obtained, the genetic donor formula is：
Wherein, L (d1)、L(d2) it is gene conserved sequence, gd is the genetic donor that is extracted；
A2：Random shearing is carried out to the genetic donor of the extraction by genetic donor fragment formula, after being sheared Genetic donor fragment, the genetic donor fragment formula is：
Wherein, gs1, gs2 are respectively the genetic donor fragment after shearing；
A3：Fragment replacement is carried out to the genetic donor fragment after the shearing by fragment replacement formula, the fragment is replaced It is changed to：
Wherein, gr (i, d1)、gr(i,d2) be respectively chromosome i repair after d1、d2Wiki because；
A4：Judge whether gene editing succeeds by genetic test formula, if so, then performing step to gene to next assemble Rapid S402, until all pairing gene editings are finished, step A1 is performed if it is not, then returning；
The genetic test formula is：
gr(i,d1)≤U(d1) and gr (i, d2)≤U(d2)；
Wherein, U (d1)、U(d2) it is the gene order upper limit.
It should be noted that all genes (dimension) progress two of chromosome (individual) neither repeats to match somebody with somebody in nucleus (population) It is right, and according to editor's Probability pcJudge whether pairing gene performs gene editing operation, if rand ＞ pcThen perform gene editing behaviour Make, it is assumed that the d in chromosome i1、d2Wiki is because of respectively x (i, d1) and x (i, d2), then to their then row gene editing operation productions The raw d repaired1、d2Wiki because.
It should be noted that the 4th step can also be expressed as chromosome (individual) all genes (dimension) in nucleus (population), Random two are carried out neither to repeat to match, and according to editor's Probability pcJudge whether pairing gene performs gene editing technical operation, If performing, gene site-directed shearing, donor restructuring in execution gene editing technical operation, genetic donor random shearing, fixed point Genetic fragment is replaced and genetic test and identification, is produced new chromosome (individual) gene pairs, secondly, is calculated all gene editings The fitness of new chromosome, performs selection operation produced by after operation, preferentially retains under the preferable chromosome entrance of fitness value An iteration.
GEDE algorithms of the present invention have merged gene editing technical operation on the Research foundation of DE algorithms, and the operation passes through Using editor's Probability pcAll chromosomes in current nucleus are controlled to participate in the gene scale edited, this is conducive to aid Point dimension avoids damage to normal dimension while breaking away from dimension local optimum, be effectively prevented from algorithm and cross Premature Convergence.Secondly, by above-mentioned base Because of editing technique operating process it is seen that, the operation can't producer drift, i.e., after gene editing technical operation match Tie up sum and keep constant.Therefore, progressively going deep into search, when occurring in nucleus, all gene sums are strict on chromosome In satisfaction during equality constraint, now performed each gene editing technical operation equally will strictly meet constraints, for The multi fuel economic load dispatching of above-mentioned meter and valve point effect is even more the i.e. gene each time while meet equality constraint and inequality constraints Globally optimal solution is searched in editing technique operation on feasible zone, this greatly improves the computational efficiency of algorithm, and this process can be by Shown in Fig. 6.
Specifically, the embodiment of the present invention can also be described as：
Step 1：Set Population Size M, maximum iteration max gen, crossover probability pDE, editor Probability pc, it is empty in solution Between in random initializtion population, and calculate the fitness value of each individual.
Step 2：Variation is performed to all individuals in population, is intersected and selection operation, Population Regeneration；
Step 3：Gene editing technical operation, Population Regeneration are performed to progeny population produced by step 2；
Step 4：Judge whether to reach maximum iteration max gen, if so, the then best solution conduct of output fitness value Final optimization pass result, otherwise, goes to step 2 and continues iterative search.
For verify GEDE Algorithm for Solving of the present invention have higher-dimension, non-convex, non-linear, discontinuous characteristic meter and valve point effect With the validity and superiority of multi fuel economic load dispatching optimization problem, the present invention using classical 40 unit meter and valve point effect and Multi fuel Economic Dispatch example carries out simulation analysis.
Referring to Fig. 7, from figure 7 it can be seen that the economic load dispatching optimization method of the embodiment of the present invention is generated in Fig. 7 GEDE curves, and common DE algorithms generate the DE curves in figure, it can be seen that the economic load dispatching optimization of the embodiment of the present invention Method can improve computational efficiency conscientiously.Less iterations is only needed to can obtain more preferable optimum results, and optimum results It is good.
