CN110309901A - The improvement artificial bee colony data processing method of solving condition nonlinear optimal perturbation - Google Patents
The improvement artificial bee colony data processing method of solving condition nonlinear optimal perturbation Download PDFInfo
- Publication number
- CN110309901A CN110309901A CN201910410054.8A CN201910410054A CN110309901A CN 110309901 A CN110309901 A CN 110309901A CN 201910410054 A CN201910410054 A CN 201910410054A CN 110309901 A CN110309901 A CN 110309901A
- Authority
- CN
- China
- Prior art keywords
- food source
- bee colony
- data processing
- artificial bee
- solving condition
- 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
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
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- General Preparation And Processing Of Foods (AREA)
Abstract
The present invention relates to a kind of improvement artificial bee colony data processing methods of solving condition nonlinear optimal perturbation, belong to computer and Atmosphere and Ocean subject crossing field, can be used for the predictability research of the Numerical Weather and climatic phenomenon of gas marine field.The present invention provides the condition nonlinear optimal perturbations (CNOP) that a kind of improvement artificial bee colony algorithm for exempting from gradient information is used for Efficient Solution Atmosphere and Ocean field, on the basis of original artificial bee colony algorithm, this problem is solved for CNOP, carried out the improvement of adaptability, main inclusive fitness function and select probability be adaptively adjusted and feature extraction and the increase of constraint condition etc..Compared with prior art, the present invention has many advantages, such as that accuracy height, solving speed are fast.
Description
Technical field
It is non-thread more particularly, to a kind of solving condition the present invention relates to computer and Atmosphere and Ocean subject crossing technical field
The improvement artificial bee colony data processing method of property Optimal Disturbance.
Background technique
Numerical Weather and weather predictability in atmosphere and Marine Sciences is carried out based on numerical model to study all the time all
It is the hot issue studied both at home and abroad.Condition nonlinear optimal perturbation (CNOP) method is because it is contemplated that non-linear gentle in weather
The effect in phenomenon is waited, the predictability research of weather and climatic phenomenon is widely used in.
The solution of CNOP is mathematically the process of the Nonlinear Optimization Problem of a solution Prescribed Properties, how to be counted
Calculating this Nonlinear Optimization Problem is the key that using CNOP technique study atmosphere and Marine Sciences nonlinear physics problem.Mesh
It is preceding generallyd use in meteorological field it is a kind of can handle constraint condition CNOP is solved based on the optimization algorithm of gradient information,
Middle objective function is highly dependent on adjoint mode about the acquisition of the gradient information of optimized variable, and the exploitation of adjoint mode and tests
Card generally requires to consume huge workload, and which greatly limits the extensive uses of CNOP.The present invention is intended to provide one kind exempts from ladder
The improvement artificial bee colony algorithm Efficient Solution CNOP of information is spent, to promote what Numerical Weather and weather predictability studied to open
Exhibition.
Summary of the invention
It is an object of the present invention to overcome the above-mentioned drawbacks of the prior art and provide a kind of solving condition is non-thread
The improvement artificial bee colony data processing method of property Optimal Disturbance.
The purpose of the present invention can be achieved through the following technical solutions:
A kind of improvement artificial bee colony data processing method of solving condition nonlinear optimal perturbation, solving condition are non-linear most
The overall process of excellent disturbance includes: first to carry out dimensionality reduction to sample data using principal component analytical method, will be representated by sample data
Original solution room be transformed into after feature space using artificial bee colony data processing method is improved, sought in feature space
It is excellent, it obtains optimal value and is finally used for the forecast of the Numerical Weather and climatic phenomenon in Atmosphere and Ocean field, the data processing
Method the following steps are included:
Step 1: initialization bee colony and numerical model；
Step 2: using employ bee search for food source；
Step 3: using observation bee according to setting probability selection food source, if the probability that food source is selected is general greater than setting
Rate, then the food source is selected, if otherwise being scanned for around the food source using observation bee；
Step 4: checking all food sources using investigation bee and carry out food source going or staying judgement；
Step 5: continuing iteration until being more than that greatest iteration step number or optimal value terminate iteration when remaining unchanged from step 2；
Step 6: the result after iteration being recorded as optimal value and reverts to luv space；
Step 7: utilizing the Numerical Weather and climatic phenomenon in the optimal value result Atmosphere and Ocean field for reverting to luv space
It is forecast.
