CN106326663B - A kind of coal yard based on coal petrology and genetic algorithm divides heaping method - Google Patents
A kind of coal yard based on coal petrology and genetic algorithm divides heaping method Download PDFInfo
- Publication number
- CN106326663B CN106326663B CN201610768559.8A CN201610768559A CN106326663B CN 106326663 B CN106326663 B CN 106326663B CN 201610768559 A CN201610768559 A CN 201610768559A CN 106326663 B CN106326663 B CN 106326663B
- Authority
- CN
- China
- Prior art keywords
- coal
- fitness
- individual
- population
- heap
- 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.)
- Expired - Fee Related
Links
Classifications
-
- 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
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16Z—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS, NOT OTHERWISE PROVIDED FOR
- G16Z99/00—Subject matter not provided for in other main groups of this subclass
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- Genetics & Genomics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Physiology (AREA)
- Feedback Control In General (AREA)
Abstract
The present invention relates to a kind of coal yards based on coal petrology and genetic algorithm to divide heaping method.The step of its technical solution is:Step 1 measures the random reflec tance of vitrinite distribution proportion of each dump in coal yard and measures the random reflec tance of vitrinite distribution proportion for carrying out coal;Step 2: establishing coal yard divides heap fitness mathematical modelFitness;Step 3 divides heap fitness mathematical model to seek optimal coal yard and divides heap scheme using genetic algorithm and coal yard;Step 4 divides heap fitness mathematical model according to most optimal sorting heap scheme with coal yardFitnessObtain the fitness of most optimal sorting heap scheme;If each overlapping angle value >=30% in the most fitness of optimal sorting heap scheme, meet production requirement;If there are overlap angle value < 30%, return to step three in the most fitness of optimal sorting heap scheme.The present invention divides the registration of each dump and the random reflec tance of vitrinite distribution proportion for carrying out coal of each dump to heap fitness mathematical model as coal yard, has the characteristics that be simple and efficient and being capable of rapid Optimum.
Description
Technical field
The invention belongs to coal yards to divide heap technical field.Divide heap more particularly to a kind of coal yard based on coal petrology and genetic algorithm
Method.
Background technology
Coal chemical enterprise is limited by place, mostly uses the guidance of the coals classification indicators such as volatile matter, caking index, thickness of colloidal matter layer greatly
It stacks, the same or similar dump that comes of these indexs is put together.But after These parameters are to carrying out coal stock pile stacked, often match
Coal and the unstable situation of coke quality, reason are that these indexs can not differentiate the single mixcibility for carrying out coal, and minority is carried out dump and put not
It can rationally cause coal blending quality chaotic.Mix caused by stacking process can make in the form of single coal stack dump formed it is mixed
Coal makes coal-blending coking theory of foundation on the basis of single coal be difficult to adapt to, meanwhile, it will also result in the wave to high-quality caking coal resource
Take.And can differentiate single coal and mixed coal well first using the random reflec tance of vitrinite distribution proportion index of coal, secondly,
On the basis of the similarity degree for comparing the random reflec tance of vitrinite distribution proportion for carrying out coal, coal yard can be instructed to carry out the reasonable heap of coal
It puts.
In recent years, the introduction with coal chemical enterprise to the lithofacies analysis instrument of coal so that coal chemical enterprise is to carrying out the mirror matter of coal
Group random reflectance distribution proportion measurement become more quickly with it is convenient.Meanwhile the random reflec tance of vitrinite of coal is distributed
The application study of figure is also more and more common, such as Yan Ruihua research (the vitrinite reflectances such as Yan Ruihua, Gao Zhijun, Geng Yinquan
Application [J] fuel and chemical industry of the distribution map in coal-blending coking, 2001,32 (5):227-230.), it is proposed that utilize the mirror of coal
Matter group distribution graph of reflectivity differentiates mixed coal, coal yard is instructed to divide heap and optimizing blending plan, and achieves good benefit;For another example Yao
(Yao Baiyuan, Wu Yadong instruct coal yard come rational stacking [J] the fuel of coal and change with coal distribution graph of reflectivity for primary yuan etc. of research
Work, 2008,39 (3):9-14.), by defining come coal divorced value, come coal distribution receiving degree Fi (%) and come coal reflectivity
Distribution map degree of overlapping Ci (%) index, come the rational stacking for instructing coal yard to carry out coal integrated.
