CN107292425A - Aquaculture dissolved oxygen prediction method and device - Google Patents
Aquaculture dissolved oxygen prediction method and device Download PDFInfo
- Publication number
- CN107292425A CN107292425A CN201710409845.XA CN201710409845A CN107292425A CN 107292425 A CN107292425 A CN 107292425A CN 201710409845 A CN201710409845 A CN 201710409845A CN 107292425 A CN107292425 A CN 107292425A
- Authority
- CN
- China
- Prior art keywords
- mrow
- dissolved oxygen
- msub
- sequence
- laypunov
- 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
- 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
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/08—Computing arrangements based on specific mathematical models using chaos models or non-linear system models
-
- 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
- G06Q50/02—Agriculture; Fishing; Mining
Abstract
The embodiment of the present invention provides a kind of aquaculture dissolved oxygen prediction method and device.Method includes:The dissolved oxygen sequence in collection prediction waters;Calculate the Laypunov exponent of dissolved oxygen sequence;If Laypunov exponent is more than 0, presetting method computing relay time and Embedded dimensions are utilized;Zero-mean processing is carried out to dissolved oxygen sequence, and is reconstructed according to time delay and Embedded dimensions, the phase space reconstruction of dissolved oxygen sequence is obtained;Predictor formula is calculated according to phase space reconstruction, and dissolved oxygen prediction value is obtained using obtained predictor formula.The embodiment of the present invention calculates time delay and Embedded dimensions and to dissolved oxygen sequence phase space reconstruction using presetting method, predictor formula is calculated according to phase space reconstruction, so as to obtain the predicted value of dissolved oxygen by gathering the dissolved oxygen sequence in waters.The predicted value to dissolved oxygen data is realized using less data sample, can small sample dissolved oxygen data be carried out with short-term accurate prediction.
Description
Technical field
The present embodiments relate to technical field of aquaculture, more particularly to a kind of aquaculture dissolved oxygen prediction method and
Device.
Background technology
China is aquaculture big country, and the guarantee of output of aquatic products and quality depends on cultivation water if appropriate for aquatic life
Thing is survived.Dissolved oxygen is aquatile existence essential condition, when dissolved oxygen concentration is less than 3mg/L, will cause fish
It is dead.Therefore, research cultivation water dissolved oxygen prediction method is significant.
Existing Forecasting Methodology has a lot, is broadly divided into two types:The first kind is traditional Forecasting Methodology, classical number
Learn as theoretical foundation, including time series forecasting, regression analysis, Markov model, simulation of water quality predicted method etc.;The
Two classes are the Forecasting Methodologies based on artificial intelligence, including gray model, neural network prediction method, Support vector regression are pre-
Survey method etc..These Forecasting Methodologies are required to be predicted on the basis of substantial amounts of dissolved oxygen sample is gathered, and prediction difficulty is larger
And Forecasting Methodology is complicated.
The prediction to waters dissolved oxygen can be realized using a small amount of dissolved oxygen sample there is presently no a kind of Forecasting Methodology,
Therefore it provides a kind of is current industry technology urgently to be resolved hurrily to the method that small sample water quality dissolved oxygen data are accurately predicted
Problem.
The content of the invention
In order to solve problems of the prior art, the embodiment of the present invention provides a kind of aquaculture dissolved oxygen prediction side
Method and device.
On the one hand, the embodiment of the present invention provides a kind of aquaculture dissolved oxygen prediction method, including:
The dissolved oxygen sequence in collection prediction waters;
Calculate the Laypunov exponent of the dissolved oxygen sequence;
If the Laypunov exponent is more than 0, presetting method computing relay time and Embedded dimensions are utilized;
Zero-mean processing is carried out to the dissolved oxygen sequence, and is reconstructed according to the time delay and Embedded dimensions,
Obtain the phase space reconstruction of the dissolved oxygen sequence;
Predictor formula is calculated according to the phase space reconstruction, and dissolved oxygen prediction is obtained using obtained predictor formula
Value.
On the other hand, the embodiment of the present invention provides a kind of aquaculture dissolved oxygen prediction device, it is characterised in that including:
Acquisition module, the dissolved oxygen sequence for gathering prediction waters;
First computing module, the Laypunov exponent for calculating the dissolved oxygen sequence;
Second computing module, for when the Laypunov exponent is more than 0, utilizing presetting method computing relay time and embedding
Enter dimension;
Reconstructed module, for carrying out zero-mean processing to the dissolved oxygen sequence, and according to the time delay and insertion
Dimension is reconstructed, and obtains the phase space reconstruction of the dissolved oxygen sequence;
3rd computing module, for calculating predictor formula according to the phase space reconstruction, and utilizes obtained prediction
Formula obtains dissolved oxygen prediction value.
Aquaculture dissolved oxygen prediction method and device provided in an embodiment of the present invention, by gathering the dissolved oxygen in waters
Sequence, calculates the Laypunov exponent of dissolved oxygen sequence, and presetting method meter is utilized if the Laypunov exponent of dissolved oxygen sequence is more than 0
Calculate time delay and Embedded dimensions and to dissolved oxygen sequence phase space reconstruction, prediction is calculated according to phase space reconstruction public
Formula, so as to obtain the predicted value of dissolved oxygen.The predicted value to dissolved oxygen data is realized using less data sample, can be to small
Sample dissolved oxygen data carry out short-term accurate prediction.