Further, principal component of the feature space using accumulative characteristic value >=90%, the dimension of the feature space
Degree is the number of principal components of the principal component.
Further, the food source going or staying judgement in the step 4 specifically includes: if the searching times of a certain food source are super
It crosses after maximum search step number still without more preferably food source, the food source can be abandoned by investigating bee, and New food source is randomly generated and replaces
Generation former food source.
Further, the calculation formula for the probability that food source is selected in the step 3 are as follows:
In formula, piIndicate the probability that food source is selected, fiIndicate the fitness function value of i-th of food source,Indicate the fitness function maximum value in all food sources.
Further, the calculation formula of the fitness function value of i-th of food source are as follows:
fi=1+abs (f (i))
In formula, f (i) indicates i-th of food source for the target function value of solving condition nonlinear optimal perturbation.
Further, probability is set as the random number between 0 to 1 in the step 3.
Compared with prior art, the invention has the following advantages that
(1) precision is high, by original artificial bee colony data processing method in the data processing method in the present invention
Fitness function and the probability that is selected of food source be adjusted, be provided with for the specific fitness function of food source with
And income degree calculation formula, so that the precision for the solving condition nonlinear optimal perturbation being directed to is improved.
(2) solving speed is fast, and the present invention solves this problem on the basis of original artificial bee colony algorithm, for CNOP,
Carried out the improvement of adaptability, main inclusive fitness function and select probability be adaptively adjusted and feature extraction and constraint
The increase etc. of condition, so that solving speed gets a promotion.
Detailed description of the invention
Fig. 1 is the overall flow that artificial bee colony data processing method solving condition nonlinear optimal perturbation is improved in the present invention
Schematic diagram；
Fig. 2 is the optimizing flow diagram that artificial bee colony data processing method is improved in the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiment is a part of the embodiments of the present invention, rather than whole embodiments.Based on this hair
Embodiment in bright, those of ordinary skill in the art's every other reality obtained without making creative work
Example is applied, all should belong to the scope of protection of the invention.
Embodiment
A. as can be seen that the Integral Thought that improvement artificial bee colony algorithm of the invention solves CNOP is first to make from attached drawing 1
Dimensionality reduction is carried out to sample data with dimensionality reduction technologies such as PCA, the original solution room of higher-dimension representated by sample data is transformed into low
Then dimensional feature space carries out optimizing in feature space using improved artificial bee colony algorithm, finally obtains optimal value.
B. the honeybee that artificial bee colony algorithm searching process relates to three types is improved, is to employ bee (employed respectively
Bees), bee (onlooker bees) and investigation bee (scout bees) are observed, Each performs its own functions for these three honeybees, in attached drawing 2
Respective search operation is executed in searching process respectively.
C. the process that artificial bee colony algorithm is an iteration optimizing is improved, iteration can all record optimal food source each time
Position, if the number of iterations is more than greatest iteration step number (being denoted as maxIter) or the optimal value found is more than limits holding
Constant, optimizing terminates, and algorithm exports to obtain optimal value, and obtained optimal value is reverted to what former space obtained being solved
CNOP。
D. the initialization of searching process mainly includes the initialization of bee colony and numerical model.The initialization of bee colony mainly includes
The dimension (PCs) of feature space, maximum iterative steps maxIter, initial bee colony are (including bee colony number, initial food source
Position), optimal value keep maximum times limits；The initialization of numerical model is mainly mode input, integrates the first of duration
Beginningization etc..
E. in the iteration each time of algorithm, peak is employed to search for new food source near food source, if find
The fitness function of New food source is better than the fitness function of old food source, then old food source can be replaced.The plan of search
The search strategy of former artificial bee colony algorithm is slightly used.