Though existing point of heaping method has many good qualities, there is also following deficiencies in coal chemical enterprise actual production:When
Simplification and timeliness that coal yard divides heap are not accounted for;Second is that not accounting for when the coal for carrying out coal is complicated and quantity is more, manually
Matched complexity.
Invention content
The present invention is directed to overcome the deficiencies of existing technologies, it is therefore an objective to provide a kind of be simple and efficient with rapid Optimum based on coal
The coal yard of rock and genetic algorithm divides heaping method.
To achieve the above object, the step of the technical solution adopted by the present invention is:
Step 1, measure coal yard in each dump random reflec tance of vitrinite distribution proportion and measure come coal vitrinite with
Machine reflectivity distribution ratio.
Step 2 establishes coal yard and divides heap fitness mathematical model Fitness;
Fitness=[f1,f2,...,fi,...,fM] (1)
In formula (1):M indicates the dump number in coal yard;
I indicates natural number, i=1,2,3 ..., M;
fiIndicate the registration of the random reflec tance of vitrinite distribution proportion for coming coal and No. i-th dump of No. i-th dump,
In formula (2):I indicates natural number, i=1,2,3 ..., M;
M indicates the dump number in coal yard;
N No. i-th dump of expression carrys out coal number;
Rei,jIndicate numerical value of the random reflec tance of vitrinite distribution proportion in jth point of No. i-th dump;
J expression natural numbers, j=1,2,3 ..., 50;
K indicates natural number, k=1,2,3 ..., n;
xi-k,jIndicate that the kth kind of No. i-th dump carrys out numerical value of the random reflec tance of vitrinite distribution proportion in jth point of coal;
Min expressions are minimized.
Step 3 divides heap fitness mathematical model Fitness to seek optimal coal yard to divide heap side using genetic algorithm and coal yard
Case comprises the concrete steps that:
Step 3.1 sets and carrys out coal number as D;Using integer coding, primary population at individual is built in [1, M] section, is obtained
Primary population;The form of primary population at individual is:
αt=[b1,b2,...,bD] (3)
In formula (3), t indicates the serial number of primary population at individual, and t is natural number, t=1,2,3 ..., P;
P indicates primary population at individual sum;
B indicates the gene loci of primary population at individual, b ∈ N ∩ b ∈ [1, M];
D indicates to carry out coal number.
Step 3.2 carries out crossover operation using the single-point cross method in genetic algorithm to primary population at individual, intersects general
Rate Pc=0.6;Mutation operation is carried out to primary population at individual by the way of mutation probability, mutation probability Pm=0.2 is formed and lost
Population at individual is passed, genetic groups are obtained.
Primary population and genetic groups are merged into total population by step 3.3;Divide heap fitness mathematical model using coal yard
Fitness calculates the fitness of total population at individual.
Step 3.4 obtains Pareto optimum individuals from total population;Judge whether Pareto optimum individuals sum is more than just
It takes and randomly selects if Pareto optimum individual sums are greater than or equal to primary population at individual sum for population at individual sum
Mode selected from Pareto optimum individuals progeny population individual;If Pareto optimum individual sums are less than primary population
Body sum, then all choose Pareto optimum individuals in progeny population, remaining individual is in [1, M] section in progeny population
Structure.
Step 3.5, calculate each Pareto optimum individual fitness addition and value, select the addition and value of fitness
Maximum Pareto optimum individuals are contemporary optimum individual, preserve contemporary optimum individual.
Step 3.6 removes individual all in primary population, and primary population is filled into individual all in progeny population
In;Judge whether primary population iterations are more than maximum iteration, if primary population iterations are more than greatest iteration time
Several then algorithms terminate, and calculate the addition and value of the fitness of each contemporary optimum individual, and the addition and value for selecting fitness is maximum
Contemporary optimum individual be optimal coal yard divide heap scheme;It is returned if primary population iterations are less than or equal to maximum iteration
Return step 3.2.
Step 4 divides heap scheme according to optimal coal yard, divides heap fitness mathematical model Fitness to calculate most using coal yard
Excellent coal yard divides the fitness of heap scheme;If optimal coal yard divides each in the fitness of heap scheme to overlap angle value and is greater than or equal to
30%, then meet production requirement;If optimal coal yard divides in the fitness of heap scheme there are overlapping angle value to be less than 30%, it is discontented with
Sufficient production requirement, return to step 3 are greater than or equal to up to each in the fitness that optimal coal yard divides heap scheme overlaps angle value
30%.