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 this hairs
Some bright embodiments, for those of ordinary skill in the art, on the premise of not paying creative work, can be with root
Other accompanying drawings are obtained according to these accompanying drawings.
Fig. 1 is aquaculture dissolved oxygen prediction method flow schematic diagram provided in an embodiment of the present invention;
Fig. 2 is aquaculture dissolved oxygen prediction apparatus structure schematic diagram provided in an embodiment of the present invention.
Embodiment
To make the purpose, technical scheme and advantage of the embodiment of the present invention clearer, below in conjunction with the embodiment of the present invention
In accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is
A part of embodiment of the present invention, rather than whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art
The every other embodiment obtained under the premise of creative work is not made, belongs to the scope of protection of the invention.
Fig. 1 is aquaculture dissolved oxygen prediction method flow schematic diagram provided in an embodiment of the present invention, as shown in figure 1, side
Method includes:
Step 10, the dissolved oxygen sequence in collection prediction waters;
Step 20, the Laypunov exponent for calculating the dissolved oxygen sequence;
If step 30, the Laypunov exponent are more than 0, presetting method computing relay time and Embedded dimensions are utilized;
Step 40, zero-mean processing is carried out to the dissolved oxygen sequence, and entered according to the time delay and Embedded dimensions
Line reconstruction, obtains the phase space reconstruction of the dissolved oxygen sequence;
Step 50, calculate predictor formula according to the phase space reconstruction, and obtain molten using obtained predictor formula
Solve oxygen predicted value.
Dissolved oxygen data are predicted by this method combination chaotic time Forecasting Methodology.It is first according between the regular hour
Dissolved oxygen sequence every collection prediction waters, method provided in an embodiment of the present invention, which is once calculated, can draw a future time
Predicted value, this future time is identical with the data collection interval of dissolved oxygen sequence, therefore between the time of gathered data
Every according to prediction goal-setting.Accordingly, time interval is shorter, and the time needed for obtaining gathered data is also fewer, and what is obtained is pre-
The corresponding future time of measured value is also just closer to present.
After dissolved oxygen sequence is obtained, the Laypunov exponent of the dissolved oxygen sequence is calculated according to dissolved oxygen sequence.Li Ya
The numerical characteristics of the average index diverging rate of general exponential representation phase space adjacent track, are the spies for recognizing some numerical value of chaotic motion
One of levy.If the Laypunov exponent of dissolved oxygen sequence is more than 0, it is that chaos time sequence can be utilized to prove the dissolved oxygen sequence
Chaotic time Forecasting Methodology is predicted to dissolved oxygen data, meanwhile, the inverse of the Laypunov exponent of gained can be used for estimation institute
State the foreseeable maximum time length of dissolved oxygen sequence.
Needs are predicted using chaotic time Forecasting Methodology phase space reconfiguration is carried out to the dissolved oxygen sequence, in reconstruct
It should be understood that the parameter of phase space reconstruction, parameter includes time delay parameter and Embedded dimensions during phase space.Therefore, trying to achieve
The Laypunov exponent for stating dissolved oxygen sequence is more than after 0, it is necessary to the parameter of phase space reconstruction:Time delay and Embedded dimensions,
Calculated.Then phase space reconfiguration is carried out to dissolved oxygen sequence according to the time delay and Embedded dimensions calculated.
When carrying out phase space reconfiguration, first have to the dissolved oxygen sequence carrying out zero-mean processing, then recycle meter
The time delay and Embedded dimensions drawn obtains phase space reconstruction.Specifically, dissolved oxygen sequence { x1, x2, x3..., xn-1, xn}
Sequence { x is obtained after being handled by zero-mean1', x2', x3' ..., xn-1', xn', the sequence after being handled using zero-mean, root
It is reconstructed according to the time delay and Embedded dimensions that calculate, obtains phase space reconstruction { Xi}。
Wherein, Xi=[xi', xi-τ', xi-2τ', xi-3τ' ..., xi-(m-1)τ’];
I=(m-1) τ+1, (m-1) τ+2, (m-1) τ+3 ..., m;
τ is the time delay of phase space reconstruction;
M is the Embedded dimensions of phase space reconstruction;
N is the number of dissolved oxygen in dissolved oxygen sequence;
For xi' tie up τ point of n- (m-1) in phase space for m.
Phase space reconstruction according to obtaining calculates the predictor formula of dissolved oxygen, so as to obtain dissolved oxygen prediction value.Root
It is predicted that formula can obtain one-step prediction value, if desired multi-step prediction, then add obtained predicted value as information former molten
Solution oxygen sequence row obtain new dissolved oxygen sequence, repeat the above steps, can obtain more predicted values.
Aquaculture dissolved oxygen prediction method provided in an embodiment of the present invention, by gathering the dissolved oxygen sequence in waters,
The Laypunov exponent of dissolved oxygen sequence is calculated, is calculated if the Laypunov exponent of dissolved oxygen sequence is more than 0 using presetting method
Time delay and Embedded dimensions and to dissolved oxygen sequence phase space reconstruction, predictor formula is calculated according to phase space reconstruction, from
And obtain the predicted value of dissolved oxygen.The predicted value to dissolved oxygen data is realized using less data sample, can be to small sample
Dissolved oxygen data carry out short-term accurate prediction.
