CN105956709A - GUI based modular support vector machine tide forecasting method - Google Patents

Abstract

The invention discloses a GUI based modular support vector machine tide forecasting method, which comprises the steps of S1, acquiring continuous-sequence tide level information monitored by a tide station and a tide value forecast by using a harmonic analysis method; S2, making a difference between a measured value acquired by the tide station and the tide value forecast by the harmonic analysis method, acquiring a time sequence of non-astronomical tide, carrying out data accumulation processing on the inputted tide level information and the tide value according to a grey model AGO algorithm, and enabling the processed data to act as input so as to be applied to regression forecasting of a support vector machine; S3, forecasting tide through the support vector machine according to tide forecasting time information set in the step S1; and S4, completing data recovery of a forecast result of the support vector machine through IAGO reverse accumulation processing, wherein the recovered data is used for modifying the tide forecasting value of the harmonic analysis method.

Description

A kind of modularity support vector machine tide prediction method based on GUI

Technical field

The present invention relates to tide prediction field, particularly relate to a kind of modularity support vector machine tide prediction method based on GUI.

Background technology

Owing to tide is affected by various factors, periodic factors such as power to lead tide, factor aperiodic such as wind-force, air pressure, seashore characteristic, precipitation, the inclination angle etc. of lunar orbit.Traditional system of harmonic analysis is by the statistics and analysis to long-term tidal data, it is calculated the parameter of each partial tide in model, and obtain long-term tide prediction based on the mathematical model setting up tide, but the forecast precision of the method is except being affected by partial tide quantity, also cannot analyze the impact of factor aperiodic.The most conventional neural network prediction method is to affect each key element of tide, if the information such as position of heavenly body parameter, wind, air pressure, precipitation are as the input of network, neural Network Model for Forecasting is set up as network output using Tidal Information, and by the study of historical data being determined the parameter of network.Neural network model and input according to establishing carry out tide prediction, although the method compensate for harmonic analysis to a certain extent cannot forecast the shortcoming of factor aperiodic, but the sample of learning training requires that data volume is big, involve a wide range of knowledge, various situations about being likely to occur can be covered, and the station historical data with factor aperiodic is the most little.

Summary of the invention

The problem existed according to prior art, the invention discloses a kind of modularity support vector machine tide prediction method based on GUI, comprises the following steps:

S1: obtain the continuous sequence formula tidal level information that tidal gaging station monitors and the tide value utilizing system of harmonic analysis to predict, the time value of tide prediction is set；

S2: the tide measured value obtained at tidal level station does difference with the tide value of system of harmonic analysis prediction, the time series of the non-astronomical tide obtained carries out data accumulation process according to gray model AGO algorithm to tidal level information and the tide value of input, and the data after process are used for the regression forecasting of support vector machine as input quantity；

S3: tide is predicted by the tide prediction temporal information according to arranging in S1 by support vector machine, select cross validation, population and genetic algorithm to be optimized respectively the penalty coefficient c and kernel function radius g of support vector machine during prediction, and the method selecting support vector machine to produce error minima during training carries out tide prediction as optimized algorithm；

The result of S4: SVM prediction completes data convert through the reverse accumulation process of IAGO, and the data after reduction are for revising the tide prediction value of system of harmonic analysis.

The tide measured value that tidal gaging station obtains is defined as the time series of non-astronomical tide, is expressed as surveying tide sequence y₀:x₀(1),x₀(2),…,x₀(n)；

The tidal data of system of harmonic analysis prediction, it is when representing actual tide level H (t) in somewhere, and computational methods are as follows:

$H\left(t\right)={\mathrm{\α}}_0+\underset{j=1}{\overset{m}{\Σ}}{R}_{j}cos({\mathrm{\σ}}_{j}t-{\mathrm{\θ}}_j)={\mathrm{\α}}_0+\underset{j=1}{\overset{m}{\Σ}}({\mathrm{\α}}_j{\mathrm{cos\σ}}_{j}t+b_j{\mathrm{sin\σ}}_{j}t)$