It is low herein for the intrinsic computational efficiency of DE algorithms itself, the defects such as local optimum are easily trapped into, are proposed a kind of Gene editing difference algorithm, and it is applied to 10 machine set systems of the multi fuel Economic Dispatch Problem classics of meter and valve point effect In, verify the validity and superiority of the algorithm.The notable of inventive algorithm is advantageous in that：
1) each decision variable promise breaking problem being likely to occur in difference algorithm optimization process is directed to, fitness premise is being calculated Preceding carry out Treatment for Default, reduces infeasible solution, improves and calculates feasibility.
2) going deep into iteration, when there is individual ownership decision variable sum and meeting equality constraint, now each base Because globally optimal solution is searched in editing technique operation in feasible zone, which greatly improves computational efficiency.
3) the gene scale of gene editing operation is participated in using editor's probability control, is assisting to be absorbed in the optimal portion of local dimension While point dimension breaks away from current quagmire, it is to avoid the normal dimension of destruction, the excessively precocious phenomenon of algorithm is effectively alleviated.
Below will be excellent to a kind of multi fuel economic load dispatching based on gene editing difference algorithm provided in an embodiment of the present invention Makeup, which is put, to be described.
A kind of multi fuel economic load dispatching optimization device based on gene editing difference algorithm provided in an embodiment of the present invention, bag Include：
Economic load dispatching model building module, by set up based on and valve point effect and multi fuel economic load dispatching model；
Initialization of population module, for setting Population Size M, maximum iteration max gen, crossover probability pDE, editor Probability pc, initial population is generated and according to the initial population meter according to the economic load dispatching model at random in solution space Calculate initial fitness value；
Make a variation cross selection module, is grasped for performing mutation operation, crossover operation and selection to all individuals in population Make, and Population Regeneration；
Gene editing module, for performing gene editing technical operation to the population after renewal, Population Regeneration and remembers again Record iterations adds one；
Loop module, for repeating variation cross selection module and gene editing module, until iterations reaches Default maximum iteration max gen.
Economic load dispatching model building module is specifically included：
Unit is set up in traditional economy scheduling, and the economic load dispatching object function traditional for setting up is：
Wherein, FjFor unit j fuel cost；N is generating set number；aj、bj、cjFor generating set j fuel cost system Number PjFor exerting oneself for generating set j；
Valve point effect amending unit, for being modified according to valve point effect to traditional economic load dispatching object function, is obtained Obtaining revised economic load dispatching object function is：
Wherein ej、fjFor generating set j valve point effect coefficient；Minimum technology for generating set j is exerted oneself；
Valve point effect and multi fuel amending unit, for according to valve point effect and multi fuel to revised economic load dispatching mesh Scalar functions carry out the economic load dispatching model of segment processing, the final meter of acquisition and valve point effect and multi fuel, the meter and valve point The economic load dispatching model of effect and multi fuel can be formulated as：
Wherein, ajK、bjK、cjK、ejK、fjKFor the cost coefficient of unit j K type fuel, the piecewise function has K Section；For unit j EIAJ；
Also, the constraints of the economic load dispatching model of the meter and valve point effect and multi fuel includes account load balancing constraints With unit output constraint；
The account load balancing constraints are：
Wherein, PDFor system load demand；
The unit output is constrained to：
Initialization of population module is specifically included：
Initialization of population unit, for by initialization of population formula according to the economic load dispatching model in solution space with Machine generates initial population, and the initialization of population formula is：
Wherein, i ∈ (1, M), j ∈ (1, N)；
Each individual represents a solution in population, is passed through for the multi fuel of the meter comprising D platform generating sets and valve point effect Helped Problems of Optimal Dispatch, and its i-th of individual is represented by
X (i)=[Xi,1,Xi,2,...Xi,N]=[Pi,1,Pi,2,...Pi,N]；
Wherein, X (i) is i-th of individual, X in populationi,1For first control variable, P in individual ii,1For individual i the 1st Generating set is exerted oneself；
Out-of-limit processing unit is constrained, for entering row constraint to the population after the initialization according to the out-of-limit processing formula of constraint Out-of-limit processing, it is described constraint it is out-of-limit processing formula be：