On the basis of above-described embodiment, further, the side of the Laypunov exponent for calculating the dissolved oxygen sequence
Method includes:
It is x to calculate two adjoint points in the dissolved oxygen sequence0And x0+ ε is after n times iteration, fN(xε) and fN(x0+ ε) distance
xn=f (xn-1);
ε → 0, N → ∞ is taken, x is obtained0And x0Laypunov exponent λ (x between 2 points of+ε0):
The Laypunov exponent of each consecutive points in the dissolved oxygen sequence is added up, the Li Yapu of the dissolved oxygen sequence is obtained
Index λ (x):
Wherein, n is the number of dissolved oxygen data in dissolved oxygen sequence;
xn=f (xn-1) it is adjacent 2 points of mapping relations in dissolved oxygen sequence;
x0For any point in dissolubility sequence;
x0+ ε is x0Neighbor point;
N is iterations;
fN(x0) it is x0Value after n times iteration;
fN(x0+ ε) it is x0Values of+the ε after n times iteration;
For fN(x0) and fN(x0+ ε) distance;
λ(x0) it is x in dissolved oxygen sequence0And x0Laypunov exponent λ (x between 2 points of+ε0);
λ (x) is the Laypunov exponent of dissolved oxygen sequence, and the unit of Laypunov exponent is:Nanotesla/an iteration.
After dissolved oxygen sequence is obtained, the Laypunov exponent of the dissolved oxygen sequence is calculated according to dissolved oxygen sequence.Li Ya
The numerical characteristics of the average index diverging rate of general exponential representation phase space adjacent track, are the spies for recognizing some numerical value of chaotic motion
One of levy.If the Laypunov exponent of dissolved oxygen sequence is more than 0, it is that chaos time sequence can be utilized to prove the dissolved oxygen sequence
Chaotic time Forecasting Methodology is predicted to dissolved oxygen data, meanwhile, the inverse of the Laypunov exponent of gained can be used for estimation institute
State the foreseeable maximum time length of dissolved oxygen sequence.
Method provided in an embodiment of the present invention is there is provided a kind of computational methods of Laypunov exponent, by calculating Li Yapu
Index can be judged the chaotic characteristic of dissolved oxygen sequence, and then is realized pre- using chaos time sequence to dissolved oxygen sequence
Survey method carries out dissolved oxygen data small sample and accurately predicted.
On the basis of the various embodiments described above, further, the method tool of the utilization presetting method computing relay time
Body is:Utilize the auto-relativity function method computing relay time.
Needs are predicted using chaotic time Forecasting Methodology phase space reconfiguration is carried out to the dissolved oxygen sequence, in reconstruct
It should be understood that the parameter of phase space reconstruction, parameter includes time delay parameter and Embedded dimensions during phase space.Therefore, trying to achieve
The Laypunov exponent for stating dissolved oxygen sequence is more than after 0, it is necessary to enter to the parameter time delay of phase space reconstruction and Embedded dimensions
Row is calculated.The method of computing relay time can use auto-relativity function method.
The basic thought of auto-relativity function method is the linear dependence for extracting dissolved oxygen sequence.
For dissolved oxygen sequence { x1, x2, x3..., xn-1, xn, auto-correlation function expression formula is:
Wherein,
Wherein, C (τ) is the auto-correlation function of dissolved oxygen sequence;
xiFor the value of i-th of dissolved oxygen in dissolved oxygen sequence;
N is the number of dissolved oxygen data in dissolved oxygen sequence;
τ is time movement value;
μ is the average of dissolved oxygen sequence;
σ is the standard deviation of dissolved oxygen sequence.
Auto-correlation function is illustrated down at the time of be designated as i and i+ τ, and thus the interrelated or similarity degree of motion may be used
Functional image of the auto-correlation function on time τ (taking τ=1,2,3 ...) is made, when auto-correlation function drops to preset value
When, resulting time τ is exactly the optimum delay time of phase space reconstruction, i.e., required time delay.
Method provided in an embodiment of the present invention, the delay time parameter of phase space reconstruction is calculated using auto-correlation function, can
Accurately to try to achieve time delay, so as to provide accurate parameter there is provided the precision of phase space reconstruction for phase space reconstruction, obtain
More accurate predictor formula, and then improve the precision of prediction of dissolved oxygen.
On the basis of the various embodiments described above, further, the utilization presetting method calculates the method tool of Embedded dimensions
Body is:Estimate Embedded dimensions using G-P algorithms.
G-P algorithm key steps are as follows:
(1) dissolved oxygen sequence { x is utilized1, x2, x3..., xn-1, xn, according to the delay time T that has determined and default
Initial Embedded dimensions m0, phase space reconstruction can obtain a corresponding phase space Y;
(2) phase space Y correlation function is calculated:
Wherein, θ (x) is Heaviside functions;yiAnd yjIt is two phase points in phase space Y;|yi-yj| it is phase point yiWith
yjBetween Euclidean distance;C (r) is an aggregation function, represents that distance between two points are less than the general of r on attractor in phase space
Rate.
For r some proper range, ATTRACTOR DIMENSIONS DmLog-linear relation should be met with cumulative distribution function C (r),
I.e.
(3) m can be corresponded to obtain by fitting0Correlation dimension estimate
(4) Embedded dimensions m is gradually increased1, m1> m0, repeat the above steps (2) and (3), until corresponding dimension estimate
D (m) no longer with m growth, but it is constant in certain error range untill, now obtain Embedded dimensions m.
Method provided in an embodiment of the present invention, the delay time parameter of phase space reconstruction is calculated using auto-correlation function, can
Accurately to try to achieve Embedded dimensions, so as to provide accurate parameter there is provided the precision of phase space reconstruction for phase space reconstruction, obtain
More accurate predictor formula, and then improve the precision of prediction of dissolved oxygen.
