CN110110339A - A kind of hydrologic forecast error calibration method and system a few days ago - Google Patents
A kind of hydrologic forecast error calibration method and system a few days ago Download PDFInfo
- Publication number
- CN110110339A CN110110339A CN201810085197.1A CN201810085197A CN110110339A CN 110110339 A CN110110339 A CN 110110339A CN 201810085197 A CN201810085197 A CN 201810085197A CN 110110339 A CN110110339 A CN 110110339A
- Authority
- CN
- China
- Prior art keywords
- hydrologic
- prediction
- forecast
- value
- error
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 132
- 230000008569 process Effects 0.000 claims abstract description 82
- 239000011159 matrix material Substances 0.000 claims abstract description 66
- 238000012937 correction Methods 0.000 claims description 72
- 238000003064 k means clustering Methods 0.000 claims description 23
- 238000005259 measurement Methods 0.000 claims description 15
- 238000010276 construction Methods 0.000 claims description 7
- 238000004458 analytical method Methods 0.000 claims description 6
- 238000009472 formulation Methods 0.000 claims description 6
- 239000000203 mixture Substances 0.000 claims description 6
- 238000005192 partition Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 4
- 230000001174 ascending effect Effects 0.000 claims description 3
- 238000000354 decomposition reaction Methods 0.000 abstract description 3
- 238000012850 discrimination method Methods 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 12
- 238000004590 computer program Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 5
- 230000001629 suppression Effects 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 230000009897 systematic effect Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000013277 forecasting method Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 238000001556 precipitation Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000011144 upstream manufacturing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G06F19/00—
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Probability & Statistics with Applications (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention relates to a kind of hydrologic forecast error calibration method and systems a few days ago, including calculate hydrologic forecast Process Forecasting value and prediction error matrix percentile section；K mean cluster is carried out to the predicted value and prediction error matrix percentile section；Establish the typical categories regression model of predicted value and prediction error matrix percentile section after K mean cluster；Establish the predicted value calibration model of typical categories regression model prediction error estimation.The present invention decomposes error dynamics process, is recognized, is positioned, then carried out characteristic matching in practical applications using meter and error time sequence mode decomposition and the error character discrimination method a few days ago of error character classification, and it is horizontal to improve hydrologic forecast a few days ago.
Description
Technical Field
The invention relates to a correction method and a correction system for a renewable energy power generation technology, in particular to a Japanese hydrologic prediction error correction method and a Japanese hydrologic prediction error correction system.
Background
Errors of hydrologic prediction have certain uncertainty, and especially in the Japanese hydrologic prediction, the adopted process-driven model method or data-driven model method cannot effectively avoid the problem of deviation uncertainty of the Japanese flow prediction. And the flood peak information obtained by the real-time hydrological survey station of the drainage basin is utilized, and the warehouse entry flow and the flood peak flow are corrected on the basis of the flood peak information, so that the method depends on the upstream occurrence time of the flood peak and the flood peak flow, and has the limitations of short forecast period, dependence on drainage basin runoff characteristics and the like. For hydrologic prediction with a long forecast period, relatively few actual measurement data can be used, and more correction schemes can be found in sequence characteristics and historical prediction rules.
For application, the extension of the forecast period brings risks in terms of error stability and information reliability, and how to improve the error suppression capability of the hydrological forecast in the future determines the practical degree of the forecast model to a certain extent. In order to solve the problem, different types of error real-time correction technologies are usually adopted to intervene in an original prediction result output by a model, wherein a common mode is analysis based on a time sequence of an error, for example, the error correction is realized by means of an autoregressive model or a BP neural network, and the like.
Secondly, in the prior art, different error characteristics are generally not regarded as each configuration of the whole error dynamic process, so that the same processing strategy with different characteristics such as systematic errors, nonlinear errors and the like is easy to occur in the aspect of error correction, the error suppression capability is reduced, and the overall performance of the hydrologic prediction model is weakened. In the case of a pre-day hydrologic forecast, there may be large estimation errors in precipitation due to the forecast period.
Disclosure of Invention
The invention aims to provide a Japanese hydrologic prediction error correction method and a Japanese hydrologic prediction error correction system, which are suitable for hydrologic prediction error correction which is longer than the conventional hydrologic prediction and keeps a higher prediction level in a longer prediction period, so as to solve the problems that different error characteristics are not generally regarded as each configuration of the whole error dynamic process, so that the same processing strategy with different characteristics such as systematic errors, nonlinear errors and the like is easy to appear in the aspect of error correction, the error suppression capability is reduced, and the overall performance of a hydrologic prediction model is weakened.