α in formula₀For mean sea level height, R_jFor amplitude of component tide, θ_jFor the initial phase of partial tide, σ_jFor the angular velocity of partial tide, α_j=R_jcosθ_j, b_j=R_jsinθ_j, m is the number of partial tide, is positive integer；The forecast part of system of harmonic analysis is regarded as astronomical tide part, and the harmonic analysis time series obtained after calculating is designated as y₁:x₁(1),x₁(2),…,x₁(n)；Will actual measurement tide sequence y₀When doing poor with harmonic analysis time series:

Computing formula is:

y₂=y₀-y₁

New sequence is:

y₂:x₃(1),x₃(2),…,x₃(n)

This sequence is firstly the need of the regression forecasting being used for support vector machine after AGO operates:

The sequence table processed through AGO is shown as x⁽¹⁾(1),x⁽¹⁾(2),…,x⁽¹⁾(n), computing formula is:

$x^{\left(1\right)}\left(k\right)={\mathrm{\Σ}}_{i=1}^k{x}_3\left(k\right),k=1,2,...,n$

At the forecasting sequence obtained after SVM prediction it is:

${\hat{x}}^{\left(1\right)}\left(1\right),{\hat{x}}^{\left(1\right)}\left(2\right),...,{\hat{x}}^{\left(1\right)}\left(n\right)$

This sequence is through the algorithm non-final forecasting sequence of astronomical tide part of acquisition of IAGO:

${\hat{x}}^{\left(0\right)}(k+1)={\hat{x}}^{\left(1\right)}(k+1)-{\hat{x}}^{\left(1\right)}\left(k\right),k=1,2,...,n$

WhereinRepresenting the data of reduction after non-astronomical tide fractional prediction, final tidal data is added with astronomical tide part by the predictive value of non-astronomical tide part and is finally predicted the outcome；

The time obtained is that the tide value of k+1 is expressed as

$y(k+1)={\hat{x}}^{\left(0\right)}(k+1)+x_1(k+1).$

Owing to have employed technique scheme, a kind of based on GUI modularity support vector machine tide prediction method that the present invention provides, the factor being possible not only to affect aperiodic such as wind direction, rainfall, Storm events, seashore characteristic etc. tide is fused in this Forecasting Methodology, and Small Sample Database can also reach more accurate result.Set up one based on support vector machines forecast model, first in MATLAB 7.8, import a SVM workbox, then utilize svmtrain function that training sample data are trained, testing, with test sample svmpredict function, the model formed, the tide of same tidal station could be predicted by the data after training and test again.

Accompanying drawing explanation

In order to be illustrated more clearly that the embodiment of the present application or technical scheme of the prior art, the accompanying drawing used required in embodiment or description of the prior art will be briefly described below, apparently, accompanying drawing in describing below is only some embodiments described in the application, for those of ordinary skill in the art, on the premise of not paying creative work, it is also possible to obtain other accompanying drawing according to these accompanying drawings.

Fig. 1 is the flow chart of this method；

Fig. 2 is the flow chart of tide prediction process.

Detailed description of the invention

For making technical scheme and advantage clearer, below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out the most complete description:

A kind of based on GUI modularity support vector machine tide prediction method as shown in Figure 1, specifically includes following steps:

S1: obtain the continuous sequence formula tidal level information that tidal gaging station monitors and the tide value utilizing system of harmonic analysis to predict, arrange the time value of tide prediction, want the tide value after predicting 10 days if this time is user, then this time is 10.

S2: the tide measured value obtained at tidal level station does difference with the tide value of system of harmonic analysis prediction, the non-astronomical tide time series obtained carries out data accumulation process according to gray model AGO algorithm to tidal level information and the tide value of input, and the data after process are used for the regression forecasting of support vector machine as input quantity.