It is further, described that prediction public affairs are calculated according to the phase space reconstruction on the basis of the various embodiments described above
The method of formula is specially:Predictor formula is determined using adding-weight one-rank local-region method.
Specifically, in phase space reconstruction { XiIn calculate each point to central point XnDistance, find out XnReference vector collection be
Xni(i=1,2,3 ..., q), calculate point XniTo XnDistance be diIf, dmIt is diIn minimum value, then defining point XniWeights
For:
It is fitted using adding-weight one-rank local-region method by reference vector collection XniIt is fitted with its weighing vector collection:
Application weighting least square method has:
Wherein, piFor point XniWeights;
Q is reference vector collection XniNumber;
diFor point XniTo XnDistance;
dmFor diIn minimum value;
WillRegard as on unknown number a, local derviation is sought simultaneously in b binary function, both sides
Abbreviation is obtained on unknown number a, and b equation group is:
Solving equations obtain a, b, then in the fitting formula above substitution, you can obtain predictor formula.
Now we obtain the one-step prediction value of dissolved oxygen sequence, need to such as carry out multi-step prediction, can regard predicted value as letter
It is that multi-step prediction can be achieved that breath, which adds former dissolved oxygen sequence and repeats above step,.
Method provided in an embodiment of the present invention, is handled phase space reconstruction using adding-weight one-rank local-region method, obtains molten
The predictor formula of oxygen sequence row is solved, dissolved oxygen prediction value then can be obtained by according to predictor formula, can also be pre- according to what is obtained
Measured value is calculated again obtains more predicted values.The predicted value to dissolved oxygen data is realized using less data sample, can
Small sample dissolved oxygen data are carried out with short-term accurate prediction.
Fig. 2 is aquaculture dissolved oxygen prediction apparatus structure schematic diagram provided in an embodiment of the present invention, as shown in Fig. 2 dress
Put including:Acquisition module 1, the first computing module 2, the second computing module 3, the computing module 5 of reconstructed module 4 and the 3rd, wherein, adopt
Collect the dissolved oxygen sequence that module 1 is used to gather prediction waters;First computing module 2 is used for the Li Ya for calculating the dissolved oxygen sequence
General index;Second computing module 3 is used to, when the Laypunov exponent is more than 0, utilize presetting method computing relay time and embedding
Enter dimension;Reconstructed module 4 is used to carry out the dissolved oxygen sequence zero-mean processing, and according to the time delay and embedded dimension
Number is reconstructed, and obtains the phase space reconstruction of the dissolved oxygen sequence;3rd computing module 5 is used for according to the phase space reconstruction
Predictor formula is calculated, and dissolved oxygen prediction value is obtained using obtained predictor formula.
Acquisition module 1 gathers the dissolved oxygen sequence for predicting waters, the embodiment of the present invention according to certain time interval first
The device of offer, which is once calculated, can draw the predicted value of a future time, this future time and the data of dissolved oxygen sequence
Acquisition time interval is identical, thus the gathered data of acquisition module 1 time interval according to prediction goal-setting.Accordingly, the time
Interval is shorter, and the time needed for obtaining gathered data is also fewer, and the obtained corresponding future time of predicted value is also just closer to existing
.
After dissolved oxygen sequence is obtained, the first computing module 2 calculates Lee of the dissolved oxygen sequence according to dissolved oxygen sequence
Refined general index.Laypunov exponent represents the numerical characteristics of the average index diverging rate of phase space adjacent track, is identification chaos fortune
Move one of feature of some numerical value.If the Laypunov exponent of dissolved oxygen sequence is more than 0, when to prove the dissolved oxygen sequence be chaos
Between sequence dissolved oxygen data can be predicted using chaotic time Forecasting Methodology, meanwhile, the Laypunov exponent of gained fall
Number can be used for estimating the foreseeable maximum time length of the dissolved oxygen sequence.
Needs are predicted using chaotic time Forecasting Methodology phase space reconfiguration is carried out to the dissolved oxygen sequence, in reconstruct
It should be understood that the parameter of phase space reconstruction, parameter includes time delay parameter and Embedded dimensions during phase space.Therefore, trying to achieve
The Laypunov exponent for stating dissolved oxygen sequence is more than after 0, it is necessary to which the second computing module 3 is to the parameter of phase space reconstruction:During delay
Between and Embedded dimensions, calculated.Then phase is carried out to dissolved oxygen sequence according to the time delay and Embedded dimensions calculated
Space Reconstruction.
Reconstructed module 4 first has to the dissolved oxygen sequence carrying out zero-mean processing, then when carrying out phase space reconfiguration
The time delay calculated and Embedded dimensions are recycled to obtain phase space reconstruction.Specifically, dissolved oxygen sequence { x1, x2,
x3..., xn-1, xnSequence { x is obtained after zero-mean processing1', x2', x3' ..., xn-1', xn', handled using zero-mean
Sequence afterwards, is reconstructed according to the time delay and Embedded dimensions that calculate, obtains phase space reconstruction { Xi}。
Wherein, Xi=[xi', xi-τ', xi-2τ', xi-3τ' ..., xi-(m-1)τ’];
I=(m-1) τ+1, (m-1) τ+2, (m-1) τ+3 ..., m;
τ is the time delay of phase space reconstruction;
M is the Embedded dimensions of phase space reconstruction;
N is the number of dissolved oxygen in dissolved oxygen sequence;
For xi' tie up τ point of n- (m-1) in phase space for m.