The purpose of the invention is realized by adopting the following technical scheme:
the invention provides a Japanese hydrologic prediction error correction method, which is improved in that:
acquiring hydrologic forecast data;
correcting the acquired hydrologic forecast data based on a pre-established correction model; wherein,
the pre-established correction model comprises: and clustering and generating the K-means value of the percentile interval based on the historical hydrologic prediction value and the prediction error matrix.
Further: the acquiring of the hydrologic forecast data comprises acquiring historical hydrologic forecast data and historical hydrologic actual measurement data.
Further: the pre-established correction model comprises:
calculating a prediction value and a prediction error matrix percentile interval of the hydrologic prediction process according to the historical hydrologic prediction data and the historical hydrologic actual measurement data;
performing K-means clustering on the forecasting values of the hydrologic forecasting process and the forecasting error matrix percentile interval;
establishing a typical category regression model of the prediction value after K-means clustering and the prediction error matrix percentile interval;
and establishing a correction model according to the typical category regression model.
Further: before calculating the hydrologic prediction value and prediction error matrix percentile interval according to the acquired historical hydrologic prediction data and the acquired historical hydrologic actual measurement data, constructing a prediction value and a prediction error matrix;
the constructing of the forecasting value and the forecasting error matrix in the hydrologic forecasting process comprises the following steps:
determining a forecast value and a forecast error in a hydrologic forecast process according to historical hydrologic forecast data and historical hydrologic actual measurement data;
constructing a forecast value and a forecast error matrix according to the forecast value and the forecast error in the hydrologic forecast process;
forecasting value P 'in hydrologic forecasting process according to forecasting value and forecasting error matrix of percentile method'ijAnd (5) grouping the components in different regions.
Further: the forecasting value and the forecasting error in the hydrologic forecasting process are as follows:
Eij＝Pij-P'ij；
wherein E isijTo predict errors, PijIs measured value, P'ijPredicting a value for the hydrologic prediction process, wherein i is more than or equal to 1 and less than or equal to a set threshold, the set threshold is the time resolution between 24 hours/day-by-day samples, and j is more than or equal to 1 and less than or equal to n; n is a scoreAnalyzing the forecast days.
Further: the hydrologic prediction process prediction value and prediction error matrix are expressed as:
A＝(P′ij,Eij)
wherein: x is a prediction value and prediction error matrix, Ai1、Ai2...AijBoth represent the hydrologic prediction process prediction value and the 1 st column of the prediction error matrix at row i and the 2 nd column of the prediction error matrix at row i.
Further: the step of grouping the forecast values in the hydrologic forecast process according to the percentile method forecast values and the forecast error matrix comprises the following steps:
forecasting value P 'through hydrologic forecasting process'ijTaking ascending order as a premise, and carrying out P 'according to a percentile method'ijPercentile inter-partition collection, P'ijThe percentile inter-partition collection is expressed as: x (T)t，Eij) Represents when P'ijAn error sample corresponding to the prediction value interval Tt of the hydrologic prediction process selected at random, wherein the prediction value interval of the hydrologic prediction process is P'ijThe error samples are classified under the condition of each percentile interval,t＝[1,100,1]。
further: and the K-means clustering of the prediction value and the prediction error matrix percentile interval is represented as follows:
wherein: f (P)k) K-means aggregation for representing prediction value and prediction error matrix percentile interval of hydrologic prediction processClass; pkTo make f (P)k) One division at a minimum, Pk＝{C1，…，Ck}；Denotes the center position of the nth cluster, n is 1, …, k, d (x)y,mn) Denotes xyTo mnA distance; x is the number ofyIs X (T)t，Eij) The elements of (1); m isnRepresents the center position of the nth cluster; t istAnd representing the forecast value interval of the randomly selected hydrological forecasting process.
Further: the typical class regression model is represented as:
Eij＝a·P'ij+b
wherein: a and b are preset regression coefficients, P'ijPredicting values for a hydrologic prediction process; eij represents the prediction error.