Fitness value calculation unit, by by fitness value calculation formula to entering based on the population after the out-of-limit processing of row constraint Fitness value is calculated, the fitness value calculation formula is：
Wherein, λ is penalty coefficient,For the actual overall-fuel cost of optimization,For for system power injustice The penalty term of weighing apparatus, it can thus be seen that when this tends to 0,I.e. fitness value is overall-fuel cost value.
Variation cross selection module is specifically included：
Mutation operation unit, it is described for carrying out mutation operation to all individuals in population by mutation operation formula Mutation operation formula is：
Wherein, g is current algebraically, r1、r2、r3∈ N (1, M) ∩ r1≠r2≠r3≠ i,Be g generation variation individual i,Be g for the different individual selected at random in population, G ∈ [0,2] are the scaling factor, for controlling difference ResoluteTo individualInfluence；
Crossover operation unit, for by crossover operation formula according to individual in current populationWith produced by mutation operation Variation individualImplement crossover operation, produce experimental subjectsThe crossover operation formula is：
Wherein, rand be [0,1] between obey equally distributed random number, pDE∈ [0,1] is crossover probability, and N is variable Dimension；
Selection operation unit, for by selection operation formula to parent individualityWith experiment individualPerform selection behaviour Make, be preferentially retained as offspring individual, the selection operation formula is：
Wherein,For i-th of offspring individual；
First population updating block, for using the corresponding population of the population at individual after selection operation as after renewal Population.
Gene editing module is specifically included：
Mark and pairing unit, for marking population to be the nucleus in gene editing technical operation, are marked in population Individual is that the individual dimension in the chromosome in gene editing technical operation, mark population is pair in gene editing technical operation All genes of chromosome are answered, all genes are carried out with two and neither repeats to match, pairing gene is obtained；
Gene editing operating unit is matched, if equally distributed random for being obeyed between pairing gene corresponding [0,1] Number rand is more than default editor's Probability pc, then pairing gene editing operation is performed to pairing gene；
Second population updating block, for being compiled to having performed the population at individual of gene editing technical operation and being not carried out gene The population at individual of volume technical operation carries out fitness value calculation and compares the size of fitness value, and it is big that selection retains fitness value Population at individual, abandons the small population at individual of fitness value；
Iterations summing elements, plus one for recording iterations；
The unit that pairing gene editing operation is carried out in the pairing gene editing operating unit is specifically included：
Genetic donor extracts subelement, for setting the d in chromosome i1、d2Wiki is because of respectively x (i, d1) and x (i, d2), and fixed point shearing and donor restructuring are carried out to gene by genetic donor formula, obtain the genetic donor extracted, the base Because donor formula is：
Wherein, L (d1)、L(d2) it is gene conserved sequence, gd is the genetic donor that is extracted；
Random shearing subelement, for being cut at random to the genetic donor of the extraction by genetic donor fragment formula Cut, the genetic donor fragment after being sheared, the genetic donor fragment formula is：
Wherein, gs1, gs2 are respectively the genetic donor fragment after shearing；
Fragment replaces subelement, for carrying out fragment to the genetic donor fragment after the shearing by fragment replacement formula Replace, the fragment is replaced with：
Wherein, gr (i, d1)、gr(i,d2) be respectively chromosome i repair after d1、d2Wiki because；
Genetic test subelement, for judging whether gene editing succeeds by genetic test formula, if so, then to next Assemble and pairing gene editing operating unit is performed to gene, until all pairing gene editings are finished, base is performed if it is not, then returning Because donor extracts subelement；
The genetic test formula is：
gr(i,d1)≤U(d1) and gr (i, d2)≤U(d2)；
Wherein, U (d1)、U(d2) it is the gene order upper limit.
It is apparent to those skilled in the art that, for convenience and simplicity of description, the system of foregoing description, The specific work process of device and unit, may be referred to the corresponding process in preceding method embodiment, will not be repeated here.
Described above, the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although with reference to before Embodiment is stated the present invention is described in detail, it will be understood by those within the art that：It still can be to preceding State the technical scheme described in each embodiment to modify, or equivalent substitution is carried out to which part technical characteristic；And these Modification is replaced, and the essence of appropriate technical solution is departed from the spirit and scope of various embodiments of the present invention technical scheme.