3rd computing module 5 calculates the predictor formula of dissolved oxygen according to obtained phase space reconstruction, so as to obtain molten
Solve oxygen predicted value.One-step prediction value can be obtained according to predictor formula, if desired multi-step prediction, then using obtained predicted value as
Information adds former dissolved oxygen sequence and obtains new dissolved oxygen sequence, repeats the above steps, can obtain more predicted values.
Aquaculture dissolved oxygen prediction device provided in an embodiment of the present invention, by gathering the dissolved oxygen sequence in waters,
The Laypunov exponent of dissolved oxygen sequence is calculated, is calculated if the Laypunov exponent of dissolved oxygen sequence is more than 0 using presetting method
Time delay and Embedded dimensions and to dissolved oxygen sequence phase space reconstruction, predictor formula is calculated according to phase space reconstruction, from
And obtain the predicted value of dissolved oxygen.The predicted value to dissolved oxygen data is realized using less data sample, can be to small sample
Dissolved oxygen data carry out short-term accurate prediction.
On the basis of the various embodiments described above, further, first computing module, specifically for:
It is x to calculate two adjoint points in the dissolved oxygen sequence0And x0+ ε is after n times iteration, fN(x0) and fN(x0+ ε) distance
xn=f (xn-1);
ε → 0, N → ∞ is taken, x is obtained0And x0Laypunov exponent λ (x between 2 points of+ε0):
The Laypunov exponent of each consecutive points in the dissolved oxygen sequence is added up, the Li Yapu of the dissolved oxygen sequence is obtained
Index λ (x):
Wherein, n is the number of dissolved oxygen data in dissolved oxygen sequence;
xn=f (xn-1) it is adjacent 2 points of mapping relations in dissolved oxygen sequence;
x0For any point in dissolubility sequence;
x0+ ε is x0Neighbor point;
N is iterations;
fN(x0) it is x0Value after n times iteration;
fN(x0+ ε) it is x0Values of+the ε after n times iteration;
For fN(x0) and fN(x0+ ε) distance;
λ(x0) it is x in dissolved oxygen sequence0And x0Laypunov exponent λ (x between 2 points of+ε0);
λ (x) is the Laypunov exponent of dissolved oxygen sequence.
After dissolved oxygen sequence is obtained, the first computing module calculates the Li Ya of the dissolved oxygen sequence according to dissolved oxygen sequence
General index.Laypunov exponent represents the numerical characteristics of the average index diverging rate of phase space adjacent track, is identification chaotic motion
One of feature of some numerical value.If the Laypunov exponent of dissolved oxygen sequence is more than 0, prove that the dissolved oxygen sequence is chaotic time
Sequence can be predicted using chaotic time Forecasting Methodology to dissolved oxygen data, meanwhile, the inverse of the Laypunov exponent of gained
Available for estimating the foreseeable maximum time length of the dissolved oxygen sequence.
Device provided in an embodiment of the present invention, the first computing module 2 can be to dissolving oxygen sequence by calculating Laypunov exponent
The chaotic characteristic of row is judged, and then realization utilizes Study on prediction technology of chaotic series to carry out dissolved oxygen number dissolved oxygen sequence
Accurately predicted according to small sample.
On the basis of the various embodiments described above, further, second computing module specifically utilizes auto-relativity function method
The computing relay time.
Second computing module is using described in the method and above method embodiment of auto-relativity function method computing relay time
Method is identical, and here is omitted.
Device provided in an embodiment of the present invention, the second computing module calculates the delay of phase space reconstruction using auto-correlation function
Time parameter, can accurately be tried to achieve time delay, so as to provide accurate parameter for phase space reconstruction, there is provided phase space reconstruction
Precision, obtain more accurate predictor formula, and then improve the precision of prediction of dissolved oxygen.
On the basis of the various embodiments described above, further, second computing module is specifically embedding using the estimation of G-P algorithms
Enter dimension.
Second computing module estimates the method and the method phase described in above method embodiment of Embedded dimensions using G-P algorithms
Together, here is omitted.
Device provided in an embodiment of the present invention, the second computing module is embedding using G-P algorithms estimation phase space reconstruction
Enter dimension, can accurately try to achieve Embedded dimensions, there is provided phase space reconstruction so as to provide accurate parameter for phase space reconstruction
Precision, obtains more accurate predictor formula, and then improve the precision of prediction of dissolved oxygen.
On the basis of the various embodiments described above, further, the 3rd computing module specifically utilizes weighing first order local area
Method determines predictor formula, and obtains dissolved oxygen prediction value using obtained predictor formula.
3rd computing module determines predictor formula using adding-weight one-rank local-region method and obtains molten using obtained predictor formula
The method for solving oxygen predicted value is identical with above method embodiment, and here is omitted.
Device provided in an embodiment of the present invention, the 3rd computing module is carried out using adding-weight one-rank local-region method to phase space reconstruction
Processing, obtains the predictor formula of dissolved oxygen sequence, then can be obtained by dissolved oxygen prediction value according to predictor formula, can be with root
Calculated again according to obtained predicted value and obtain more predicted values.Realized using less data sample to the pre- of dissolved oxygen data
Small sample dissolved oxygen data can be carried out short-term accurate prediction by measured value.