F. observation bee can select food source according to the income degree of food source, the difference with original artificial bee colony algorithm, this
The Probability p that food source is selected in the improvement artificial bee colony algorithm of inventioniIt is calculated with following publicity, wherein fiIndicate i-th of food
The fitness function value in source.The income degree of probability i.e. food source that food source is selected, income degree is higher to represent food source
It is more excellent, then the probability selected is higher.
In formula, piIndicate the probability that food source is selected, fiIndicate the fitness function value of i-th of food source,Indicate the fitness function maximum value in all food sources.
Wherein, the calculation formula of the fitness function value of i-th of food source are as follows:
fi=1+abs (f (i))
In formula, f (i) indicates i-th of food source for the target function value of solving condition nonlinear optimal perturbation.
G. in every single-step iteration, bee is observed with certain probability (being denoted as P, value is the random number between 0 to 1) choosing
Food source is selected, and is scanned near the food source.Specific strategy is if the Probability p that food source is selectediGreater than P,
So the food source is observed bee and chooses, and then observes bee and searches for around it, if the fitness letter of the New food source found
Number is better than the fitness function of old food source, then old food source can be replaced by new food source.
H. maximum search step number (being denoted as maxSearch) defined for each food source, in every single-step iteration, investigation
Bee can check all food sources, once discovery is more than maxSearch to the searching times of a certain food source, still not find more
Then the food source is abandoned, and a new food source is randomly generated to replace original food source in excellent food source.
I. improvement artificial bee colony algorithm of the invention is in searching process, it is necessary to guarantee it is all employ bee, observation bee with
And investigation bee is searched in restriction range, in an iterative process, if the food source searched exceeds restrained boundary, it is necessary to will eat
Material resource is tied in solution room.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can readily occur in various equivalent modifications or replace
It changes, these modifications or substitutions should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with right
It is required that protection scope subject to.
Claims (6)
1. a kind of improvement artificial bee colony data processing method of solving condition nonlinear optimal perturbation, solving condition are non-linear optimal
The overall process of disturbance includes: first to carry out dimensionality reduction to sample data using principal component analytical method, will be representated by sample data
Original solution room is transformed into after feature space using artificial bee colony data processing method is improved, and carries out optimizing in feature space,
It obtains optimal value and is finally used for the forecast of the Numerical Weather and climatic phenomenon in Atmosphere and Ocean field, which is characterized in that should
Data processing method the following steps are included:
Step 1: initialization bee colony and numerical model；
Step 2: using employ bee search for food source；
Step 3: using observation bee according to setting probability selection food source, if the probability that food source is selected is greater than setting probability,
The food source is selected, if otherwise being scanned for around the food source using observation bee；
Step 4: checking all food sources using investigation bee and carry out food source going or staying judgement；
Step 5: continuing iteration until being more than that greatest iteration step number or optimal value terminate iteration when remaining unchanged from step 2；
Step 6: the result after iteration being recorded as optimal value and reverts to luv space；
Step 7: being carried out using the Numerical Weather and climatic phenomenon in the optimal value result Atmosphere and Ocean field for reverting to luv space
Forecast.
2. a kind of improvement artificial bee colony data processing side of solving condition nonlinear optimal perturbation according to claim 1
Method, which is characterized in that the feature space uses the principal component of accumulative characteristic value >=90%, and the dimension of the feature space is
The number of principal components of the principal component.
3. a kind of improvement artificial bee colony data processing side of solving condition nonlinear optimal perturbation according to claim 1
Method, which is characterized in that the food source going or staying judgement in the step 4 specifically includes: if the searching times of a certain food source are more than
Still without more preferably food source after maximum search step number, the food source can be abandoned by investigating bee, and New food source substitution is randomly generated
Former food source.
4. a kind of improvement artificial bee colony data processing side of solving condition nonlinear optimal perturbation according to claim 1
Method, which is characterized in that the calculation formula for the probability that food source is selected in the step 3 are as follows:
In formula, piIndicate the probability that food source is selected, fiIndicate the fitness function value of i-th of food source,Table
Show the fitness function maximum value in all food sources.