Modularity forecast journey is in running, it is necessary first to obtaining two inputs, one is the tidal gaging station measured data that user is loaded into, and these Data Sources are in the tidal gaging station record to tidal level information every day, and in expression is a time series in form, and this sequence is designated as y₀:x₀(1),x₀(2),…,x₀(n).Its two be system of harmonic analysis prediction tidal data, it represent actual tide level H (t) in somewhere time, computational methods are as follows:

$H\left(t\right)={\mathrm{\α}}_0+\stackrel{m}{\Σ}{R}_{j}cos({\mathrm{\σ}}_{j}t-{\mathrm{\θ}}_j)={\mathrm{\α}}_0+\stackrel{m}{\Σ}({\mathrm{\α}}_j{\mathrm{cos\σ}}_{j}t+b_j{\mathrm{sin\σ}}_{j}t)$

α in formula₀For mean sea level height, R_jFor amplitude of component tide, θ_jFor the initial phase of partial tide, σ_jFor the angular velocity of partial tide, α_j=R_jcosθ_j, b_j=R_jsinθ_j, m is the number of partial tide, is positive integer.Owing to system of harmonic analysis primary concern is that, when predicting tidal level, the effect that tide is formed by Between Celestial Tide-generating Forces, so the forecast part of system of harmonic analysis is regarded as astronomical tide part, the harmonic analysis time series obtained after calculating is designated as y₁:x₁(1),x₁(2),…,x₁(n)。

The most two-part difference, then regarded as the non-astronomical tide part affected by other factors such as environment by us, and computational methods are: y₀-y₁, this sequence table is shown as x₃(1),x₃(2),…,x₃Sequence n () processes after is expressed as " non-astronomical tide tide sequence " in fig. 2, list entries is processed to improve precision of prediction by this sequence as the prediction being originally inputted for the non-i.e. support vector machine of astronomical tide part, the AGO algorithm first passing through gray model before forecasting.AGO is as follows to seasonal effect in time series processing method:

The original series divided due to non-astronomical tide is x₃(1),x₃(2),…,x₃N (), then the sequence table processed through AGO is shown as x⁽¹⁾(1),x⁽¹⁾(2),…,x⁽¹⁾(n), whereinAfter the treatment, the tide time series of new non-astronomical tide part is applied in the forecast of support vector machine.

S3: tide is predicted by the tide prediction temporal information according to arranging in S1 by support vector machine, select cross validation, population and genetic algorithm to be optimized respectively the penalty coefficient c and kernel function radius g of support vector machine during prediction, and the method selecting support vector machine to produce error minima during training carries out tide prediction as optimized algorithm.

The result of S4: SVM prediction completes data convert through the reverse accumulation process of IAGO, and the data after reduction are for revising the tide prediction value of system of harmonic analysis.

During prediction, program will automatically be selected three kinds of optimized algorithms to be optimized support vector machine and chosen the minimum algorithm of error as when time optimization method of prediction, and these three method is cross validation selection method, particle swarm optimization algorithm and genetic algorithm respectively.Wherein, program is root-mean-square error RMSE that training process produces to the judgment criteria of error, and computational methods are:

$RMSE=\sqrt{\frac{{\mathrm{\Σ}}_{i=1}^L{(y-\hat{y})}^2}{L}}$

Wherein L is training sample number, y andRepresent the training data of tide and the predicted value of this part respectively.

The predictive value of tide non-astronomical tide part is just obtained after support vector machine is forecast, owing to data having been carried out AGO process before forecast, so also needing to carry out the reduction of data before obtaining final predicting the outcome, i.e. IAGO operates, being shown as " acquisition predicts the outcome and carries out data convert " in fig. 2, the computational methods of IAGO are:

${\hat{x}}^{\left(0\right)}(k+1)={\hat{x}}^{\left(1\right)}(k+1)-{\hat{x}}^{\left(1\right)}\left(k\right),k=1,2,...,n$

WhereinThe data reduced after representing prediction,Represent unreduced data after prediction.

The forecast data of non-astronomical tide part is done with the data that astronomical tide part system of harmonic analysis calculates and obtained final forecast result.I.e. complete the operation of whole prediction program.

The above; it is only the present invention preferably detailed description of the invention; but protection scope of the present invention is not limited thereto; any those familiar with the art is in the technical scope that the invention discloses; according to technical scheme and inventive concept equivalent or change in addition thereof, all should contain within protection scope of the present invention.