Further: correcting the acquired hydrologic forecast based on a pre-established correction model, wherein the correction is carried out according to the following formula:
P't＝f(P't)
wherein: p 'on the left'tRepresenting a specific forecast value in a forecast value interval of the hydrologic forecast process; f (P't) Representing the correction values for a typical subset of the forecasted time series tracks given by K-means clustering.
The invention also provides a Japanese hydrologic prediction error correction system, which is improved in that: the method comprises the following steps:
the acquisition module is used for acquiring hydrologic forecast data;
the correction module is used for correcting the acquired hydrologic forecast data based on a pre-established correction model; wherein,
the pre-established correction model comprises: and clustering and generating the K-means value of the percentile interval based on the historical hydrologic prediction value and the prediction error matrix.
Further: the device also comprises a formulation module which is used for formulating the correction model in advance.
Further: the formulation module comprises:
the calculation submodule is used for calculating a forecast value and a forecast error matrix percentile interval in the hydrologic forecast process according to the historical hydrologic forecast data and the historical hydrologic actual measurement data;
the clustering submodule is used for carrying out K-means clustering on the forecasting values of the hydrologic forecasting process and the forecasting error matrix percentile interval;
the first establishing submodule is used for establishing a typical category regression model of the prediction value after the K-means clustering and the prediction error matrix percentile interval;
and the second establishing submodule is used for establishing a correction model according to the typical category regression model.
And further, the method also comprises a construction submodule for constructing the hydrologic prediction process forecast value and the forecast error matrix before the hydrologic prediction process forecast value and the forecast error matrix percentile interval are calculated according to the obtained hydrologic prediction.
Further: the construction submodule, comprising:
the first determination unit is used for determining a prediction value and a prediction error in a hydrologic prediction process according to historical hydrologic prediction data and historical hydrologic actual measurement data;
the first establishing unit is used for constructing a forecast value and a forecast error matrix according to the forecast value and the forecast error;
a collecting unit for forecasting value P 'in hydrologic forecasting process according to percentile forecasting value and forecasting error matrix'ijAnd (5) grouping the components in different regions.
Further: the correction module is further used for correcting the acquired hydrologic forecast based on a pre-established correction model, and the correction comprises the following steps:
P't＝f(P't)
wherein: p 'on the left'tRepresenting a specific forecast value in a forecast value interval of the hydrologic forecast process; f (P't) Representing the correction values for a typical subset of the forecasted time series tracks given by K-means clustering.
Compared with the closest prior art, the technical scheme provided by the invention has the beneficial effects that:
the invention provides a Japanese hydrologic forecast error correction method, which comprises the steps of obtaining hydrologic forecast data; correcting the acquired hydrologic forecast data based on a pre-established correction model; the pre-established correction model comprises: based on the historical hydrologic prediction value and the K-means clustering of the percentile interval of the prediction error matrix, the problems that different error characteristics are not generally regarded as each configuration of the whole error dynamic process in the prior art, so that the same processing strategy with different characteristics such as systematic errors, nonlinear errors and the like is easy to appear in the aspect of error correction, the error suppression capability is reduced, and the whole performance of a hydrologic prediction model is weakened are solved.
On the basis of a traditional hydrological forecasting method, diagnosis research is carried out on time sequence tracks of error features of the hydrological forecasting in the day ahead, the method is different from a black box type error statistical correction, a machine learning algorithm for precision optimization and the like, a day ahead error feature identification method for considering error time series modal decomposition and error feature classification is adopted, and the purpose of reducing the error of the hydrological forecasting in the day ahead is achieved by decomposing, identifying and positioning an error dynamic process and then carrying out feature matching in practical application.
Drawings
FIG. 1 is a simple block diagram of a method for correcting a Japanese hydrological prediction error based on time series characteristic clustering according to the present invention;
FIG. 2 is a detailed block diagram of the flow of the method for correcting the error of the Japanese hydrological prediction based on the time series characteristic clustering provided by the invention;
fig. 3 is a structural block diagram of a system for correcting the error of the Japanese hydrologic prediction based on the time series feature clustering provided by the invention.
Detailed Description
The following describes embodiments of the present invention in further detail with reference to the accompanying drawings.