Due to the adoption of the above technical scheme, compared with prior art, the present invention it significantly has the beneficial effect that:
1, the present invention proposes overlapping each dump and the random reflec tance of vitrinite distribution proportion for carrying out coal of each dump
Degree divides heap fitness mathematical model Fitness, the model that can go out the quality that coal yard divides heap scheme with efficient evaluations as coal yard, from
And accelerate the selection that optimal coal yard divides heap scheme.
2, the present invention is used divides heap scheme with Pareto theories with the method optimizing coal yard that genetic algorithm is combined, the algorithm
It is stable convergence in 100~300 ranges in iterations, to realize that coal yard divides the rapid Optimum of heap scheme.
3, the present invention since coal random reflec tance of vitrinite distribution proportion and dump random reflec tance of vitrinite distribution
Ratio is not necessarily to other evaluation indexes, after genetic algorithm optimization as input, you can and it obtains optimal coal yard and divides heap scheme, to
It realizes and is simple and efficient.
Therefore, the present invention has the characteristics that be simple and efficient and being capable of rapid Optimum.
Description of the drawings:
Fig. 1 is the random reflec tance of vitrinite distribution histogram for carrying out coal of No. 1 dump and No. 1 dump of the present invention;
Fig. 2 is the random reflec tance of vitrinite distribution histogram for carrying out coal of No. 2 dumps and No. 2 dumps of the present invention;
Fig. 3 is the random reflec tance of vitrinite distribution histogram for carrying out coal of No. 3 dumps and No. 3 dumps of the present invention;
Fig. 4 is the random reflec tance of vitrinite distribution histogram for carrying out coal of No. 4 dumps and No. 4 dumps of the present invention;
Fig. 5 is the random reflec tance of vitrinite distribution histogram for carrying out coal of No. 5 dumps and No. 5 dumps of the present invention.
Specific implementation mode:
The invention will be further described with reference to the accompanying drawings and detailed description, not to the limit of its protection domain
System.
Embodiment 1
A kind of coal yard based on coal petrology and genetic algorithm divides heaping method.This method comprises the concrete steps that:
Step 1, the present embodiment the dump number M of certain coal yard be 5, it is 14 that carry out coal number, which be D, measures in the coal yard 5
14 kinds of random reflec tance of vitrinite distribution proportions for carrying out coal of random reflec tance of vitrinite distribution proportion and measurement of dump are (due to coal
Random reflec tance of vitrinite distribution proportion data it is numerous and jumbled, therefore do not enumerate), the described 14 kinds coal classifications for carrying out coal are as shown in table 1.
The coal classification that 1 14 kinds of table carrys out coal
Step 2 establishes coal yard and divides heap fitness mathematical model Fitness.
Fitness=[f1,f2,f3,f4,f5] (1)
In formula (1), fiIndicate the random reflec tance of vitrinite distribution proportion for coming coal and No. i-th dump of No. i-th dump
Registration,
In formula (2), i is natural number, i=1,2,3,4,5;
N No. i-th dump of expression carrys out coal number;
J expression natural numbers, j=1,2,3 ..., 50;
Rei,jIndicate numerical value of the random reflec tance of vitrinite distribution proportion in jth point of No. i-th dump;
K indicates natural number, k=1,2,3 ..., n;
xi-k,jIndicate that the kth kind of No. i-th dump carrys out numerical value of the random reflec tance of vitrinite distribution proportion in jth point of coal;
Min expressions are minimized.
Step 3 divides heap fitness mathematical model Fitness to seek optimal coal yard to divide heap side using genetic algorithm and coal yard
Case is as follows:
Step 3.1, to carry out coal number D be 14, and setting primary population at individual sum P is 50, maximum iteration 300.It adopts
With integer coding, primary population at individual is built in [1,5] section, obtains primary population.The form of primary population at individual is:
αt=[b1,b2,...,b14] (3)
In formula (3), t is the serial number of primary population at individual, and t is natural number, t=1,2,3 ..., 50;
B is the gene loci of primary population at individual, b ∈ N ∩ b ∈ [1,5].