Claims (10)

1. a kind of multi fuel economic load dispatching optimization method based on gene editing difference algorithm, it is characterised in that including：
S1：Set up the economic load dispatching model of meter and valve point effect and multi fuel；
S2：Population Size M, maximum iteration maxgen, crossover probability p are setDE, editor Probability pc, according to the economic tune Degree model generates initial population and calculates initial fitness value according to the initial population at random in solution space；
S3：Mutation operation, crossover operation and selection operation, and Population Regeneration are performed to all individuals in population；
S4：Gene editing technical operation is performed to the population after renewal, Population Regeneration and iterations is recorded again and plus one；
S5：Step S3 to S4 is repeated, until iterations reaches default maximum iteration maxgen.
2. a kind of multi fuel economic load dispatching optimization method based on gene editing difference algorithm according to claim 1, its It is characterised by, the step S1 is specifically included：
Setting up traditional economic load dispatching object function is：
Wherein, FjFor unit j fuel cost；N is generating set number；aj、bj、cjFor generating set j fuel cost FACTOR Pj For exerting oneself for generating set j；
Traditional economic load dispatching object function is modified according to valve point effect, revised economic load dispatching object function is obtained For：
Wherein ej、fjFor generating set j valve point effect coefficient；Minimum technology for generating set j is exerted oneself；
Segment processing is carried out to revised economic load dispatching object function according to valve point effect and multi fuel, obtain final meter and The economic load dispatching model of valve point effect and multi fuel, the economic load dispatching model of the meter and valve point effect and multi fuel can use formula It is expressed as：
Wherein, ajK、bjK、cjK、ejK、fjKFor the cost coefficient of unit j K type fuel, the piecewise function has K sections；For unit j EIAJ；
Also, the constraints of the economic load dispatching model of the meter and valve point effect and multi fuel includes account load balancing constraints and machine Group units limits；
The account load balancing constraints are：
Wherein, PDFor system load demand；
The unit output is constrained to：
3. a kind of multi fuel economic load dispatching optimization method based on gene editing difference algorithm according to claim 1, its It is characterised by, the step S2 is specifically included：
Initial population, the kind are generated by initialization of population formula at random according to the economic load dispatching model in solution space Group initializes formula：
Wherein, i ∈ (1, M), j ∈ (1, N)；
Each individual represents a solution in population, and the multi fuel economy for the meter comprising D platform generating sets and valve point effect is adjusted Optimization problem is spent, its i-th of individual is represented by：
X (i)=[Xi,1,Xi,2,...Xi,N]=[Pi,1,Pi,2,...Pi,N]；
Wherein, X (i) is i-th of individual, X in populationi,1For first control variable, P in individual ii,1For the 1st generating of individual i Unit is exerted oneself；
Enter the out-of-limit processing of row constraint, the out-of-limit processing of constraint to the population after the initialization according to out-of-limit processing formula is constrained Formula is：
Fitness value, the fitness value meter are calculated to entering the population after the out-of-limit processing of row constraint by fitness value calculation formula Calculating formula is：
Wherein, λ is penalty coefficient,For the actual overall-fuel cost of optimization,To be unbalanced for system power Penalty term, it can thus be seen that when this tends to 0,I.e. fitness value is overall-fuel cost value.
4. a kind of multi fuel economic load dispatching optimization method based on gene editing difference algorithm according to claim 1, its It is characterised by, the step S3 is specifically included：
Mutation operation is carried out to all individuals in population by mutation operation formula, the mutation operation formula is：
Wherein, g is current algebraically, r1、r2、r3∈ N (1, M) ∩ r1≠r2≠r3≠ i,Be g generation variation individual i, Be g for the different individual selected at random in population, G ∈ [0,2] are the scaling factor, for controlling differential vectorTo individualInfluence；
By crossover operation formula according to individual in current populationWith the variation individual produced by mutation operationImplement to intersect behaviour Make, produce experimental subjectsThe crossover operation formula is：
Wherein, rand be [0,1] between obey equally distributed random number, pDE∈[0,1] it is crossover probability, N is the dimension of variable Number；
By selection operation formula to parent individualityWith experiment individualSelection operation is performed, filial generation is preferentially retained as Body, the selection operation formula is：
Wherein,For i-th of offspring individual；
It regard the corresponding population of the population at individual after selection operation as the population after renewal.
5. a kind of multi fuel economic load dispatching optimization method based on gene editing difference algorithm according to claim 1, its It is characterised by, the step S4 is specifically included：
S401：It is the nucleus in gene editing technical operation to mark population, and it is gene editing technology to mark the individual in population Chromosome in operation, it is all bases of the homologue in gene editing technical operation to mark the individual dimension in population All genes are carried out two and neither repeat to match, obtain pairing gene by cause；