The following description and the drawings sufficiently illustrate specific embodiments of the invention to enable those skilled in the art to practice them. Other embodiments may incorporate structural, logical, electrical, process, and other changes. The examples merely typify possible variations. Individual components and functions are optional unless explicitly required, and the sequence of operations may vary. Portions and features of some embodiments may be included in or substituted for those of others. The scope of embodiments of the invention encompasses the full ambit of the claims, as well as all available equivalents of the claims. Embodiments of the invention may be referred to herein, individually or collectively, by the term "invention" merely for convenience and without intending to voluntarily limit the scope of this application to any single invention or inventive concept if more than one is in fact disclosed.
The first embodiment,
The invention provides a method for identifying the day-ahead error characteristics by taking the analysis of the error time sequence track characteristics as the premise and considering the modal decomposition of the error time sequence and the classification of the error characteristics, thereby realizing the error correction of the day-ahead hydrological prediction. The invention provides a flow chart of a Japanese hydrologic forecast error correction method based on time sequence feature clustering, as shown in figures 1 and 2, the method comprises the following steps:
step 11, acquiring hydrologic forecast data;
step 12, correcting the acquired hydrologic forecast data based on a pre-established correction model; wherein,
the pre-established correction model comprises: and clustering and generating the K-means value of the percentile interval based on the historical hydrologic prediction value and the prediction error matrix.
The acquiring of the hydrologic forecast data in step 11 includes acquiring historical hydrologic forecast data and historical hydrologic actual measurement data.
The step 12 of correcting the acquired hydrologic forecast data by the pre-established correction model includes:
1. and (4) collecting hydrologic prediction time sequence process data, constructing a prediction value and a prediction error matrix, and calculating a percentile interval of the prediction value.
2. And clustering the prediction value with the prediction error matrix K mean value.
3. And establishing a typical category regression model of the forecast value and the forecast error matrix.
4. A prediction value correction model based on the prediction error estimate.
The specific contents are as follows:
to accurately illustrate the method and steps of practicing the present invention, the following examples are given:
(1) hydrologic prediction process data collection
Usually, the forecast period of hydrologic forecast is the accumulation of convergence time given by statistical data, and can reach the day ahead and above under the condition of forecasting rainfall by application.
In the embodiment, a hydrologic prediction process and corresponding actually-measured warehousing flow rate form a prediction time sequence, and a prediction value is defined to be P'ijAnd measured value is PijThe prediction error is Eij，Eij＝Pij-P'ij
Wherein i is more than or equal to 1 and less than or equal to 96, j is more than or equal to 1 and less than or equal to n
N is the forecast day number of the input analysis, i.e. the day number of the input analysis samples is N, the time resolution between the day-by-day samples is 15min, and the total number of the samples N is N96
Wherein: eijTo predict errors, PijIs measured value, P'ijPredicting a value for the hydrologic prediction process, wherein i is more than or equal to 1 and less than or equal to a set threshold, the set threshold is the time resolution between 24 hours/day-by-day samples, and j is more than or equal to 1 and less than or equal to n; n is the forecast day number of the analysis, and the total number of samples N is N.
First, a total of N historical samples are data-aggregated.
The forecast value and the forecast error form a matrix which is defined as X (P'ij，Eij)
From P'ijTaking ascending order as a premise, and carrying out P 'according to a percentile method'ijInter-division collection
Thus, X (P'ij，Eij) Further rewritten as X (T)t，Eij)，t＝[1,100,1]
(2)X(Tt，Eij) K-means clustering of
By randomly selected intervals TtAnd taking the K samples as initial center points, classifying the rest samples into the cluster where the center point with the highest similarity is located, then determining the mean value of the sample coordinates in the current cluster as a new center point, and sequentially and circularly iterating until the categories of all the samples do not change.
Thus, the original data set X (T)t，Eij) By using the K-means clustering method, X (T) can be obtainedt，Eij) Treated as a clustered data set and one potentially each cluster satisfying the integer K. Outputting one of the partitions Pk＝{C1，…，CkIs xyIs X (T)t，Eij) Then the problem of K-means clustering as described in this example is to find such a partition Pk＝{C1，…，CkSuch that the objective functionAnd minimum.
Wherein,denotes the center position of the nth cluster, n is 1, …, k, d (x)y，mn) Denotes xyTo mnDistance.
(3)X(P'ij，Eij) Regression modeling of canonical classes
In the interval TtIn, X (T)t，Eij) The method is divided into clusters with similar object characteristics by the application of a K-means clustering algorithm, and objects are different among different clusters.