For example, No. 1 individual α of primary population1Gene it is as shown in table 2, No. 1 individual α of primary population1What is represented divides heap
Scheme is:The coal that comes of No. 1 dump is 3#, 9# and 14#, and the coal that comes of No. 2 dumps is 2#, 4#, 6# and 10#, and the coal that comes of No. 3 dumps is
The coal that comes of 5#, 12# and 13#, No. 4 dumps are 7# and 11#, and the coal that comes of No. 5 dumps is 1# and 8#.
No. 1 individual α of 2 primary population of table1Gene
Step 3.2 carries out crossover operation using the single-point cross method in genetic algorithm to primary population at individual, intersects general
Rate Pc=0.6;Mutation operation is carried out to primary population at individual by the way of mutation probability, mutation probability Pm=0.2 is formed and lost
Population at individual is passed, genetic groups are obtained.
Primary population and genetic groups are merged into total population by step 3.3;Divide heap fitness mathematical model using coal yard
Fitness calculates the fitness of total population at individual.
For example, No. 1 individual β of total population1Gene it is as shown in table 3.
No. 1 individual β of 3 total population of table1Gene
Heap fitness mathematical model Fitness is divided to calculate No. 1 individual β of total population using coal yard1Fitness be:
[6.6%, 75.1%, 1%, 0%, 0%].
Step 3.4 obtains Pareto optimum individuals from total population;Judge whether Pareto optimum individuals sum is more than just
It takes and randomly selects if Pareto optimum individual sums are greater than or equal to primary population at individual sum for population at individual sum
Mode selected from Pareto optimum individuals progeny population individual;If Pareto optimum individual sums are less than primary population
Body sum, then all choose Pareto optimum individuals in progeny population, remaining individual is in [1,5] section in progeny population
Structure.
Such as:No. 2 individual β of total population2No. 3 individual β of gene and total population3Gene it is as shown in table 4.
No. 2 individual β of 4 total population of table2No. 3 individual β of gene and total population3Gene
Heap fitness mathematical model Fitness is divided to calculate separately out using coal yard:No. 2 individual β of total population2Fitness
For [0%, 0%, 2.9%, 0%, 0%];No. 3 individual β of total population3Fitness be [7.3%, 0%, 8.4%, 0%,
31.4%].It can be seen that No. 3 individual β of total population3Fitness in respectively overlap angle value accordingly be greater than or equal to total kind
No. 2 individual β of group2Fitness in respectively overlap angle value, then No. 3 individual β of total population3For Pareto optimum individuals.
Step 3.5, calculate each Pareto optimum individual fitness addition and value, select the addition and value of fitness
Maximum Pareto optimum individuals are contemporary optimum individual, preserve contemporary optimum individual.
Step 3.6 removes individual all in primary population, and primary population is filled into individual all in progeny population
In;Judge whether primary population iterations are more than maximum iteration, if primary population iterations are more than greatest iteration time
Several then algorithms terminate, and calculate the addition and value of the fitness of each contemporary optimum individual, and the addition and value for selecting fitness is maximum
Contemporary optimum individual be optimal coal yard divide heap scheme;It is returned if primary population iterations are less than or equal to maximum iteration
Return step 3.2.
The step 3 of the present embodiment carrys out coal using 5 dumps and 14 kinds and divides heap object as the optimization of genetic algorithm, which exists
When iterations were 100 generation, algorithmic stability convergence.When iterations are 300, algorithm terminates, and the optimal coal yard of output divides heap side
Case is as shown in table 5.
5 optimal coal yard of table divides heap scheme
As can be seen from Table 5, optimal coal yard divides the heap scheme to be:The coal that comes of No. 1 dump is 2# and 7#;No. 2 dumps carry out coal
It is 1#, 3# and 5#;The coal that comes of No. 3 dumps is 9#, 10# and 11#;The coal that comes of No. 4 dumps is 12#, 13# and 14#;No. 5 dumps
It is 4#, 6# and 8# to carry out coal.