5. a kind of improvement artificial bee colony data processing side of solving condition nonlinear optimal perturbation according to claim 4
Method, which is characterized in that the calculation formula of the fitness function value of i-th of food source are as follows:
fi=1+abs (f (i))
In formula, f (i) indicates i-th of food source for the target function value of solving condition nonlinear optimal perturbation.
6. a kind of improvement artificial bee colony data processing side of solving condition nonlinear optimal perturbation according to claim 1
Method, which is characterized in that set probability as the random number between 0 to 1 in the step 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910410054.8A CN110309901A (en) | 2019-05-16 | 2019-05-16 | The improvement artificial bee colony data processing method of solving condition nonlinear optimal perturbation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910410054.8A CN110309901A (en) | 2019-05-16 | 2019-05-16 | The improvement artificial bee colony data processing method of solving condition nonlinear optimal perturbation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110309901A true CN110309901A (en) | 2019-10-08 |
Family
ID=68075530
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910410054.8A Pending CN110309901A (en) | 2019-05-16 | 2019-05-16 | The improvement artificial bee colony data processing method of solving condition nonlinear optimal perturbation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110309901A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113221385A (en) * | 2021-06-08 | 2021-08-06 | 上海交通大学 | Initialization method and system for dating forecast |
-
2019
- 2019-05-16 CN CN201910410054.8A patent/CN110309901A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113221385A (en) * | 2021-06-08 | 2021-08-06 | 上海交通大学 | Initialization method and system for dating forecast |
| CN113221385B (en) * | 2021-06-08 | 2022-09-23 | 上海交通大学 | Initialization method and system for dating forecast |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Farkas et al. | Networks in life: Scaling properties and eigenvalue spectra | |
| Srinivasan | Microfinance India: state of the sector report 2011;(Microfinance in India (MFI)) | |
| Liu et al. | A global optimization algorithm for simulation-based problems via the extended DIRECT scheme | |
| AU2015347304B2 (en) | Testing insecure computing environments using random data sets generated from characterizations of real data sets | |
| Seely et al. | Application of a hybrid forest growth model to evaluate climate change impacts on productivity, nutrient cycling and mortality in a montane forest ecosystem | |
| TW201235865A (en) | Data structure for tiling and packetizing a sparse matrix | |
| CN109492355A (en) | A kind of software analysis resistant method and system based on deep learning | |
| He et al. | A study of Feynman integrals with uniform transcendental weights and their symbology | |
| CN110309901A (en) | The improvement artificial bee colony data processing method of solving condition nonlinear optimal perturbation | |
| CN107180043A (en) | Paging implementation method and paging system | |
| Zhao | Vegetation and climate reconstructions on different time scales in China: a review of Chinese palynological research | |
| CN106250453A (en) | The cipher text retrieval method of numeric type data based on cloud storage and device | |
| CN109977977A (en) | A kind of method and corresponding intrument identifying potential user | |
| Lin et al. | Learning to rank with groups | |
| Mauro et al. | Comparison of sampling methods for estimation of nearest-neighbor index values | |
| Wu et al. | Learning to rank using query-level regression | |
| CN108108596A (en) | A kind of method and apparatus for the digital finger-print for being used to generate writing | |
| Zhao et al. | Visibility graph analysis of the sea surface temperature irreversibility during El Niño events | |
| Lin et al. | Learning to rank with cross entropy | |
| Chuter et al. | A dynamical systems analysis of the data assimilation linked ecosystem carbon (DALEC) models | |
| Mkwizu et al. | Regional and Domestic Industrialisation Policies in Africa: Case Study of Two SADC States | |
| Yang et al. | Development Methodology of Web Crawling Based on Physical Properties DB of Building Materials for the Efficiency of Building Energy Simulation | |
| Wang et al. | An optimization algorithm of Bayesian network classifiers by derivatives of conditional log likelihood | |
| CN107741879A (en) | A kind of big data processing method and its device | |
| Amir Entekhabi et al. | Analysis of the effects of pilot project implementation on improving the quality of life in rural areas (Case study: Khalajistan ward, Qom County) |
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 |
Application publication date: 20191008 |
|
| RJ01 | Rejection of invention patent application after publication |