Obtaining a certain typical characteristic subset of an original prediction value matrix and a prediction error value matrix with similar characteristics through a clustering algorithm, and establishing a regression equation aiming at the subset of the typical prediction time sequence track:
Eij＝a·P'ij+ b type (1.1)
Solving the linear relation between the two by using a least square method to give P'ijSatisfies the interval TtAnd the values of a and b under the condition of the typical feature subset, namely P is givenijAnd EijSatisfies the interval TtAnd a subset of typical features, existA given, historical statistics based, typical relationship.
(4) Based on EijEstimated P'ijAnd (3) calculating a correction value:
using the formula (1.1), interval TtAnd in the characteristic feature subset, corresponding EijFrom the measured value in the history time sequence track as PijAnd predicted value is P'ijThe difference is equivalent to an estimate in a linear relationship.
Therefore, in the interval TtAnd in the characteristic feature subset, PijCan be regarded as P'ijI.e. corresponding to an interval, a characteristic feature subset, there is a certain Pij＝f(P'ij)。
At a future moment, the hydrologic forecast model gives a specific forecast value P't
Is P'tSubject to TtIn a specific interval of (1), and P'tIn single day-ahead prediction, the belonged interval and the exhibited error statistics accord with a typical prediction time sequence track subset given by K-means clustering, and then
P't＝f(P't) Formula (1.2)
Equation 1.2 is the error correction model given in this example.
Example II,
Based on the same inventive concept, the invention also provides a time series characteristic clustering-based Japanese hydrologic forecast error correction system, the structure diagram of which is shown in fig. 3, and the system comprises:
an obtaining module 201, configured to obtain hydrologic forecast data;
the correction module 202 is configured to correct the acquired hydrologic forecast data based on a pre-established correction model; wherein,
the pre-established correction model comprises: and (4) clustering and generating the percentile interval K mean value based on the forecast value and the forecast error matrix in the historical hydrologic forecast process.
Further: the device also comprises a formulation module which is used for formulating the correction model in advance.
Further: the formulation module comprises:
the calculation submodule is used for calculating a forecast value and a forecast error matrix percentile interval in the hydrologic forecast process according to the historical hydrologic forecast data and the historical hydrologic actual measurement data;
the clustering submodule is used for carrying out K-means clustering on the forecasting values of the hydrologic forecasting process and the forecasting error matrix percentile interval;
the first establishing submodule is used for establishing a typical category regression model of the prediction value after the K-means clustering and the prediction error matrix percentile interval;
and the second establishing submodule is used for establishing a correction model according to the typical category regression model.
And further, the method also comprises a construction submodule for constructing the forecasting value and the forecasting error matrix in the hydrologic forecasting process before the hydrologic forecasting value and the forecasting error matrix percentile interval are calculated according to the obtained hydrologic forecasting.
Further: the construction submodule, comprising:
the first determination unit is used for determining a prediction value and a prediction error in a hydrologic prediction process according to historical hydrologic prediction data and historical hydrologic actual measurement data;
the first establishing unit is used for constructing a forecast value and a forecast error matrix according to the forecast value and the forecast error;
a collecting unit for forecasting value P 'in hydrologic forecasting process according to percentile forecasting value and forecasting error matrix'ijAnd (5) grouping the components in different regions.
Further: the correction module is further used for correcting the acquired hydrologic forecast based on a pre-established correction model, and the correction comprises the following steps:
P't＝f(P't)
wherein: p 'on the left'tRepresenting a specific forecast value in a forecast value interval of the hydrologic forecast process; f (P't) Representing the correction values for a typical subset of the forecasted time series tracks given by K-means clustering.
The method adopts K-means clustering to carry out typical feature subset division on a prediction value and prediction error value matrix, and realizes prediction value correction by using the steps and the given equation of the method through screening the numerical value interval to which the prediction value belongs and the corresponding feature subset.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art can make modifications and equivalents to the embodiments of the present invention without departing from the spirit and scope of the present invention, which is set forth in the claims of the present application.
Claims (17)
1. A Japanese hydrologic prediction error correction method is characterized by comprising the following steps:
acquiring hydrologic forecast data;
correcting the acquired hydrologic forecast data based on a pre-established correction model;
wherein the pre-established correction model comprises: and clustering and generating the K-means value of the percentile interval based on the historical hydrologic prediction value and the prediction error matrix.