S402：If obeying equally distributed random number rand between pairing gene corresponding [0,1] more than default editor's probability pc, then pairing gene editing operation is performed to pairing gene；
S403：To the population at individual for having performed the population at individual of gene editing technical operation Yu being not carried out gene editing technical operation Carry out fitness value calculation and compare the size of fitness value, selection retains the big population at individual of fitness value, abandons fitness The small population at individual of value；
S404：Record iterations adds one；
Pairing gene editing operation in the step S402 is specifically included：
A1：D in chromosome i is set1、d2Wiki is because of respectively x (i, d1) and x (i, d2), and by genetic donor formula to base Because carrying out fixed point shearing and donor restructuring, the genetic donor extracted is obtained, the genetic donor formula is：
Wherein, L (d1)、L(d2) it is gene conserved sequence, gd is the genetic donor that is extracted；
A2：Random shearing, the gene after being sheared are carried out to the genetic donor of the extraction by genetic donor fragment formula Donor fragment, the genetic donor fragment formula is：
Wherein, gs1, gs2 are respectively the genetic donor fragment after shearing；
A3：Fragment replacement is carried out to the genetic donor fragment after the shearing by fragment replacement formula, the fragment is replaced with：
Wherein, gr (i, d1)、gr(i,d2) be respectively chromosome i repair after d1、d2Wiki because；
A4：Judge whether gene editing succeeds by genetic test formula, if so, then performing step to gene to next assemble S402, until all pairing gene editings are finished, step A1 is performed if it is not, then returning；
The genetic test formula is：
gr(i,d1)≤U(d1) and gr (i, d2)≤U(d2)；
Wherein, U (d1)、U(d2) it is the gene order upper limit.
6. a kind of multi fuel economic load dispatching optimization device based on gene editing difference algorithm, it is characterised in that including：
Economic load dispatching model building module, by set up based on and valve point effect and multi fuel economic load dispatching model；
Initialization of population module, for setting Population Size M, maximum iteration maxgen, crossover probability pDE, editor probability pc, initial population is generated according to the economic load dispatching model at random in solution space and calculated just according to the initial population Beginning fitness value；
Make a variation cross selection module, for performing mutation operation, crossover operation and selection operation to all individuals in population, and Population Regeneration；
Gene editing module, for after renewal population perform gene editing technical operation, again Population Regeneration and record change Generation number adds one；
Loop module, for repeating variation cross selection module and gene editing module, until iterations reach it is default Maximum iteration maxgen.
7. a kind of multi fuel economic load dispatching optimization device based on gene editing difference algorithm according to claim 6, its It is characterised by, the economic load dispatching model building module is specifically included：
Unit is set up in traditional economy scheduling, and the economic load dispatching object function traditional for setting up is：
Wherein, FjFor unit j fuel cost；N is generating set number；aj、bj、cjFor generating set j fuel cost FACTOR Pj For exerting oneself for generating set j；
Valve point effect amending unit, for being modified according to valve point effect to traditional economic load dispatching object function, is repaiied Economic load dispatching object function after just is：
Wherein ej、fjFor generating set j valve point effect coefficient；Minimum technology for generating set j is exerted oneself；
Valve point effect and multi fuel amending unit, for according to valve point effect and multi fuel to revised economic load dispatching target letter Number carries out the economic load dispatching model of segment processing, the final meter of acquisition and valve point effect and multi fuel, the meter and valve point effect It can be formulated as with the economic load dispatching model of multi fuel：
Wherein, ajK、bjK、cjK、ejK、fjKFor the cost coefficient of unit j K type fuel, the piecewise function has K sections；For unit j EIAJ；
Also, the constraints of the economic load dispatching model of the meter and valve point effect and multi fuel includes account load balancing constraints and machine Group units limits；
The account load balancing constraints are：
Wherein, PDFor system load demand；
The unit output is constrained to：
8. a kind of multi fuel economic load dispatching optimization device based on gene editing difference algorithm according to claim 6, its It is characterised by, the initialization of population module is specifically included：
Initialization of population unit, for being given birth at random in solution space according to the economic load dispatching model by initialization of population formula Into initial population, the initialization of population formula is：
Wherein, i ∈ (1, M), j ∈ (1, N)；
Each individual represents a solution in population, and the multi fuel economy for the meter comprising D platform generating sets and valve point effect is adjusted Optimization problem is spent, its i-th of individual is represented by：
X (i)=[Xi,1,Xi,2,...Xi,N]=[Pi,1,Pi,2,...Pi,N]；
Wherein, X (i) is i-th of individual, X in populationi,1For first control variable, P in individual ii,1For the 1st generating of individual i Unit is exerted oneself；