Device embodiment described above is only schematical, wherein the unit illustrated as separating component can
To be or may not be physically separate, the part shown as unit can be or may not be physics list
Member, you can with positioned at a place, or can also be distributed on multiple NEs.It can be selected according to the actual needs
In some or all of module realize the purpose of this embodiment scheme.Those of ordinary skill in the art are not paying creativeness
Work in the case of, you can to understand and implement.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can
Realized by the mode of software plus required general hardware platform, naturally it is also possible to which some functions are realized by hardware processor
Module.Understood based on such, the part that above-mentioned technical proposal substantially contributes to prior art in other words can be with soft
The form of part product is embodied, and the computer software product can be stored in a computer-readable storage medium, such as ROM/
RAM, magnetic disc, CD etc., including some instructions to cause a computer equipment (can be personal computer, server, or
Person's network equipment etc.) perform method described in some parts of each embodiment or embodiment.
Finally it should be noted that:The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations;Although
The present invention is described in detail with reference to the foregoing embodiments, it will be understood by those within the art that:It still may be used
To be modified to the technical scheme described in foregoing embodiments, or equivalent substitution is carried out to which part technical characteristic;
And these modification or replace, do not make appropriate technical solution essence depart from various embodiments of the present invention technical scheme spirit and
Scope.
Claims (10)
1. a kind of aquaculture dissolved oxygen prediction method, it is characterised in that including:
The dissolved oxygen sequence in collection prediction waters;
Calculate the Laypunov exponent of the dissolved oxygen sequence;
If the Laypunov exponent is more than 0, presetting method computing relay time and Embedded dimensions are utilized;
Zero-mean processing is carried out to the dissolved oxygen sequence, and is reconstructed according to the time delay and Embedded dimensions, is obtained
The phase space reconstruction of the dissolved oxygen sequence;
Predictor formula is calculated according to the phase space reconstruction, and dissolved oxygen prediction value is obtained using obtained predictor formula.
2. according to the method described in claim 1, it is characterised in that the Laypunov exponent for calculating the dissolved oxygen sequence
Method includes:
It is x to calculate two adjoint points in the dissolved oxygen sequence0And x0+ ε is after n times iteration, fN(x0) and fN(x0+ ε) distance
xn=f (xn-1);
<mrow>
<msup>
<mi>&epsiv;e</mi>
<mrow>
<mi>N</mi>
<mi>&lambda;</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mn>0</mn>
</msub>
<mo>)</mo>
</mrow>
</mrow>
</msup>
<mo>=</mo>
<mo>|</mo>
<msup>
<mi>f</mi>
<mi>N</mi>
</msup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mn>0</mn>
</msub>
<mo>+</mo>
<mi>&epsiv;</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<msup>
<mi>f</mi>
<mi>N</mi>
</msup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mn>0</mn>
</msub>
<mo>)</mo>
</mrow>
<mo>|</mo>
<mo>;</mo>
</mrow>
ε → 0, N → ∞ is taken, x is obtained0And x0Laypunov exponent λ (x between 2 points of+ε0):
<mrow>
<mi>&lambda;</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mn>0</mn>
</msub>
<mo>)</mo>
</mrow>
<mo>=</mo>
<msub>
<mi>lim</mi>
<mrow>
<mi>N</mi>
<mo>&RightArrow;</mo>
<mi>&infin;</mi>
</mrow>
</msub>
<mfrac>
<mn>1</mn>
<mi>N</mi>
</mfrac>
<mi>l</mi>
<mi>n</mi>
<mo>|</mo>
<mrow>
<mfrac>
<mrow>
<msup>
<mi>df</mi>
<mi>N</mi>
</msup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mn>0</mn>
</msub>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mi>d</mi>
<mi>x</mi>
</mrow>
</mfrac>
<msub>
<mo>|</mo>
<msub>
<mi>x</mi>
<mn>0</mn>
</msub>
</msub>
</mrow>
<mo>|</mo>
<mo>;</mo>
</mrow>
The Laypunov exponent of each consecutive points in the dissolved oxygen sequence is added up, the Laypunov exponent of the dissolved oxygen sequence is obtained
λ(x):
<mrow>
<mi>&lambda;</mi>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<msub>
<mi>lim</mi>
<mrow>
<mi>N</mi>
<mo>&RightArrow;</mo>
<mi>&infin;</mi>
</mrow>
</msub>
<mfrac>
<mn>1</mn>
<mi>N</mi>
</mfrac>
<msubsup>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<mi>l</mi>
<mi>n</mi>
<mo>|</mo>
<mrow>
<mfrac>
<mrow>
<msup>
<mi>df</mi>
<mi>N</mi>
</msup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mi>d</mi>
<mi>x</mi>
</mrow>
</mfrac>
<msub>
<mo>|</mo>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
</msub>
</mrow>
<mo>|</mo>
<mo>;</mo>
</mrow>
Wherein, n is the number of dissolved oxygen data in dissolved oxygen sequence;
xn=f (xn-1) it is adjacent 2 points of mapping relations in dissolved oxygen sequence;
x0For any point in dissolubility sequence;
x0+ ε is x0Neighbor point;
N is iterations;
fN(x0) it is x0Value after n times iteration;
fN(x0+ ε) it is x0Values of+the ε after n times iteration;
For fN(x0) and fN(x0+ ε) distance;
λ(x0) it is x in dissolved oxygen sequence0And x0Laypunov exponent λ (x between 2 points of+ε0);
λ (x) is the Laypunov exponent of dissolved oxygen sequence.