2. The method of correcting a Japanese hydrologic prediction error as set forth in claim 1, wherein: the construction process of the pre-established correction model comprises the following steps:
calculating a prediction value and a prediction error matrix percentile interval in the hydrologic prediction process according to the historical hydrologic prediction data and the historical hydrologic actual measurement data;
performing K-means clustering on the forecasting values of the hydrologic forecasting process and the forecasting error matrix percentile interval;
establishing a typical category regression model of the prediction value after K-means clustering and the prediction error matrix percentile interval;
and establishing a correction model according to the typical category regression model.
3. The method of correcting a Japanese hydrologic prediction error as set forth in claim 2, wherein: before calculating the prediction value and the prediction error matrix percentile interval in the hydrologic prediction process according to the acquired historical hydrologic prediction data and the acquired historical hydrologic actual measurement data, the method further comprises the following steps:
determining a forecast value and a forecast error in a hydrologic forecast process according to historical hydrologic forecast data and historical hydrologic actual measurement data;
constructing a prediction value and prediction error matrix according to the prediction value and the prediction error;
and (4) performing interval collection of forecast values in the hydrologic forecast process according to the forecast values and the forecast error matrix by a percentile method.
4. The method of correcting a Japanese hydrologic prediction error as claimed in claim 3, characterized in that: the hydrologic prediction error calculation formula is as follows:
Eij＝Pij-P'ij；
wherein E isijTo predict errors, PijIs measured value, P'ijPredicting a value for the hydrologic prediction process, wherein i is more than or equal to 1 and less than or equal to a set threshold, the set threshold is the time resolution between 24 hours/day-by-day samples, and j is more than or equal to 1 and less than or equal to n; n is the number of days of forecast that are counted in the analysis.
5. The method of correcting a Japanese hydrologic prediction error as claimed in claim 4, characterized in that: the hydrologic prediction process prediction value and prediction error matrix are expressed as:
A＝(Pij′,Eij)
wherein: x is a prediction value and prediction error matrix, Ai1、Ai2...AijBoth represent the hydrologic prediction process prediction value and the 1 st column of the prediction error matrix at row i and the 2 nd column of the prediction error matrix at row i.
6. The method of correcting a Japanese hydrologic prediction error as claimed in claim 5, characterized in that: the step of grouping the forecast values in the hydrologic forecast process according to the percentile method forecast values and the forecast error matrix comprises the following steps:
forecasting value P 'through hydrologic forecasting process'ijTaking ascending order as a premise, and carrying out P 'according to a percentile method'ijPercentile inter-partition collection, P'ijThe percentile inter-partition collection is expressed as: x (T)t，Eij) Represents when P'ijAn error sample corresponding to the prediction value interval Tt of the hydrologic prediction process selected at random, wherein the prediction value interval of the hydrologic prediction process is P'ijThe error samples are classified under the condition of each percentile interval,t＝[1,100,1]。
7. the method of correcting a Japanese hydrologic prediction error as set forth in claim 2, wherein: the K-means clustering generation of the prediction value and the prediction error matrix percentile interval in the hydrologic prediction process is represented as follows:
wherein: f (P)k) K-means clustering representing prediction values of the hydrologic prediction process and the prediction error matrix percentile interval; pkTo make f (P)k) One division at a minimum, Pk＝{C1，…，Ck}；Denotes the center position of the nth cluster, n is 1, …, k, d (x)y,mn) Denotes xyTo mnA distance; x is the number ofyIs X (T)t，Eij) The elements of (1); m isnRepresents the center position of the nth cluster; t istAnd representing the forecast value interval of the randomly selected hydrological forecasting process.
8. The method of correcting a Japanese hydrologic prediction error as set forth in claim 2, wherein: the typical class regression model is represented as:
Eij＝a·P'ij+b
wherein: a and b are preset regression coefficients, P'ijPredicting values for a hydrologic prediction process; eijA typical class regression model, representing the prediction error.
9. The method of correcting a Japanese hydrologic prediction error as set forth in claim 2, wherein: the making of the correction model according to the typical category regression model comprises:
and determining a correction model between the forecast value of the hydrologic prediction process and the measured value of the hydrologic prediction process of the typical category regression model.