Step 4 divides heap scheme according to optimal coal yard, divides heap fitness mathematical model Fitness to calculate most using coal yard
Excellent coal yard divides the fitness of heap scheme:
The registration calculation formula of (1) No. 1 dump and 2# and 7# the random reflec tance of vitrinite distribution proportion for carrying out coal is:
In formula (4), j expression natural numbers, j=1,2,3 ..., 50;
Re1,jFor No. 1 dump random reflec tance of vitrinite distribution proportion jth point numerical value;
x1-1,jCarry out numerical value of the random reflec tance of vitrinite distribution proportion in jth point of coal for the 2# of No. 1 dump;
x1-2,jCarry out numerical value of the random reflec tance of vitrinite distribution proportion in jth point of coal for the 7# of No. 1 dump;
Min expressions are minimized.
The random reflec tance of vitrinite distribution proportion that No. 1 dump carrys out coal with 2# and 7# is as shown in table 6.
6 No. 1 dumps of table carry out the random reflec tance of vitrinite distribution proportion of coal with 2# and 7#
By comparing, the minimum value of each row is taken.Therefore, No. 1 dump carrys out the random reflec tance of vitrinite point of coal with 2# and 7#
The registration f of cloth ratio1=1%+15.7%+18%+6.7%=41.4%.No. 1 dump and 2# and 7# are come to the vitrinite of coal
Random reflectance distribution proportion is mapped, as shown in Figure 1.It will be seen from figure 1 that coal classification for the 2# of 1/3 coking coal come coal,
The random reflec tance of vitrinite distribution registration that coal classification comes coal and No. 1 dump for the 7# of bottle coal is larger, and satisfaction divides heap
It is required that.
(2) No. 2 dumps carry out the registration calculation formula of the random reflec tance of vitrinite distribution proportion of coal with 1#, 3# and 5#
For:
In formula (5), j expression natural numbers, j=1,2,3 ..., 50;
Re2,jFor No. 2 dumps random reflec tance of vitrinite distribution proportion jth point numerical value;
x2-1,jCarry out numerical value of the random reflec tance of vitrinite distribution proportion in jth point of coal for the 1# of No. 2 dumps;
x2-2,jCarry out numerical value of the random reflec tance of vitrinite distribution proportion in jth point of coal for the 3# of No. 2 dumps;
x2-3,jCarry out numerical value of the random reflec tance of vitrinite distribution proportion in jth point of coal for the 5# of No. 2 dumps;
Min is to be minimized.
Carry out the method phase of the registration of the random reflec tance of vitrinite distribution proportion of coal with 2# and 7# with No. 1 dump of calculating
Together, No. 2 dumps carry out the registration f of the random reflec tance of vitrinite distribution proportion of coal with 1#, 3# and 5#2It is 68.4%.By No. 2 coals
The random reflec tance of vitrinite distribution proportion that heap carrys out coal with 1#, 3# and 5# is mapped, as shown in Figure 2.Figure it is seen that
Coal classification is that the 1# of rich coal carrys out coal, the 5# that the 3# that coal classification is rich coal carrys out coal, coal classification is 1/3 coking coal comes coal and No. 2
The random reflec tance of vitrinite distribution of dump essentially coincides, and satisfaction divides heap requirement.
(3) carry out the method for the registration of the random reflec tance of vitrinite distribution proportion of coal with 2# and 7# with No. 1 dump of calculating
Identical, No. 3 dumps carry out the registration f of the random reflec tance of vitrinite distribution proportion of coal with 9#, 10# and 11#3It is 59.9%.By 3
The random reflec tance of vitrinite distribution proportion that number dump carrys out coal with 9#, 10# and 11# is mapped, as shown in Figure 3.It can be with from Fig. 3
Find out, coal classification be that the 9# of coking coal carrys out coal, the 11# that the 10# that coal classification is coking coal carrys out coal, coal classification is coking coal carrys out coal and
The random reflec tance of vitrinite distribution of No. 3 dumps essentially coincides, and satisfaction divides heap requirement.
(4) carry out the method for the registration of the random reflec tance of vitrinite distribution proportion of coal with 2# and 7# with No. 1 dump of calculating
Identical, No. 4 dumps carry out the registration f of the random reflec tance of vitrinite distribution proportion of coal with 12#, 13# and 14#4It is 41.9%.It will
The random reflec tance of vitrinite distribution proportion that No. 4 dumps carry out coal with 12#, 13# and 14# is mapped, as shown in Figure 4.It can from Fig. 4
To find out, coal classification is that the 14# that the 12# of lean coal carrys out coal, the 13# that coal classification is lean coal carrys out coal, coal classification is lean coal comes
The random reflec tance of vitrinite distribution of coal and No. 4 dumps essentially coincides, and satisfaction divides heap requirement.
(5) carry out the method for the registration of the random reflec tance of vitrinite distribution proportion of coal with 2# and 7# with No. 1 dump of calculating
Identical, No. 5 dumps carry out the registration f of the random reflec tance of vitrinite distribution proportion of coal with 4#, 6# and 8#5It is 57.3%.By No. 4
The random reflec tance of vitrinite distribution proportion that dump carrys out coal with 4#, 6# and 8# is mapped, as shown in Figure 5.It can from Fig. 5
Go out, coal classification be the 4# of 1/3 coking coal carry out coal, the 8# that the 6# that coal classification is 1/3 coking coal carrys out coal, coal classification is 1/3 coking coal
The random reflec tance of vitrinite distribution for coming coal and No. 5 dumps essentially coincides, and satisfaction divides heap requirement.
As can be seen that each coincidence angle value that the optimal coal yard of the present embodiment divides in the fitness of heap scheme is all higher than 30%, therefore
Meet production requirement.
It is worth noting that:
Due to the registration computational methods phase of each dump and the random reflec tance of vitrinite distribution proportion for carrying out coal of the dump
Together, present embodiment by do not repeat each dump in other embodiment it is corresponding with the dump come coal vitrinite it is anti-at random
Penetrate the calculating of the registration of rate distribution proportion.
It is same as Example 1, if each in the fitness of the most optimal sorting heap scheme of other embodiment overlaps angle value and is more than
Or be equal to 30%, then meet production requirement;If there are overlap angle value in the fitness of the most optimal sorting heap scheme of other embodiment
When less than 30%, then production requirement, return to step 3, until each registration in the most fitness of optimal sorting heap scheme are unsatisfactory for
Value is greater than or equal to 30%.
Present embodiment compared with prior art, significantly has the beneficial effect that:
1, the present invention proposes overlapping each dump and the random reflec tance of vitrinite distribution proportion for carrying out coal of each dump
Degree divides heap fitness mathematical model Fitness, the model that can go out the quality that coal yard divides heap scheme with efficient evaluations as coal yard, from
And accelerate the selection that optimal coal yard divides heap scheme.
2, the present invention is used divides heap scheme with Pareto theories with the method optimizing coal yard that genetic algorithm is combined, the algorithm
It is stable convergence in 100~300 ranges in iterations, to realize that coal yard divides the rapid Optimum of heap scheme.
3, the present invention since coal random reflec tance of vitrinite distribution proportion and dump random reflec tance of vitrinite distribution
Ratio is not necessarily to other evaluation indexes, after genetic algorithm optimization as input, you can and it obtains optimal coal yard and divides heap scheme, to
The method of realizing is simple and efficient.
Therefore, the present invention has the characteristics that be simple and efficient and being capable of rapid Optimum.
Claims (1)
1. a kind of coal yard based on coal petrology and genetic algorithm divides heaping method, it is characterised in that the coal yard divides the specific of heaping method
Step is:
Step 1, measure coal yard in each dump random reflec tance of vitrinite distribution proportion and measure come coal vitrinite it is anti-at random
Penetrate rate distribution proportion;
Step 2 establishes coal yard and divides heap fitness mathematical model Fitness;
Fitness=[f1,f2,...,fi,...,fM] (1)
In formula (1):M indicates the dump number in coal yard,
I expression natural numbers, i=1,2,3 ..., M,
fiIndicate vitrinite's random reflected of the random reflec tance of vitrinite distribution proportion for carrying out coal and No. i-th dump of No. i-th dump
The registration of rate distribution proportion,
In formula (2):I expression natural numbers, i=1,2,3 ..., M,
M indicates the dump number in coal yard,
N No. i-th dump of expression carrys out coal number,
Rei,jIndicate numerical value of the random reflec tance of vitrinite distribution proportion in jth point of No. i-th dump,
J expression natural numbers, j=1,2,3 ..., 50,
K expression natural numbers, k=1,2,3 ..., n,
xi-k,jIndicate that the kth kind of No. i-th dump carrys out numerical value of the random reflec tance of vitrinite distribution proportion in jth point of coal,
Min expressions are minimized;
Step 3 divides heap fitness mathematical model Fitness to seek optimal coal yard to divide heap scheme using genetic algorithm and coal yard, has
Body step is:
Step 3.1 sets and carrys out coal number as D;Using integer coding, primary population at individual is built in [1, M] section, obtains primary
Population;The form of primary population at individual is:
αt=[b1,b2,...,bD] (3)
In formula (3), t indicates that the serial number of primary population at individual, t are natural number, t=1,2,3 ..., P,
P indicates primary population at individual sum,
B indicates the gene loci of primary population at individual, b ∈ N and b ∈ [1, M],
D indicates to carry out coal number;
Step 3.2 carries out crossover operation, crossover probability Pc using the single-point cross method in genetic algorithm to primary population at individual
=0.6;Mutation operation is carried out to primary population at individual by the way of mutation probability, mutation probability Pm=0.2 forms heredity kind
Group's individual, obtains genetic groups;
Primary population and genetic groups are merged into total population by step 3.3;Divide heap fitness mathematical model using coal yard
Fitness calculates the fitness of total population at individual;
Step 3.4 obtains Pareto optimum individuals from total population;Judge whether Pareto optimum individuals sum is more than primary kind
Group's individual sum takes the side randomly selected if Pareto optimum individual sums are greater than or equal to primary population at individual sum
Formula selects progeny population individual from Pareto optimum individuals;If it is total that Pareto optimum individual sums are less than primary population at individual
Number, then all choose Pareto optimum individuals in progeny population, remaining individual structure in [1, M] section in progeny population
It builds;
Step 3.5, calculate each Pareto optimum individual fitness addition and value, the addition and value for selecting fitness is maximum
Pareto optimum individuals be contemporary optimum individual, preserve contemporary optimum individual;
Step 3.6 removes individual all in primary population, is filled into primary population with individual all in progeny population;
Judge whether primary population iterations are more than maximum iteration, if primary population iterations are more than maximum iteration
Algorithm terminates, and calculates the addition and value of the fitness of each contemporary optimum individual, and the addition and value for selecting fitness maximum is worked as
It is that optimal coal yard divides heap scheme for optimum individual;Step is returned if primary population iterations are less than or equal to maximum iteration
Rapid 3.2;
Step 4 divides heap scheme according to optimal coal yard, divides heap fitness mathematical model Fitness to calculate optimal coal using coal yard
The fitness of heap scheme is divided in field;If optimal coal yard divides each in the fitness of heap scheme to overlap angle value and is greater than or equal to 30%,
Then meet production requirement;If optimal coal yard divides in the fitness of heap scheme there are overlapping angle value to be less than 30%, it is unsatisfactory for giving birth to
Production demand, return to step 3 are greater than or equal to 30% up to each in the fitness that optimal coal yard divides heap scheme overlaps angle value.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610768559.8A CN106326663B (en) | 2016-08-30 | 2016-08-30 | A kind of coal yard based on coal petrology and genetic algorithm divides heaping method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610768559.8A CN106326663B (en) | 2016-08-30 | 2016-08-30 | A kind of coal yard based on coal petrology and genetic algorithm divides heaping method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106326663A CN106326663A (en) | 2017-01-11 |
CN106326663B true CN106326663B (en) | 2018-08-21 |
Family
ID=57789937
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610768559.8A Expired - Fee Related CN106326663B (en) | 2016-08-30 | 2016-08-30 | A kind of coal yard based on coal petrology and genetic algorithm divides heaping method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106326663B (en) |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102816577A (en) * | 2012-08-30 | 2012-12-12 | 武汉科技大学 | Coal-rock blending method suitable for tamping coking |