3. method according to claim 1 or 2, it is characterised in that the side of the utilization presetting method computing relay time
Method is specially:Utilize the auto-relativity function method computing relay time.
4. method according to claim 3, it is characterised in that the utilization presetting method calculates the method tool of Embedded dimensions
Body is:Estimate Embedded dimensions using G-P algorithms.
5. method according to claim 3, it is characterised in that described that prediction public affairs are calculated according to the phase space reconstruction
The method of formula is specially:Predictor formula is determined using adding-weight one-rank local-region method.
6. a kind of aquaculture dissolved oxygen prediction device, it is characterised in that including:
Acquisition module, the dissolved oxygen sequence for gathering prediction waters;
First computing module, the Laypunov exponent for calculating the dissolved oxygen sequence;
Second computing module, for when the Laypunov exponent is more than 0, utilizing presetting method computing relay time and embedded dimension
Number;
Reconstructed module, for carrying out zero-mean processing to the dissolved oxygen sequence, and according to the time delay and Embedded dimensions
It is reconstructed, obtains the phase space reconstruction of the dissolved oxygen sequence;
3rd computing module, for calculating predictor formula according to the phase space reconstruction, and utilizes obtained predictor formula
Obtain dissolved oxygen prediction value.
7. device according to claim 6, it is characterised in that first computing module, specifically for:
It is x to calculate two adjoint points in the dissolved oxygen sequence0And x0+ ε is after n times iteration, fN(x0) and fN(x0+ ε) distance
xn=f (xn-1);
<mrow>
<msup>
<mi>&epsiv;e</mi>
<mrow>
<mi>N</mi>
<mi>&lambda;</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mn>0</mn>
</msub>
<mo>)</mo>
</mrow>
</mrow>
</msup>
<mo>=</mo>
<mo>|</mo>
<msup>
<mi>f</mi>
<mi>N</mi>
</msup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mn>0</mn>
</msub>
<mo>+</mo>
<mi>&epsiv;</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<msup>
<mi>f</mi>
<mi>N</mi>
</msup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mn>0</mn>
</msub>
<mo>)</mo>
</mrow>
<mo>|</mo>
<mo>;</mo>
</mrow>
ε → 0, N → ∞ is taken, x is obtained0And x0Laypunov exponent λ (x between 2 points of+ε0):
<mrow>
<mi>&lambda;</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mn>0</mn>
</msub>
<mo>)</mo>
</mrow>
<mo>=</mo>
<msub>
<mi>lim</mi>
<mrow>
<mi>N</mi>
<mo>&RightArrow;</mo>
<mi>&infin;</mi>
</mrow>
</msub>
<mfrac>
<mn>1</mn>
<mi>N</mi>
</mfrac>
<mi>l</mi>
<mi>n</mi>
<mo>|</mo>
<mrow>
<mfrac>
<mrow>
<msup>
<mi>df</mi>
<mi>N</mi>
</msup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mn>0</mn>
</msub>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mi>d</mi>
<mi>x</mi>
</mrow>
</mfrac>
<msub>
<mo>|</mo>
<msub>
<mi>x</mi>
<mn>0</mn>
</msub>
</msub>
</mrow>
<mo>|</mo>
<mo>;</mo>
</mrow>
The Laypunov exponent of each consecutive points in the dissolved oxygen sequence is added up, the Laypunov exponent of the dissolved oxygen sequence is obtained
λ(x):
<mrow>
<mi>&lambda;</mi>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<msub>
<mi>lim</mi>
<mrow>
<mi>N</mi>
<mo>&RightArrow;</mo>
<mi>&infin;</mi>
</mrow>
</msub>
<mfrac>
<mn>1</mn>
<mi>N</mi>
</mfrac>
<msubsup>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</msubsup>
<mi>l</mi>
<mi>n</mi>
<mo>|</mo>
<mrow>
<mfrac>
<mrow>
<msup>
<mi>df</mi>
<mi>N</mi>
</msup>
<mrow>
<mo>(</mo>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mi>d</mi>
<mi>x</mi>
</mrow>
</mfrac>
<msub>
<mo>|</mo>
<msub>
<mi>x</mi>
<mi>i</mi>
</msub>
</msub>
</mrow>
<mo>|</mo>
<mo>;</mo>
</mrow>
Wherein, n is the number of dissolved oxygen data in dissolved oxygen sequence;
xn=f (xn-1) it is adjacent 2 points of mapping relations in dissolved oxygen sequence;
x0For any point in dissolubility sequence;
x0+ ε is x0Neighbor point;
N is iterations;
fN(x0) it is x0Value after n times iteration;
fN(x0+ ε) it is x0Values of+the ε after n times iteration;
For fN(x0) and fN(x0+ ε) distance;
λ(x0) it is x in dissolved oxygen sequence0And x0Laypunov exponent λ (x between 2 points of+ε0);
λ (x) is the Laypunov exponent of dissolved oxygen sequence.
8. the device according to claim 6 or 7, it is characterised in that second computing module specifically utilizes auto-correlation letter
The number method computing relay time.
9. device according to claim 8, it is characterised in that second computing module specifically utilizes the estimation of G-P algorithms
Embedded dimensions.