Out-of-limit processing unit is constrained, it is out-of-limit for entering row constraint to the population after the initialization according to the out-of-limit processing formula of constraint Handle, the out-of-limit processing formula of constraint is：
Fitness value calculation unit, for calculating suitable to entering the population after the out-of-limit processing of row constraint by fitness value calculation formula Angle value is answered, the fitness value calculation formula is：
Wherein, λ is penalty coefficient,For the actual overall-fuel cost of optimization,To be unbalanced for system power Penalty term, it can thus be seen that when this tends to 0,I.e. fitness value is overall-fuel cost value.
9. a kind of multi fuel economic load dispatching optimization device based on gene editing difference algorithm according to claim 6, its It is characterised by, the variation cross selection module is specifically included：
Mutation operation unit, for carrying out mutation operation, the variation to all individuals in population by mutation operation formula Operation formula be：
Wherein, g is current algebraically, r1、r2、r3∈ N (1, M) ∩ r1≠r2≠r3≠ i,Be g generation variation individual i, Be g for the different individual selected at random in population, G ∈ [0,2] are the scaling factor, for controlling differential vectorTo individualInfluence；
Crossover operation unit, for by crossover operation formula according to individual in current populationWith the change produced by mutation operation Different individualImplement crossover operation, produce experimental subjectsThe crossover operation formula is：
Wherein, rand be [0,1] between obey equally distributed random number, pDE∈ [0,1] is crossover probability, and N is the dimension of variable Number；
Selection operation unit, for by selection operation formula to parent individualityWith experiment individualSelection operation is performed, is selected Excellent to be retained as offspring individual, the selection operation formula is：
Wherein,For i-th of offspring individual；
First population updating block, for regarding the corresponding population of the population at individual after selection operation as the kind after renewal Group.
10. a kind of multi fuel economic load dispatching optimization device based on gene editing difference algorithm according to claim 6, its It is characterised by, the gene editing module is specifically included：
Mark and pairing unit, for marking population to be the nucleus in gene editing technical operation, mark the individual in population For the chromosome in gene editing technical operation, it is the correspondence dye in gene editing technical operation to mark the individual dimension in population All genes are carried out two and neither repeat to match, obtain pairing gene by all genes of colour solid；
Gene editing operating unit is matched, if for obeying equally distributed random number between pairing gene corresponding [0,1] Rand is more than default editor's Probability pc, then pairing gene editing operation is performed to pairing gene；
Second population updating block, for having performed the population at individual of gene editing technical operation and being not carried out gene editing skill The population at individual of art operation carries out fitness value calculation and compares the size of fitness value, and selection retains the big population of fitness value Individual, abandons the small population at individual of fitness value；
Iterations summing elements, plus one for recording iterations；
The unit that pairing gene editing operation is carried out in the pairing gene editing operating unit is specifically included：
Genetic donor extracts subelement, for setting the d in chromosome i1、d2Wiki is because of respectively x (i, d1) and x (i, d2), and Fixed point shearing is carried out to gene by genetic donor formula and donor is recombinated, the genetic donor extracted, the genetic donor is obtained Formula is：
Wherein, L (d1)、L(d2) it is gene conserved sequence, gd is the genetic donor that is extracted；
Random shearing subelement, for carrying out random shearing to the genetic donor of the extraction by genetic donor fragment formula, Genetic donor fragment after being sheared, the genetic donor fragment formula is：
Wherein, gs1, gs2 are respectively the genetic donor fragment after shearing；
Fragment replaces subelement, is replaced for carrying out fragment to the genetic donor fragment after the shearing by fragment replacement formula Change, the fragment is replaced with：
Wherein, gr (i, d1)、gr(i,d2) be respectively chromosome i repair after d1、d2Wiki because；
Genetic test subelement, for judging whether gene editing succeeds by genetic test formula, if so, then being assembled to next Pairing gene editing operating unit is performed to gene, until all pairing gene editings are finished, is supplied if it is not, then returning and performing gene Body extracts subelement；
The genetic test formula is：
gr(i,d1)≤U(d1) and gr (i, d2)≤U(d2)；
Wherein, U (d1)、U(d2) it is the gene order upper limit.
CN201710237099.0A 2017-04-12 2017-04-12 A kind of multi fuel economic load dispatching optimization method based on gene editing difference algorithm Pending CN107025501A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710237099.0A CN107025501A (en) 2017-04-12 2017-04-12 A kind of multi fuel economic load dispatching optimization method based on gene editing difference algorithm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710237099.0A CN107025501A (en) 2017-04-12 2017-04-12 A kind of multi fuel economic load dispatching optimization method based on gene editing difference algorithm