10. The method of correcting a Japanese hydrologic prediction error as claimed in claim 9, characterized in that: the calibration model for determining the distance between the predicted value of the hydrologic prediction process and the measured value of the hydrologic prediction process is expressed by the following formula:
Pij＝f(P'ij)
wherein: p'ijPredicting values for hydrologic prediction processes，PijIs a hydrologic forecast process measured value f (P'ij) Is a function with the hydrologic prediction process prediction value as an independent variable.
11. The method of correcting a Japanese hydrologic prediction error as claimed in claim 10, characterized in that: correcting the acquired hydrologic forecast based on a pre-established correction model, comprising: when the forecast value of the hydrologic forecast process is a specific forecast value, correcting according to the following formula:
P't＝f(P't)
wherein: p 'on the left'tA correction result representing a specific prediction value of the hydrologic prediction process; f (P't) Representing a function with hydrologically predicted process specific prediction values as arguments.
12. A Japanese hydrologic forecast error correction system is characterized in that: the method comprises the following steps:
the acquisition module is used for acquiring hydrologic forecast data;
the correction module is used for correcting the acquired hydrologic forecast data based on a pre-established correction model; wherein,
the pre-established correction model comprises: and clustering and generating the K-means value of the percentile interval based on the historical hydrologic prediction value and the prediction error matrix.
13. The system for correcting errors in forecasting hydrology before day of claim 12, wherein: the device also comprises a formulation module which is used for formulating the correction model in advance.
14. The system for correcting errors in forecasting hydrology before day of claim 13, wherein: the formulation module comprises:
the calculation submodule is used for calculating a forecast value and a forecast error matrix percentile interval in the hydrologic forecast process according to the historical hydrologic forecast data and the historical hydrologic actual measurement data;
the clustering submodule is used for carrying out K-means clustering on the forecasting values of the hydrologic forecasting process and the forecasting error matrix percentile interval;
the first establishing submodule is used for establishing a typical category regression model of the prediction value after the K-means clustering and the prediction error matrix percentile interval;
and the second establishing submodule is used for establishing a correction model according to the typical category regression model.
15. The system for correcting errors of Japanese hydrologic prediction according to claim 12, further comprising a construction sub-module for constructing a hydrologic prediction process prediction value and prediction error matrix before said calculating of hydrologic prediction process prediction value and prediction error matrix percentile interval from the acquired hydrologic prediction.
16. The method of correcting a Japanese hydrologic prediction error as claimed in claim 15, wherein: the construction submodule, comprising:
the first determination unit is used for determining a prediction value and a prediction error in a hydrologic prediction process according to historical hydrologic prediction data and historical hydrologic actual measurement data;
the first establishing unit is used for constructing a hydrologic prediction process forecast value and a forecast error matrix according to the hydrologic prediction process forecast value and the forecast error;
a collecting unit for forecasting value P 'in hydrologic forecasting process according to percentile forecasting value and forecasting error matrix'ijAnd (5) grouping the components in different regions.
17. The system for correcting errors in forecasting hydrology before day of claim 12, wherein: the correction module is further used for correcting the hydrologic forecast process forecast value according to the following formula when the forecast value is a specific forecast value:
P't＝f(P't)
wherein: p 'on the left'tA correction result representing a specific prediction value of the hydrologic prediction process; f (P't) Representing a function with hydrologically predicted process specific prediction values as arguments.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810085197.1A CN110110339B (en) | 2018-01-29 | 2018-01-29 | Japanese hydrologic forecast error correction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810085197.1A CN110110339B (en) | 2018-01-29 | 2018-01-29 | Japanese hydrologic forecast error correction method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110110339A true CN110110339A (en) | 2019-08-09 |
| CN110110339B CN110110339B (en) | 2022-10-11 |
Family
ID=67483022
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810085197.1A Active CN110110339B (en) | 2018-01-29 | 2018-01-29 | Japanese hydrologic forecast error correction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110110339B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113158394A (en) * | 2020-01-22 | 2021-07-23 | 河海大学 | Basin hydrological model error correction method and device based on evaporation error compensation |