10. device according to claim 8, it is characterised in that the 3rd computing module specifically utilizes weighing first order office
Domain method determines predictor formula, and obtains dissolved oxygen prediction value using obtained predictor formula.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710409845.XA CN107292425A (en) | 2017-06-02 | 2017-06-02 | Aquaculture dissolved oxygen prediction method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710409845.XA CN107292425A (en) | 2017-06-02 | 2017-06-02 | Aquaculture dissolved oxygen prediction method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107292425A true CN107292425A (en) | 2017-10-24 |
Family
ID=60094203
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710409845.XA Pending CN107292425A (en) | 2017-06-02 | 2017-06-02 | Aquaculture dissolved oxygen prediction method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107292425A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108614422A (en) * | 2018-05-23 | 2018-10-02 | 中国农业大学 | Dissolved oxygen optimal control method, apparatus and system in the cultivation of continental rise plant circulation water |
CN114675010A (en) * | 2022-05-31 | 2022-06-28 | 卡松科技股份有限公司 | Intelligent analysis method for oxidation resistance of lubricating oil |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103778482A (en) * | 2014-02-12 | 2014-05-07 | 中国农业大学 | Aquaculture dissolved oxygen short-term prediction method based on multi-scale analysis |
CN105488335A (en) * | 2015-11-23 | 2016-04-13 | 广东工业大学 | Lyapunov exponent based power system load prediction method and apparatus |
CN105956702A (en) * | 2016-04-29 | 2016-09-21 | 中国农业大学 | Aquaculture water quality dissolved oxygen prediction method based on data fusion |
-
2017
- 2017-06-02 CN CN201710409845.XA patent/CN107292425A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103778482A (en) * | 2014-02-12 | 2014-05-07 | 中国农业大学 | Aquaculture dissolved oxygen short-term prediction method based on multi-scale analysis |
CN105488335A (en) * | 2015-11-23 | 2016-04-13 | 广东工业大学 | Lyapunov exponent based power system load prediction method and apparatus |
CN105956702A (en) * | 2016-04-29 | 2016-09-21 | 中国农业大学 | Aquaculture water quality dissolved oxygen prediction method based on data fusion |
Non-Patent Citations (4)
Title |
---|
丁涛 等: "混沌时间序列局域预测模型及其应用", 《大连理工大学学报》 * |
朱广利 等: "基于混沌理论的水质预测方法研究", 《湖北农业科学》 * |
邹志红 等: "基于相空间重构的一阶加权局域预测法在溶解氧预测中的应用", 《数学的实践与认识》 * |
陆仁强 等: "基于相空间重构理论的天津市近海水质混沌特性研究", 《海洋环境科学》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108614422A (en) * | 2018-05-23 | 2018-10-02 | 中国农业大学 | Dissolved oxygen optimal control method, apparatus and system in the cultivation of continental rise plant circulation water |
CN108614422B (en) * | 2018-05-23 | 2020-07-31 | 中国农业大学 | Method, device and system for optimally controlling dissolved oxygen in land-based factory circulating water aquaculture |
CN114675010A (en) * | 2022-05-31 | 2022-06-28 | 卡松科技股份有限公司 | Intelligent analysis method for oxidation resistance of lubricating oil |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Yang et al. | Semisupervised multilabel deep learning based nonintrusive load monitoring in smart grids | |
Liu et al. | A hybrid WA–CPSO-LSSVR model for dissolved oxygen content prediction in crab culture | |
CN105163326B (en) | A kind of cell clustering method and system based on wireless network traffic feature | |
CN110852515B (en) | Water quality index prediction method based on mixed long-time and short-time memory neural network | |
CN108564129A (en) | A kind of track data sorting technique based on generation confrontation network | |
CN110751318A (en) | IPSO-LSTM-based ultra-short-term power load prediction method | |
CN104050242A (en) | Feature selection and classification method based on maximum information coefficient and feature selection and classification device based on maximum information coefficient | |
CN107705556A (en) | A kind of traffic flow forecasting method combined based on SVMs and BP neural network | |
CN105760952A (en) | Load prediction method based on Kalman filtering and self-adaptive fuzzy neural network | |
Samui | Application of least square support vector machine (LSSVM) for determination of evaporation losses in reservoirs | |
Li et al. | Multi-step ahead wind speed forecasting approach coupling maximal overlap discrete wavelet transform, improved grey wolf optimization algorithm and long short-term memory | |
CN107038292A (en) | A kind of many output of wind electric field correlation modeling methods based on adaptive multivariable nonparametric probability | |
CN112288193A (en) | Ocean station surface salinity prediction method based on GRU deep learning of attention mechanism | |
CN109214503A (en) | Project of transmitting and converting electricity cost forecasting method based on KPCA-LA-RBM | |
CN110969290A (en) | Runoff probability prediction method and system based on deep learning | |
Tasfi et al. | Deep neural networks with confidence sampling for electrical anomaly detection | |
CN113592144A (en) | Medium-and-long-term runoff probability forecasting method and system | |
CN112396152A (en) | Flood forecasting method based on CS-LSTM | |
CN107453921A (en) | Smart city system artificial intelligence evaluation method based on nonlinear neural network | |
CN107292425A (en) | Aquaculture dissolved oxygen prediction method and device | |
CN115906954A (en) | Multivariate time sequence prediction method and device based on graph neural network | |
Yang et al. | Chaotic bayesian method based on multiple criteria decision making (MCDM) for forecasting nonlinear hydrological time series | |
CN110516792A (en) | Non-stable time series forecasting method based on wavelet decomposition and shallow-layer neural network | |
CN104239689B (en) | Short-term wind speed forecasting method based on optimization Method Using Relevance Vector Machine | |
Badyalina et al. | Streamflow estimation at ungauged basin using modified group method of data handling |
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: 20171024 |