Claims (2)

1. a modularity support vector machine tide prediction method based on GUI, it is characterised in that: include with Lower step:

S1: obtain the continuous sequence formula tidal level information that tidal gaging station monitors and the tide utilizing system of harmonic analysis to predict Nighttide value, arranges the time value of tide prediction；

S2: the tide measured value obtained at tidal level station does difference with the tide value of system of harmonic analysis prediction, obtains The time series of non-astronomical tide according to gray model AGO algorithm to input tidal level information and tide value number According to accumulation process, the data after process are used for the regression forecasting of support vector machine as input quantity；

S3: tide is predicted by the tide prediction temporal information according to arranging in S1 by support vector machine, During prediction, the penalty coefficient c and kernel function radius g of support vector machine are selected respectively cross validation, Population and genetic algorithm are optimized, and select support vector machine to produce error minima during training Method carries out tide prediction as optimized algorithm；

The result of S4: SVM prediction completes data convert, after reduction through the reverse accumulation process of IAGO Data for revising the tide prediction value of system of harmonic analysis.

A kind of modularity support vector machine tide prediction method based on GUI the most according to claim 1, It is further characterized in that: the tide measured value that tidal gaging station obtains is defined as the time series of non-astronomical tide, represents For actual measurement tide sequence y₀:x₀(1),x₀(2),…,x₀(n)；

The tidal data of system of harmonic analysis prediction, it is when representing actual tide level H (t) in somewhere, computational methods As follows:

$H\left(t\right)={\mathrm{\α}}_0+\underset{j=1}{\overset{m}{\mathrm{\Σ}}}{R}_{j}cos({\mathrm{\σ}}_{j}t-{\mathrm{\θ}}_j)={\mathrm{\α}}_0+\underset{j=1}{\overset{m}{\mathrm{\Σ}}}({\mathrm{\α}}_j{\mathrm{cos\σ}}_{j}t+b_j{\mathrm{sin\σ}}_{j}t)$

α in formula₀For mean sea level height, R_jFor amplitude of component tide, θ_jFor the initial phase of partial tide, σ_jFor partial tide Angular velocity, α_j=R_jcosθ_j, b_j=R_jsinθ_j, m is the number of partial tide, is positive integer；System of harmonic analysis Forecast part regard astronomical tide part as, after calculating, the harmonic analysis time series that obtains is designated as y₁: x₁(1),x₁(2),…,x₁(n)；Will actual measurement tide sequence y₀When doing poor with harmonic analysis time series:

Computing formula is:

y₂=y₀-y₁

New sequence is:

y₂:x₃(1),x₃(2),…,x₃(n)

This sequence is firstly the need of the regression forecasting being used for support vector machine after AGO operates:

The sequence table processed through AGO is shown as x⁽¹⁾(1),x⁽¹⁾(2),…,x⁽¹⁾(n), computing formula is:

$x^{\left(1\right)}\left(k\right)={\mathrm{\Σ}}_{i=1}^k{x}_3\left(k\right),k=1,2,...,n$

At the forecasting sequence obtained after SVM prediction it is:

${\hat{x}}^{\left(1\right)}\left(1\right),{\hat{x}}^{\left(1\right)}\left(2\right),...,{\hat{x}}^{\left(1\right)}\left(n\right)$

This sequence is through the algorithm non-final forecasting sequence of astronomical tide part of acquisition of IAGO:

${\hat{x}}^{\left(0\right)}(k+1)={\hat{x}}^{\left(1\right)}(k+1)-{\hat{x}}^{\left(1\right)}\left(k\right),k=1,2,...,n$

WhereinThe data reduced after representing non-astronomical tide fractional prediction, final tidal data is by non-astronomical tide The predictive value of part is added with astronomical tide part and is finally predicted the outcome；

The time obtained is that the tide value of k+1 is expressed as

$y(k+1)={\hat{x}}^{\left(0\right)}(k+1)+x_1(k+1).$