Publications (1)

Publication Number Publication Date
CN107025501A true CN107025501A (en) 2017-08-08

Family

ID=59527658

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710237099.0A Pending CN107025501A (en) 2017-04-12 2017-04-12 A kind of multi fuel economic load dispatching optimization method based on gene editing difference algorithm

Country Status (1)

Country Link
CN (1) CN107025501A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108565857A (en) * 2018-05-07 2018-09-21 江南大学 A kind of Economic Dispatch method based on information interchange strategy ACS in continuous space
WO2023065399A1 (en) * 2021-10-22 2023-04-27 中冶南方工程技术有限公司 Blast furnace batching optimization method based on differential evolution algorithm, and electronic device and storage medium

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104239961A (en) * 2014-07-16 2014-12-24 广东工业大学 Electric system economic dispatching optimization method based on criss-cross algorithm
CN105976052A (en) * 2016-04-28 2016-09-28 广东工业大学 Improved quantum-behaved particle swarm optimization algorithm-based multi-region economic dispatch method
CN106251043A (en) * 2016-07-20 2016-12-21 广东工业大学 Multizone economic environment method for optimizing scheduling and device
CN106339817A (en) * 2016-08-29 2017-01-18 广东工业大学 Multi-region dynamic economy scheduling method and system
CN106355277A (en) * 2016-08-30 2017-01-25 广东工业大学 Multi-area dynamic economic dispatch optimization method and system

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104239961A (en) * 2014-07-16 2014-12-24 广东工业大学 Electric system economic dispatching optimization method based on criss-cross algorithm
CN105976052A (en) * 2016-04-28 2016-09-28 广东工业大学 Improved quantum-behaved particle swarm optimization algorithm-based multi-region economic dispatch method
CN106251043A (en) * 2016-07-20 2016-12-21 广东工业大学 Multizone economic environment method for optimizing scheduling and device
CN106339817A (en) * 2016-08-29 2017-01-18 广东工业大学 Multi-region dynamic economy scheduling method and system
CN106355277A (en) * 2016-08-30 2017-01-25 广东工业大学 Multi-area dynamic economic dispatch optimization method and system

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108565857A (en) * 2018-05-07 2018-09-21 江南大学 A kind of Economic Dispatch method based on information interchange strategy ACS in continuous space
CN108565857B (en) * 2018-05-07 2021-01-22 江南大学 Electric power system scheduling method based on information exchange strategy continuous domain ant colony algorithm
WO2023065399A1 (en) * 2021-10-22 2023-04-27 中冶南方工程技术有限公司 Blast furnace batching optimization method based on differential evolution algorithm, and electronic device and storage medium

Similar Documents

Publication Publication Date Title
CN107067121A (en) A kind of improvement grey wolf optimized algorithm based on multiple target
Naderi et al. A comprehensive study of practical economic dispatch problems by a new hybrid evolutionary algorithm
CN106599519B (en) The cut deal motherboard and slab collaborative design method and system of facing to manufacture order combination optimization
Han et al. Information-utilization-method-assisted multimodal multiobjective optimization and application to credit card fraud detection
CN105719091B (en) A kind of parallel Multiobjective Optimal Operation method of Hydropower Stations
CN106971049A (en) A kind of new multi objective optimization method of catalytic cracking piece-rate system
Sedighizadeh et al. Hybrid approach to FACTS devices allocation using multi-objective function with NSPSO and NSGA-II algorithms in Fuzzy framework
CN105487496B (en) The optimization method of Power Plant Thermal on-line process identification and control algolithm based on Bi-objective Parallel I SLAND-HFC mixed model genetic programming algorithms
CN108122079A (en) Computational methods, system and the storage medium of thermal power plant's sharing of load
CN106372756A (en) Thermal power plant load optimization distribution method based on breeding particle swarm optimization
CN106815660B (en) User load combination optimization method based on simulated annealing algorithm
CN107400935A (en) Adjusting method based on the melt-spinning technology for improving ELM
CN106505624B (en) Determine the regulator control system and method for the optimal ability to arrange jobs of power distribution network distributed generation resource
CN112906292A (en) Method, system, equipment and storage medium for plant-level heat and power load online optimal distribution of cogeneration unit
CN106992551A (en) Photovoltaic inversion controller parameter discrimination method based on fuzzy C-mean algorithm and differential evolution hybrid algorithm
CN111416356A (en) Transmission and distribution network linkage optimization method based on alternating direction multiplier method and optimal power flow
Shang et al. Production scheduling optimization method based on hybrid particle swarm optimization algorithm
CN108710970A (en) A kind of parallel dimension reduction method of Multiobjective Scheduling of huge Hydro Power Systems with Cascaded Reservoirs
Liu et al. Opposition-mutual learning differential evolution with hybrid mutation strategy for large-scale economic load dispatch problems with valve-point effects and multi-fuel options
CN107025501A (en) A kind of multi fuel economic load dispatching optimization method based on gene editing difference algorithm
CN104732067A (en) Industrial process modeling forecasting method oriented at flow object
CN108647772A (en) A method of it is rejected for slope monitoring data error
CN108448659A (en) A method of reactive power optimization of power system is solved based on HACS algorithms
CN108154266A (en) A kind of algorithm of Economic Dispatch containing wind-powered electricity generation
He et al. Biobjective optimization-based frequency regulation of power grids with high-participated renewable energy and energy storage systems

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
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20170808