CN105243437A (en) * | 2015-09-21 | 2016-01-13 | 武汉科技大学 | Method for predicting coke quality and optimizing coal blending ratio for tamping coking |
-
2016
- 2016-08-30 CN CN201610768559.8A patent/CN106326663B/en not_active Expired - Fee Related
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102816577A (en) * | 2012-08-30 | 2012-12-12 | 武汉科技大学 | Coal-rock blending method suitable for tamping coking |
CN105243437A (en) * | 2015-09-21 | 2016-01-13 | 武汉科技大学 | Method for predicting coke quality and optimizing coal blending ratio for tamping coking |
Non-Patent Citations (9)
Title |
---|
"Estimation of Some Coal Parameters Depending on Petrographic and Inorganic Analyses by Using Genetic Algorithm and Adaptive Neuro-Fuzzy Inference Systems";S. Chehreh Chelgani等;《ENERGY EXPLORATION & EXPLOITATION》;20110901;第29卷(第4期);第479-494页 * |
"Study on the Effect of Adding Coke Powder on Coke Quality for Coal Blending";Qigang Zhang等;《Advanced Materials Research》;20111122;第396-398卷;第1106-1110页 * |
"一种遗传算法在焦化配煤优化方案中的应用研究";杨俊杰;《中国优秀硕士学位论文全文数据库 工程科技Ⅰ辑》;20150815;第2015年卷(第8期);B017-1 * |
"基于遗传算法的炼焦配煤最优化配置研究";刘春梅;《制造业自动化》;20120229;第34卷(第2期);第113-117页 * |
"思维进化和支持向量机理论及其在炼焦配煤优化中的应用研究";吕青;《中国博士学位论文全文数据库 工程科技Ⅰ辑》;20110815;第2011年卷(第8期);B017-3 * |
"焦炭质量预测与优化配比算法的研究";陈红军;《中国优秀硕士学位论文全文数据库 经济与管理科学辑》;20080915;第2008年卷(第9期);J150-13 * |
"配合煤反射率分布图在指导炼焦配煤中的作用";姚伯元;《燃料与化工》;20080930;第39卷(第5期);第11-l5页 * |
"镜质组反射率分布图在配煤炼焦中的应用";燕瑞华等;《燃料与化工》;20010930;第32卷(第5期);第227-230页 * |
"镜质组反射率指导优化配煤炼焦方案的研究";田英奇等;《煤炭科学技术》;20160430;第44卷(第4期);第162-168页 * |
Also Published As
Publication number | Publication date |
---|---|
CN106326663A (en) | 2017-01-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Karasakal et al. | A multicriteria sorting approach based on data envelopment analysis for R&D project selection problem | |
CN105243437B (en) | The method of prediction coke quality and Optimized Coal Blending ratio for tamping coking | |
Ren et al. | Intuitionistic multiplicative analytic hierarchy process in group decision making | |
CN109118067A (en) | A kind of Renewable Energy Development potential evaluation method | |
CN105678607A (en) | Order batching method based on improved K-Means algorithm | |
Zoroja et al. | Impact of information and communication technology to the competitiveness of European countries-cluster analysis approach | |
Kadziński et al. | DIS-CARD: a new method of multiple criteria sorting to classes with desired cardinality | |
CN110111024A (en) | Scientific and technological achievement market valuation method based on AHP model of fuzzy synthetic evaluation | |
CN107526975A (en) | A kind of method based on difference secret protection decision tree | |
CN108510050A (en) | It is a kind of based on shuffling the feature selection approach to leapfrog | |
CN102799954A (en) | Method and system for multi-objective optimization applied to risk assessment | |
Köksalan et al. | A DEA-based approach to ranking multi-criteria alternatives | |
CN106326663B (en) | A kind of coal yard based on coal petrology and genetic algorithm divides heaping method | |
CN105976099A (en) | Fuzzy information-based aerospace model scientific research and production management level evaluation method | |
CN105956741A (en) | Intelligent power grid technology maturity evaluation method based on TRL | |
CN109492931A (en) | A kind of determining method of railway speed target value scheme evaluation | |
CN107730072A (en) | Based on the parallel machine gang scheduling method and system for improving Artificial Immune Algorithm | |
CN104182853A (en) | Pilot assignment model establishing and solving method | |
CN105727777B (en) | A kind of heavy bunker fuel oil beending optimization method | |
CN106651094A (en) | Enterprise efficiency evaluation method based on equivalent energy saving amount | |
Nielsen | Global relative poverty | |
CN110472877A (en) | A kind of comprehensive performance evaluation method of chain products | |
CN105335799A (en) | Off-peak power consumption plan compiling method based on mixed multi-attribute evaluation | |
CN107832933A (en) | A kind of index simplifying method of power distribution network input-output evaluation system | |
CN107180308A (en) | The method for building up of water transport engineering project market credit system model |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
C10 | Entry into 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 | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180821 Termination date: 20190830 |