| WO2021218457A1 (en) * | 2020-04-28 | 2021-11-04 | 中国长江三峡集团有限公司 | Method for performing runoff forecast under influence of upstream reservoir group by using forecast errors |
| WO2022104709A1 (en) * | 2020-11-19 | 2022-05-27 | 中山大学 | Monthly-scale rainfall forecast correction method coupled with gamma and gaussian distribution |
| CN116595945A (en) * | 2023-07-14 | 2023-08-15 | 浙江大华技术股份有限公司 | High-precision simulation scattering parameter extraction method, electronic equipment and storage medium |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106650767A (en) * | 2016-09-20 | 2017-05-10 | 河海大学 | Flood forecasting method based on cluster analysis and real time correction |
-
2018
- 2018-01-29 CN CN201810085197.1A patent/CN110110339B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106650767A (en) * | 2016-09-20 | 2017-05-10 | 河海大学 | Flood forecasting method based on cluster analysis and real time correction |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113158394A (en) * | 2020-01-22 | 2021-07-23 | 河海大学 | Basin hydrological model error correction method and device based on evaporation error compensation |
| CN113158394B (en) * | 2020-01-22 | 2022-08-19 | 河海大学 | Basin hydrological model error correction method and device based on evaporation error compensation |
| WO2021218457A1 (en) * | 2020-04-28 | 2021-11-04 | 中国长江三峡集团有限公司 | Method for performing runoff forecast under influence of upstream reservoir group by using forecast errors |
| GB2599862A (en) * | 2020-04-28 | 2022-04-13 | China Three Gorges Corp | Method for performing runoff forecast under influence of upstream reservoir group by using forecast errors |
| GB2599862B (en) * | 2020-04-28 | 2022-09-07 | China Three Gorges Corp | Method for forecasting runoff under influence of upstream reservoir group by utilizing forecasting errors |
| WO2022104709A1 (en) * | 2020-11-19 | 2022-05-27 | 中山大学 | Monthly-scale rainfall forecast correction method coupled with gamma and gaussian distribution |
| CN116595945A (en) * | 2023-07-14 | 2023-08-15 | 浙江大华技术股份有限公司 | High-precision simulation scattering parameter extraction method, electronic equipment and storage medium |
| CN116595945B (en) * | 2023-07-14 | 2023-10-20 | 浙江大华技术股份有限公司 | High-precision simulation scattering parameter extraction method, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110110339B (en) | 2022-10-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Oprea et al. | Machine learning algorithms for short-term load forecast in residential buildings using smart meters, sensors and big data solutions | |
| Ceci et al. | Predictive modeling of PV energy production: How to set up the learning task for a better prediction? | |
| CN110110339B (en) | Japanese hydrologic forecast error correction method and system | |
| Corizzo et al. | Anomaly detection and repair for accurate predictions in geo-distributed big data | |
| US20150317589A1 (en) | Forecasting system using machine learning and ensemble methods | |
| CN116596044B (en) | Power generation load prediction model training method and device based on multi-source data | |
| CN114548509B (en) | Multi-type load joint prediction method and system for multi-energy system | |
| Meng et al. | A two-stage short-term traffic flow prediction method based on AVL and AKNN techniques | |
| CN114819374B (en) | Regional new energy ultra-short term power prediction method and system | |
| CN113379168B (en) | Time series prediction processing method, device and equipment | |
| CN113449919B (en) | Power consumption prediction method and system based on feature and trend perception | |
| CN116307291B (en) | Distributed photovoltaic power generation prediction method and prediction terminal based on wavelet decomposition | |
| CN115034485A (en) | Wind power interval prediction method and device based on data space | |
| Kim et al. | Short-term electric load forecasting using data mining technique | |
| CN115577752A (en) | Ultra-short-term photovoltaic power prediction method and system | |
| CN116187835A (en) | Data-driven-based method and system for estimating theoretical line loss interval of transformer area | |
| CN109460893B (en) | Photovoltaic power station weather type correlation index calculation method and system | |
| CN114186733A (en) | Short-term load prediction method and device | |
| Tang et al. | On missing traffic data imputation based on fuzzy C-means method by considering spatial–temporal correlation | |
| CN109635008B (en) | Equipment fault detection method based on machine learning | |
| Dos Santos et al. | Preprocessing in fuzzy time series to improve the forecasting accuracy | |
| CN111310121A (en) | New energy output probability prediction method and system | |
| CN111262248A (en) | Random power flow analysis and calculation method and system | |
| WO2022222230A1 (en) | Indicator prediction method and apparatus based on machine learning, and device and storage medium | |
| CN114676931A (en) | Electric quantity prediction system based on data relay technology |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |