CN112989705A - Method and device for predicting reservoir entry flow value, electronic device and medium - Google Patents
Method and device for predicting reservoir entry flow value, electronic device and medium Download PDFInfo
- Publication number
- CN112989705A CN112989705A CN202110341420.6A CN202110341420A CN112989705A CN 112989705 A CN112989705 A CN 112989705A CN 202110341420 A CN202110341420 A CN 202110341420A CN 112989705 A CN112989705 A CN 112989705A
- Authority
- CN
- China
- Prior art keywords
- prediction
- reservoir
- target
- predicted
- sequence
- 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
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000012549 training Methods 0.000 claims description 45
- 230000007613 environmental effect Effects 0.000 claims description 15
- 238000003860 storage Methods 0.000 claims description 15
- 238000012216 screening Methods 0.000 claims description 8
- 238000010801 machine learning Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 4
- 238000011144 upstream manufacturing Methods 0.000 claims description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 abstract description 3
- 230000015654 memory Effects 0.000 description 12
- 238000004364 calculation method Methods 0.000 description 8
- 238000012545 processing Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000012417 linear regression Methods 0.000 description 4
- 238000010248 power generation Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000009826 distribution Methods 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 238000004220 aggregation Methods 0.000 description 2
- 230000002776 aggregation Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000003646 Spearman's rank correlation coefficient Methods 0.000 description 1
- 230000003190 augmentative effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000013277 forecasting method Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 230000008707 rearrangement Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- 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
- 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—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Entrepreneurship & Innovation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Artificial Intelligence (AREA)
- Quality & Reliability (AREA)
- Development Economics (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the invention discloses a method and a device for predicting a reservoir entry flow value, electronic equipment and a medium. The method comprises the following steps: determining a plurality of historical moments according to the predicted moment corresponding to the target reservoir, and forming an initialized predicted characteristic sequence according to a plurality of description information of the target reservoir at each historical moment; iteratively updating the prediction characteristic sequence by using at least one pre-trained prediction model until an iteration ending condition is met; and when the iteration ending condition is met, the final prediction results corresponding to each prediction model are obtained, and the predicted warehousing flow value of the target reservoir at the prediction time is determined according to each final prediction result. According to the scheme of the embodiment of the invention, the flow value flowing into the water reservoir is accurately predicted.
Description
Technical Field
The embodiment of the invention relates to the technical field of computers, in particular to a method and a device for predicting a reservoir warehousing flow value, electronic equipment and a medium.
Background
For hydroelectric power plants, electricity generation is the main source of economic benefit, and water is the raw material for production. How to accurately predict the flow value entering a reservoir of a hydropower station, help the hydropower station to reasonably arrange flood control and power generation plan scheduling work, avoid flood disasters and improve power generation economic benefits is a key problem of industry attention.
At the present stage, the flow value entering a reservoir of the hydropower station is mainly predicted by a time sequence prediction method, for example, a traditional statistical time sequence model, a linear regression model or a long-short term memory artificial neural network; however, the methods are difficult to accurately predict the flow value entering the reservoir of the hydropower station; for example, the influence factor corresponding to each parameter cannot be accurately determined for the conventional statistical timing model, and the model operation efficiency is low; long-sequence periodic prediction is difficult to realize for machine learning models such as linear regression; the method aims at the problems that deep learning models such as a long-term and short-term memory artificial neural network are poor in stability and the flow value entering a reservoir of a hydropower station is difficult to accurately predict.
Disclosure of Invention
The embodiment of the invention provides a method and a device for predicting a warehousing flow value of a reservoir, electronic equipment and a medium, so as to accurately predict the flow value flowing into the reservoir.
In a first aspect, an embodiment of the present invention provides a method for predicting a reservoir entry flow value, including:
determining a plurality of historical moments according to the predicted moment corresponding to the target reservoir, and forming an initialized predicted characteristic sequence according to a plurality of description information of the target reservoir at each historical moment;
iteratively updating the prediction characteristic sequence by using at least one pre-trained prediction model until an iteration ending condition is met;
and when the iteration ending condition is met, the final prediction results corresponding to each prediction model are obtained, and the predicted warehousing flow value of the target reservoir at the prediction time is determined according to each final prediction result.
In a second aspect, an embodiment of the present invention further provides a device for predicting a reservoir entry flow value, including:
the initialized prediction characteristic sequence generation module is used for determining a plurality of historical moments according to the prediction moments corresponding to the target reservoir and forming an initialized prediction characteristic sequence according to a plurality of description information of the target reservoir at each historical moment;
the prediction characteristic sequence iteration module is used for carrying out iteration updating on the prediction characteristic sequence by using at least one pre-trained prediction model until an iteration ending condition is met;
and the warehousing flow value prediction module is used for acquiring final prediction results respectively corresponding to each prediction model when the iteration ending condition is met, and determining the predicted warehousing flow value of the target reservoir at the prediction time according to each final prediction result.
In a third aspect, an embodiment of the present invention further provides an electronic device, where the electronic device includes:
one or more processors;
a storage device for storing one or more programs,
when the one or more programs are executed by the one or more processors, the one or more processors implement the method for predicting the warehousing traffic value of the reservoir according to any embodiment of the invention.
In a fourth aspect, the present invention further provides a storage medium containing computer-executable instructions, where the computer-executable instructions, when executed by a computer processor, are configured to perform a method for predicting a reservoir entry flow value of a reservoir according to any one of the embodiments of the present invention.
According to the embodiment of the invention, a plurality of historical moments are determined according to the predicted moment corresponding to the target reservoir, and an initialized predicted characteristic sequence is formed according to a plurality of description information of the target reservoir at each historical moment; iteratively updating the prediction characteristic sequence by using at least one pre-trained prediction model until an iteration ending condition is met; and when the iteration ending condition is met, the final prediction results corresponding to each prediction model are obtained, and the predicted warehousing flow value of the target reservoir at the prediction time is determined according to each final prediction result, so that the flow value flowing into the reservoir is accurately predicted.
Drawings
Fig. 1 is a flowchart of a method for predicting a reservoir entry flow value in a first embodiment of the present invention;
fig. 2 is a flowchart of a method for predicting a reservoir warehousing flow rate value in the second embodiment of the invention;
fig. 3 is a flowchart of a method for predicting a flow value of a reservoir entering a reservoir in a third embodiment of the present invention;
fig. 4 is a flowchart of a method for predicting a flow value of a reservoir entering a reservoir in a third embodiment of the present invention;
fig. 5 is a schematic structural diagram of an apparatus for predicting a warehousing traffic value of a reservoir in the fourth embodiment of the present invention;
fig. 6 is a schematic structural diagram of an electronic device in a fifth embodiment of the present invention.
Detailed Description
The embodiments of the present invention will be described in further detail with reference to the drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of and not restrictive on the broad invention. It should be further noted that, for convenience of description, only some structures, not all structures, relating to the embodiments of the present invention are shown in the drawings.
Example one
Fig. 1 is a flowchart of a method for predicting a warehousing flow value of a reservoir according to an embodiment of the present invention, where the embodiment is applicable to a case of predicting a flow value flowing into the reservoir at a prediction time, and the method may be implemented by a device for predicting a warehousing flow value of a reservoir, where the device may be implemented by software and/or hardware, and is integrated in an electronic device for implementing the method, where the electronic device related in the embodiment may be a computer, a server, a tablet computer, or the like. Specifically, referring to fig. 1, the method specifically includes the following steps:
and step 110, determining a plurality of historical moments according to the predicted moment corresponding to the target reservoir, and forming an initialized predicted characteristic sequence according to a plurality of description information of the target reservoir at each historical moment.
The target reservoir may be any reservoir or lake, for example, a power station reservoir or a natural lake, and the present embodiment is not limited thereto. The predicted time may be any future time, such as 8 am tomorrow or 22 pm tomorrow, which is not limited in this embodiment.
The description information describes statistical values or time series characteristics and the like for processing the confluence flow, rainfall flow, forecast rainfall amount and environmental data of the target reservoir at each historical time, which is not limited in this embodiment. Wherein, the statistical value can be mean value, sum, maximum value, minimum value, variance, skewness or kurtosis, etc.; the timing characteristics may be absolute energy values, first order difference absolute sums, auto-regressive coefficients, or approximate entropy, etc.
In an optional implementation manner of this embodiment, in the process of determining the warehousing flow rate value of the target reservoir at the prediction time, a plurality of historical times may be determined, and an initialized prediction feature sequence corresponding to the prediction time may be formed according to a plurality of description information of the target reservoir at each historical time.
For example, if the predicted time is 2 months and 26 days, the determined historical time may be 2 months and 19 days to 2 months and 25 days; furthermore, historical warehousing flow values of the target reservoir of 2 months, 19 days to 2 months, 25 days can be obtained, description information at each historical moment is generated according to the historical warehousing flow values, and the description information is arranged in sequence to form an initialized prediction characteristic sequence.
In the present embodiment, the actual rainfall at the prediction time is represented by the forecast rainfall at the prediction time in the initialized prediction feature sequence.
And step 120, iteratively updating the prediction characteristic sequence by using at least one pre-trained prediction model until an iteration ending condition is met.
The prediction model may be a machine learning model such as xgboot, lightgm, or catboost, which is not limited in this embodiment.
The iteration condition may be a period condition, for example, the period is set to 7, and after 7 iterations, it may be determined that the end iteration condition is satisfied.
In an optional implementation manner of this embodiment, after the initialized predicted feature sequence is formed, the predicted feature sequence may be iteratively updated by using at least one pre-trained prediction model until an end iteration condition is satisfied.
Exemplarily, after the initialized prediction feature sequence is formed, the prediction feature sequence may be input into an xgboot and lightgbm prediction model to obtain 2 prediction results, and the 2 prediction results are subjected to weighted average calculation, and the calculation results are spliced with the prediction feature sequence to generate a new prediction feature sequence; further, the operation of inputting the new predicted feature sequence into the xgboot and lightgbm prediction models is continuously performed until the end iteration condition is satisfied.
And step 130, obtaining final prediction results corresponding to each prediction model when the iteration ending condition is met, and determining the predicted warehousing flow value of the target reservoir at the prediction time according to each final prediction result.
In an optional implementation manner of this embodiment, after the iteration update is finished, when the iteration condition is satisfied, the final prediction results corresponding to each prediction model may be obtained, and the predicted warehousing flow value of the target reservoir at the prediction time is determined according to each final prediction result.
In the above example, the predicted feature sequence is input into the xgboot and lightgbm prediction models to obtain 2 prediction results, and the 2 prediction results are subjected to weighted average calculation, and the calculation results are spliced with the predicted feature sequence to generate a new predicted feature sequence; and further, continuously executing the operation of inputting the new prediction characteristic sequence into the xgboot and lightgbm prediction models until the iteration ending condition is met, stopping the iteration updating of the prediction characteristic sequence, obtaining the final prediction results corresponding to the xgboot and lightgbm prediction models when the last iteration is ended, and determining the predicted warehousing flow value of the target reservoir at the prediction time according to each final prediction result.
According to the scheme of the embodiment, a plurality of historical moments are determined according to the prediction moment corresponding to the target reservoir, and an initialized prediction characteristic sequence is formed according to a plurality of description information of the target reservoir at each historical moment; iteratively updating the prediction characteristic sequence by using at least one pre-trained prediction model until an iteration ending condition is met; and when the iteration finishing condition is met, the final prediction results corresponding to each prediction model are obtained, and the predicted warehousing flow value of the target reservoir at the prediction moment is determined according to each final prediction result, so that the flow value flowing into the reservoir is accurately predicted, the hydropower station can be helped to reasonably arrange flood control and power generation plan scheduling work, flood disasters are avoided, and the power generation economic benefit is improved.
Example two
Fig. 2 is a flowchart of a method for predicting a warehousing traffic value of a reservoir in the second embodiment of the present invention, which is a further refinement of the above technical solutions, and the technical solutions in this embodiment may be combined with various alternatives in one or more of the above embodiments. As shown in fig. 2, the method for predicting the reservoir storage flow rate value may include the following steps:
And step 220, iteratively updating the prediction characteristic sequence by using a plurality of pre-trained similar prediction models until an iteration ending condition is met.
Wherein the model structures and/or model parameters of different prediction models are not completely the same; in this embodiment, the different prediction models may be xgboot, lightgbm, and catboost models based on a GBDT (Gradient Boosting Tree) framework, or may be other models, which is not limited in this embodiment.
In an optional implementation manner of this embodiment, after the initialized predicted feature sequence is formed, the predicted feature sequence may be iteratively updated by using the pre-trained xgboot, lightgbm and catboost prediction models, respectively.
Optionally, the iteratively updating the predicted feature sequence by using a plurality of pre-trained similar prediction models until an iteration ending condition is met, which may include: respectively inputting the prediction characteristic sequences into each prediction model, and respectively obtaining prediction results output in each prediction model; and after updating the prediction characteristic sequences by using the prediction results, returning to execute the operation of respectively inputting the prediction characteristic sequences into the prediction models until the iteration ending condition is met.
In an optional implementation manner of this embodiment, updating the prediction feature sequence using each prediction result may include: carrying out weighted average operation on the prediction result output by each prediction model to generate a reference prediction result; and splicing the reference prediction result with the prediction characteristic sequence to generate a new prediction characteristic sequence, and updating the prediction characteristic sequence into a new prediction characteristic sequence.
In a specific example of this embodiment, after the initialized prediction feature sequence is formed, the initialized prediction feature sequence may be respectively input into the xgboot, lightgbm, and catboost prediction models, and the xgboot, lightgbm, and catboost prediction models respectively output prediction results corresponding to the prediction feature sequence.
Further, the prediction results corresponding to the prediction feature sequences can be output by the xgboot prediction model, the lightgbm prediction model and the catboost prediction model respectively to perform weighted average operation, the prediction results are spliced with the prediction feature sequences to generate new prediction feature sequences, and the generated new prediction feature sequences are input into the xgboot prediction model, the lightgbm prediction model and the catboost prediction model respectively until the iteration ending condition is met.
The prediction weights corresponding to the prediction models may be the same or different, and may be any value, which is not limited in this embodiment.
In an optional implementation manner of this embodiment, after the iteration is ended, when the iteration ending condition is met, a prediction result currently output by each prediction model may be further obtained, and the prediction result is used as a final prediction result; and according to the prediction weights respectively corresponding to the prediction models, performing weighted average calculation on the prediction results to obtain the predicted warehousing flow value of the target reservoir at the prediction time.
For example, weighted average calculation may be performed on final prediction results output by the xgboot, lightgbm and catboost prediction models, so as to obtain a predicted warehousing flow value of the target reservoir at the prediction time.
In a specific example of this embodiment, the single prediction results of the three models xgboost, lightgbm and catboost can be weighted-averaged by the following formula:
Ypredict＝αYpredict_xgboost+βYpredict_lightgbm+γYpredict_catboost
where α, β, and γ are prediction weight values.
According to the scheme of the embodiment, after the plurality of historical moments are determined according to the prediction moment corresponding to the target reservoir, the initialized prediction characteristic sequence is formed according to the plurality of description information of the target reservoir at each historical moment, the prediction characteristic sequence can be iteratively updated by using a plurality of similar pre-trained prediction models until the iteration ending condition is met, the warehousing flow value can be predicted according to the prediction results of different prediction models, and the prediction accuracy of the warehousing flow value is improved.
EXAMPLE III
Fig. 3 is a flowchart of a method for predicting a warehousing traffic value of a reservoir in a third embodiment of the present invention, which is a further refinement of the above technical solutions, and the technical solutions in this embodiment may be combined with various alternatives in one or more of the above embodiments. As shown in fig. 3, the method for predicting the reservoir storage flow rate value may include the following steps:
and 310, determining a plurality of historical moments according to the predicted moment corresponding to the target reservoir, and forming an initialized predicted characteristic sequence according to a plurality of description information of the target reservoir at each historical moment.
Wherein the sample data comprises: confluence flow, rainfall flow, forecast rainfall amount, environmental data and the like, wherein the confluence flow can be the water flow of other rivers flowing into a target reservoir at a certain moment; the rainfall flow can be the rainfall monitored by an observation station at a certain moment; the forecast rainfall can be the forecast of the target rainfall at a certain moment by the weather forecast; the environmental forecast may include temperature, wind power or wind direction, etc. at a certain time.
In an optional implementation manner of this embodiment, before the prediction feature sequence is iteratively updated by using at least one pre-trained prediction model, sample data of a historical time corresponding to the target reservoir may be further generated, so as to generate training data; furthermore, a plurality of prediction models are obtained through simultaneous training according to the training data.
In an optional implementation manner of this embodiment, acquiring sample data corresponding to the target reservoir may include: acquiring confluence flow from an upstream and/or river channel to a target reservoir at intervals of a first set time interval; acquiring rainfall flow monitored by each observation station related to the target reservoir at each second set time interval; acquiring the forecast rainfall of the rainfall forecast prediction in a third set time interval; and acquiring environmental data monitored by the environmental sensor every fourth time interval, wherein the environmental data comprises temperature, wind speed or wind direction.
The first set time interval, the second set time interval, the third set time interval, and the fourth time interval may be the same, and for example, all of them may be 3 hours or 1 hour; for example, the first set time interval is 3 hours, the second set time interval, the third set time interval, and the fourth set time interval is 1 hour, and the like, which is not limited in this embodiment.
In an optional implementation manner of this embodiment, generating training data according to the sample data may include: preprocessing each sample data to obtain target sample data; extracting features of each target sample data to obtain a plurality of description information corresponding to each historical moment, screening each description information, and generating a training feature sequence corresponding to each historical moment; each training signature sequence is determined as training data.
For example, preprocessing each sample data to obtain target sample data may include: performing global maximum and minimum normalization processing on the confluent flow, the rainfall flow, the air temperature and the average wind speed data; according to the national standard: the technical specification of GB/T37301 and 2019 ground meteorological data service products is used for encoding wind direction data; and performing missing value processing on the confluence flow, the rainfall flow and the forecast rainfall, and the like.
In an optional implementation manner of this embodiment, performing feature extraction on each target sample data to obtain a plurality of description information corresponding to each time may include: determining target warehousing flow corresponding to the target historical time and reference warehousing flow at each time before the target historical time; determining a statistical value of the warehousing flow within each set time range according to the target warehousing flow and each reference warehousing flow; the statistical values include: mean, sum, maximum, minimum, variance, skewness, or kurtosis; and/or determining at least one time sequence characteristic according to the target warehousing flow and each reference warehousing flow; the timing characteristics include: absolute energy values, first order difference absolute sums, autoregressive coefficients, or approximate entropy. It should be noted that, when long-term prediction is performed on the warehousing traffic, the actual rainfall is an important factor affecting the prediction result, but the actual rainfall is absent (observed value) within the prediction time, so the target warehousing traffic related in this embodiment can be filled by forecasting the rainfall.
For example, if the target historical time is t, the warehousing traffic value at the time t and the reference warehousing traffic value at each time (e.g., t-1, t-2, t-3, …, t-w1, where w1 is any positive integer, e.g., 56 or 60, etc.) before the time t may be determined; further, a time sequence difference sliding window technology, a sliding window size w1 and a calculation interval [ t-i, t ], wherein i is period1 j, j is 1,2, … and w1/period1, can be adopted for calculating the statistical value of the warehousing flow rate; the period1 may be any value, or may be obtained by calculating an autocorrelation coefficient and a partial correlation coefficient of the original warehousing traffic sequence, which is not limited in this embodiment.
The skewness is a measure of the direction and degree of skew of the statistical data distribution, and is a numerical feature of the degree of asymmetry of the statistical data distribution. Defining the degree of skewness as the third-order normalized moment of the sample; the kurtosis is also called kurtosis coefficient; and characterizing the characteristic number of the peak value of the probability density distribution curve at the average value. Intuitively, the kurtosis reflects the sharpness of the peak, and the kurtosis calculation method of the random variable comprises the following steps: the ratio of the fourth central moment of the random variable to the square of the variance.
Furthermore, the target warehousing flow rate in each sliding window and each reference warehousing flow rate can be extracted to determine at least one time sequence characteristic, and it needs to be noted that the number of the target warehousing flow rate value in each sliding window is the same as that of the reference warehousing flow rate values; illustratively, the extracted timing characteristics may include: abs _ energy, the absolute energy of the returned time series data; absolute _ sum _ of _ changes (first order difference absolute sum) that returns the sum of absolute values of the first order difference results of the time series data; agg _ autocorrelation (aggregate statistical characteristics of autocorrelation coefficients of each order), aggregate statistical characteristics of autocorrelation coefficients of each order; agg _ linear _ true (linear regression based on the block timing aggregation value), and returning the linear regression of the block timing data after block aggregation; approximate entropy, which measures the periodicity, unpredictability and volatility of time series data; ar _ coefficient (autoregressive coefficient) which measures the periodicity, unpredictability and volatility of time series data; the method comprises the steps that (1) an augmented _ dickey _ fuller (extended diky-Fowler inspection-ADF inspection) is used for testing whether an autoregressive model has a unit root or not and measuring the stationarity of time sequence data; autocorrelation (lag order autocorrelation), calculating autocorrelation of lag order lag time series data; binned _ entropy, dividing the whole sequence into max _ bins buckets according to values, putting each value into a corresponding bucket, and solving entropy; transform _ correlation, dividing the entire sequence into max _ bins by value, then putting each value into a corresponding bucket, and then calculating the correlation; c3 (time series data non-linearity measure), time series data non-linearity measure; change _ quantiles (timing data description statistics for a given interval), timing data description statistics for a given interval; cid _ ce (time series data complexity) to evaluate the complexity of a time series, the more complex the series having more valleys; count _ above (over a specified value of a ratio), returning the percentage of the sequence above the specified value; counting _ above _ mean (the number is higher than the mean value), and counting the number which is higher than the mean value of the time series data; count _ below (less than a specified value of a ratio), returning the percentage of the sequence that is below the specified value; cwt _ coefficients (Ricker wavelet analysis), continuous wavelet analysis, Ricker wavelets are a commonly used wavelet type in seismic exploration, and Ricker wavelets are strictly derived based on wave equations; energy _ ratio _ by _ chunks, and after the time-series data is chunked, the entropy ratio of the target block data to the entire is calculated. When the data is not uniform enough, the redundant data is scattered in the previous block; fft _ aggregated (spectrum statistics of absolute fourier transform), returning the values of spectrum centroid, kurtosis, skewness after absolute fourier transform; fft _ coefficient (fourier transform coefficient), calculating the coefficient of one-dimensional discrete fourier sequence based on fast fourier transform algorithm; friedrich _ coefficients (polynomial coefficients fitted to Langevin model) based on determining polynomial coefficients fitted to Langevin model of dynamics; large _ standard _ definition (whether the standard deviation is multiple of the range of the standard deviation), and whether the standard deviation is r times of the data range; changest _ strike _ above _ mean (the longest continuous self-column length on the mean), returns the length of the longest continuous subsequence in x that is greater than the mean of x; change _ strike _ below _ mean (the longest continuous self-column length under the mean), return the length of the longest continuous subsequence in x that is less than the mean of x; mean _ second _ derivative _ central (mean of center of second derivative), return the mean of the center approximation of the second derivative.
Further, the screening of each description information to generate a training feature sequence corresponding to each historical time may include: and calculating correlation coefficients among the features, eliminating the features with the correlation lower than a set threshold value, and generating a training feature sequence according to the residual features.
In specific implementation, the correlation coefficient of each training characteristic sequence and the target warehousing flow value can be calculated, and characteristic variables with low correlation are removed; calculating the spearman rank correlation coefficient among the training characteristic sequences, and then selecting a training characteristic sequence (which can be selected randomly) from the characteristic information of which the linear correlation is greater than a preset threshold value; feature importance output by the GBDT algorithm may also be used to screen for features whose feature importance is greater than a specified threshold.
In an optional implementation manner of this embodiment, after a plurality of training feature sequences are generated, each training feature sequence may be randomly grouped to form a training data set, a verification data set, and a test data set; further, training three prediction models of xgboost, lightgbm and catboost respectively by using a training set; during the training process of the prediction model, the hyper-parameter optimization can be carried out by using a random search and Bayesian optimization method, and 5-fold cross validation is used, wherein the validation function is MAPE (mean absolute percentage error).
And step 330, iteratively updating the prediction characteristic sequence by using at least one pre-trained prediction model until an iteration ending condition is met.
And 340, acquiring final prediction results corresponding to each prediction model when the iteration ending condition is met, and determining the predicted warehousing flow value of the target reservoir at the prediction time according to each final prediction result.
According to the scheme of the embodiment, before the prediction characteristic sequence is iteratively updated by using at least one pre-trained prediction model, sample data corresponding to the target reservoir can be obtained, and training data are generated according to the sample data; and training at least two set machine learning models respectively by using the training data to obtain each prediction model, and providing a basis for accurately determining the warehousing flow value flowing into the target reservoir at the prediction moment subsequently.
In order to make those skilled in the art better understand the method for predicting the warehousing flow rate value of the reservoir in this embodiment, fig. 4 is a flowchart of a method for predicting the warehousing flow rate value of the reservoir in the third embodiment of the present invention, and referring to fig. 4, the method specifically includes the following steps:
and step 410, acquiring confluence flow, rainfall flow, forecast rainfall and environmental data within a specified time range.
And step 420, processing each data through the data processing module.
And 430, extracting the characteristics of the processed data through a characteristic extraction module.
And 440, screening each feature through a feature screening module, and generating an initial training set and a verification set.
And step 450, specifying a prediction period and the prediction times.
And step 460, training the model.
And 470, model prediction.
And 471, summarizing prediction results.
If yes, go to step 481;
otherwise, step 482 is performed.
And step 481, splicing the grade warehousing flow predicted value to warehousing flow sequence data, and updating the training set by using a feature extraction module and a feature screening module.
And training according to the training set and the verification set to obtain a plurality of prediction models.
And step 482, outputting a storage flow prediction result.
According to the embodiment of the invention, based on a large amount of historical data and obtainable monitoring data, the reservoir warehousing flow can be accurately predicted, and the obvious safety and economic value are brought to reservoir-based production.
Example four
Fig. 5 is a schematic structural diagram of an apparatus for predicting a warehousing traffic value of a reservoir according to a fourth embodiment of the present invention, which may execute the method for predicting a warehousing traffic value of a reservoir according to the foregoing embodiments. Referring to fig. 5, the apparatus includes: an initialized predicted feature sequence generation module 510, a predicted feature sequence iteration module 520, and a warehousing traffic value prediction module 530.
An initialized prediction feature sequence generating module 510, configured to determine multiple historical times according to prediction times corresponding to the target reservoir, and form an initialized prediction feature sequence according to multiple description information of the target reservoir at each historical time;
a predicted feature sequence iteration module 520, configured to iteratively update the predicted feature sequence using at least one pre-trained prediction model until an end iteration condition is met;
and the warehousing flow value prediction module 530 is configured to obtain final prediction results corresponding to each prediction model when the iteration ending condition is met, and determine a predicted warehousing flow value of the target reservoir at the prediction time according to each final prediction result.
According to the scheme of the embodiment, a plurality of historical moments are determined according to the prediction moment corresponding to the target reservoir through an initialized prediction characteristic sequence generation module, and an initialized prediction characteristic sequence is formed according to a plurality of description information of the target reservoir at each historical moment; iteratively updating the predicted characteristic sequence by using at least one pre-trained prediction model through a predicted characteristic sequence iteration module until an iteration ending condition is met; and when the iteration ending condition is met, the final prediction results corresponding to each prediction model are obtained through the warehousing flow value prediction module, and the predicted warehousing flow value of the target reservoir at the prediction time is determined according to each final prediction result, so that the flow value flowing into the reservoir is accurately predicted.
In an alternative implementation of this embodiment, the predicted feature sequence iteration module 520 is specifically configured to
Using a plurality of pre-trained similar prediction models to iteratively update the prediction characteristic sequence until an iteration ending condition is met;
wherein the model structures, and/or model parameters of the different prediction models are not identical.
In an optional implementation manner of this embodiment, the predicted feature sequence iteration module 520 is further specifically configured to
Inputting the prediction characteristic sequences into the prediction models respectively, and obtaining prediction results output in the prediction models respectively;
and after updating the prediction characteristic sequences by using the prediction results, returning to execute the operation of respectively inputting the prediction characteristic sequences into the prediction models until the iteration ending condition is met.
In an optional implementation manner of this embodiment, the predicted feature sequence iteration module 520 is further specifically configured to
Carrying out weighted average operation on the prediction result output by each prediction model to generate a reference prediction result;
and splicing the reference prediction result with the prediction characteristic sequence to generate a new prediction characteristic sequence, and updating the prediction characteristic sequence into the new prediction characteristic sequence.
In an optional implementation manner of this embodiment, the warehousing flow value prediction module 530 is specifically configured to
When the iteration ending condition is met, the currently output prediction result of each prediction model is used as each final prediction result;
and carrying out weighted average on each final prediction result according to the prediction weight corresponding to each prediction model respectively to obtain the predicted warehousing flow value of the target reservoir at the prediction time.
In an optional implementation manner of this embodiment, the apparatus for predicting a storage flow value of a reservoir further includes: a prediction model generation module for
Acquiring sample data corresponding to the target reservoir, and generating training data according to the sample data; the sample data includes: confluent flow, rainfall flow, forecast rainfall and environmental data;
and training at least two set machine learning models respectively by using the training data to obtain each prediction model.
In an optional implementation manner of this embodiment, the prediction model generation module is specifically configured to
Acquiring confluence flow converged to the target reservoir from an upstream and/or river channel at intervals of a first set time interval;
acquiring rainfall flow monitored by each observation station related to the target reservoir at each second set time interval;
acquiring the forecast rainfall of the rainfall forecast prediction in a third set time interval;
acquiring environmental data monitored by an environmental sensor at every fourth time interval, wherein the environmental data comprises temperature, wind speed or wind direction;
accordingly, the predictive model generation module is also particularly useful for
Preprocessing each sample data to obtain target sample data;
extracting features of the target sample data to obtain a plurality of description information corresponding to each historical moment, screening the description information to generate a training feature sequence corresponding to each historical moment;
determining each training feature sequence as training data;
wherein the prediction model generation module is further specifically used for
Determining target warehousing flow corresponding to the target historical time and reference warehousing flow at each time before the target historical time; wherein the target warehousing flow is determined by the forecast rainfall;
determining a statistical value of the warehousing flow within each set time range according to the target warehousing flow and each reference warehousing flow; the statistical values include: mean, sum, maximum, minimum, variance, skewness, or kurtosis; and/or the presence of a gas in the gas,
determining at least one time sequence characteristic according to the target warehousing traffic and each reference warehousing traffic; the timing characteristics include: absolute energy values, first order difference absolute sums, autoregressive coefficients, or approximate entropy;
accordingly, the predictive model generation module is also particularly useful for
And calculating a correlation coefficient among the features, eliminating the features with the correlation lower than a set threshold value, and generating the training feature sequence according to the residual features.
The forecasting device for the warehousing flow value of the reservoir provided by the embodiment of the invention can execute the forecasting method for the warehousing flow value of the reservoir provided by any embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method.
EXAMPLE five
Fig. 6 is a schematic structural diagram of an electronic device according to a fifth embodiment of the present invention, as shown in fig. 6, the electronic device includes a processor 60, a memory 61, an input device 62, and an output device 63; the number of the processors 60 in the electronic device may be one or more, and one processor 60 is taken as an example in fig. 6; the processor 60, the memory 61, the input device 62 and the output device 63 in the electronic apparatus may be connected by a bus or other means, and the bus connection is exemplified in fig. 6.
The memory 61 may be used as a computer-readable storage medium for storing software programs, computer-executable programs, and modules, such as program instructions/modules corresponding to the method for predicting the warehousing traffic value of the reservoir in the embodiment of the present invention (for example, the initialized prediction feature sequence generation module 510, the prediction feature sequence iteration module 520, and the warehousing traffic value prediction module 530 in the prediction device for the warehousing traffic value of the reservoir). The processor 60 executes various functional applications and data processing of the electronic device by running software programs, instructions and modules stored in the memory 61, that is, the method for predicting the reservoir warehousing traffic value is realized.
The memory 61 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the terminal, and the like. Further, the memory 61 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some examples, the memory 61 may further include memory located remotely from the processor 60, which may be connected to the electronic device through a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The input device 62 may be used to receive input numeric or character information and generate key signal inputs related to user settings and function controls of the electronic apparatus. The output device 63 may include a display device such as a display screen.
EXAMPLE six
An embodiment of the present invention further provides a storage medium containing computer-executable instructions, where the computer-executable instructions are executed by a computer processor to perform a method for predicting a warehousing flow rate value of a reservoir, and the method includes:
determining a plurality of historical moments according to the predicted moment corresponding to the target reservoir, and forming an initialized predicted characteristic sequence according to a plurality of description information of the target reservoir at each historical moment;
iteratively updating the prediction characteristic sequence by using at least one pre-trained prediction model until an iteration ending condition is met;
and when the iteration ending condition is met, the final prediction results corresponding to each prediction model are obtained, and the predicted warehousing flow value of the target reservoir at the prediction time is determined according to each final prediction result.
Of course, the storage medium provided by the embodiment of the present invention includes computer-executable instructions, and the computer-executable instructions are not limited to the operations of the method described above, and may also perform related operations in the method for predicting the warehousing flow rate value of the reservoir provided by any embodiment of the present invention.
From the above description of the embodiments, it is obvious for those skilled in the art that the present invention can be implemented by software and necessary general hardware, and certainly, can also be implemented by hardware, but the former is a better embodiment in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as a floppy disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a FLASH Memory (FLASH), a hard disk or an optical disk of a computer, and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute the methods according to the embodiments of the present invention.
It should be noted that, in the embodiment of the prediction apparatus for the warehousing flow rate value of the reservoir, the included units and modules are only divided according to the functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, specific names of the functional units are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present invention.
It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.
Claims (10)
1. A method for predicting a reservoir entry flow value is characterized by comprising the following steps:
determining a plurality of historical moments according to the predicted moment corresponding to the target reservoir, and forming an initialized predicted characteristic sequence according to a plurality of description information of the target reservoir at each historical moment;
iteratively updating the prediction characteristic sequence by using at least one pre-trained prediction model until an iteration ending condition is met;
and when the iteration ending condition is met, the final prediction results corresponding to each prediction model are obtained, and the predicted warehousing flow value of the target reservoir at the prediction time is determined according to each final prediction result.
2. The method of claim 1, wherein iteratively updating the predicted feature sequence using at least one pre-trained predictive model until an end-iteration condition is satisfied comprises:
using a plurality of pre-trained similar prediction models to iteratively update the prediction characteristic sequence until an iteration ending condition is met;
wherein the model structures, and/or model parameters of the different prediction models are not identical.
3. The method of claim 2, wherein iteratively updating the predicted feature sequence using a plurality of pre-trained homogeneous prediction models until an end-of-iteration condition is satisfied comprises:
inputting the prediction characteristic sequences into the prediction models respectively, and obtaining prediction results output in the prediction models respectively;
and after updating the prediction characteristic sequences by using the prediction results, returning to execute the operation of respectively inputting the prediction characteristic sequences into the prediction models until the iteration ending condition is met.
4. The method of claim 3, wherein updating the sequence of predicted features using each of the predicted outcomes comprises:
carrying out weighted average operation on the prediction result output by each prediction model to generate a reference prediction result;
and splicing the reference prediction result with the prediction characteristic sequence to generate a new prediction characteristic sequence, and updating the prediction characteristic sequence into the new prediction characteristic sequence.
5. The method according to claim 2, wherein the step of obtaining final prediction results respectively corresponding to each prediction model when the iteration ending condition is met, and determining the predicted warehousing flow value of the target reservoir at the prediction time according to each final prediction result comprises the steps of:
when the iteration ending condition is met, the currently output prediction result of each prediction model is used as each final prediction result;
and carrying out weighted average on each final prediction result according to the prediction weight corresponding to each prediction model respectively to obtain the predicted warehousing flow value of the target reservoir at the prediction time.
6. The method according to any one of claims 1-5, further comprising, prior to iteratively updating the sequence of predicted features using at least one pre-trained predictive model:
acquiring sample data corresponding to the target reservoir, and generating training data according to the sample data; the sample data includes: confluent flow, rainfall flow, forecast rainfall and environmental data;
and training at least two set machine learning models respectively by using the training data to obtain each prediction model.
7. The method of claim 6, wherein said obtaining sample data corresponding to said target reservoir comprises at least one of:
acquiring confluence flow converged to the target reservoir from an upstream and/or river channel at intervals of a first set time interval;
acquiring rainfall flow monitored by each observation station related to the target reservoir at each second set time interval;
acquiring the forecast rainfall of the rainfall forecast prediction in a third set time interval;
acquiring environmental data monitored by an environmental sensor at every fourth time interval, wherein the environmental data comprises temperature, wind speed or wind direction;
correspondingly, the generating training data according to the sample data includes:
preprocessing each sample data to obtain target sample data;
extracting features of the target sample data to obtain a plurality of description information corresponding to each historical moment, screening the description information to generate a training feature sequence corresponding to each historical moment;
determining each training feature sequence as training data;
the performing feature extraction on each target sample data to obtain a plurality of description information corresponding to each moment includes:
determining target warehousing flow corresponding to the target historical time and reference warehousing flow at each time before the target historical time; wherein the target warehousing flow is determined by the forecast rainfall;
determining a statistical value of the warehousing flow within each set time range according to the target warehousing flow and each reference warehousing flow; the statistical values include: mean, sum, maximum, minimum, variance, skewness, or kurtosis; and/or the presence of a gas in the gas,
determining at least one time sequence characteristic according to the target warehousing traffic and each reference warehousing traffic; the timing characteristics include: absolute energy values, first order difference absolute sums, autoregressive coefficients, or approximate entropy;
correspondingly, the screening of each piece of description information to generate a training feature sequence corresponding to each historical time includes:
and calculating a correlation coefficient among the features, eliminating the features with the correlation lower than a set threshold value, and generating the training feature sequence according to the residual features.
8. An apparatus for predicting a flow value of a reservoir into which a reservoir is put, comprising:
the initialized prediction characteristic sequence generation module is used for determining a plurality of historical moments according to the prediction moments corresponding to the target reservoir and forming an initialized prediction characteristic sequence according to a plurality of description information of the target reservoir at each historical moment;
the prediction characteristic sequence iteration module is used for carrying out iteration updating on the prediction characteristic sequence by using at least one pre-trained prediction model until an iteration ending condition is met;
and the warehousing flow value prediction module is used for acquiring final prediction results respectively corresponding to each prediction model when the iteration ending condition is met, and determining the predicted warehousing flow value of the target reservoir at the prediction time according to each final prediction result.
9. An electronic device, characterized in that the electronic device comprises:
one or more processors;
a storage device for storing one or more programs,
when executed by the one or more processors, cause the one or more processors to implement the method of predicting a flow rate at which a reservoir enters the reservoir of any of claims 1-7.
10. A storage medium containing computer-executable instructions for performing the method of predicting a warehousing traffic value of a reservoir as claimed in any one of claims 1 to 7 when executed by a computer processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110341420.6A CN112989705A (en) | 2021-03-30 | 2021-03-30 | Method and device for predicting reservoir entry flow value, electronic device and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110341420.6A CN112989705A (en) | 2021-03-30 | 2021-03-30 | Method and device for predicting reservoir entry flow value, electronic device and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112989705A true CN112989705A (en) | 2021-06-18 |
Family
ID=76338412
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110341420.6A Pending CN112989705A (en) | 2021-03-30 | 2021-03-30 | Method and device for predicting reservoir entry flow value, electronic device and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112989705A (en) |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104462861A (en) * | 2014-12-31 | 2015-03-25 | 武汉大学 | Reservoir regulation decision-making method based on reservoir regulation rule synthesis |
| CN105139093A (en) * | 2015-09-07 | 2015-12-09 | 河海大学 | Method for forecasting flood based on Boosting algorithm and support vector machine |
| CN105894138A (en) * | 2016-06-28 | 2016-08-24 | 山东大学 | Optimum weighted composite prediction method for shipment amount of manufacturing industry |
| CN107885951A (en) * | 2017-11-27 | 2018-04-06 | 河海大学 | A kind of Time series hydrological forecasting method based on built-up pattern |
| CN108875161A (en) * | 2018-05-31 | 2018-11-23 | 长江勘测规划设计研究有限责任公司 | Flow grade prediction technique based on convolutional neural networks deep learning |
| CN109840587A (en) * | 2019-01-04 | 2019-06-04 | 长江勘测规划设计研究有限责任公司 | Reservoir reservoir inflow prediction technique based on deep learning |
| CN110348622A (en) * | 2019-07-02 | 2019-10-18 | 创新奇智(成都)科技有限公司 | A kind of Time Series Forecasting Methods based on machine learning, system and electronic equipment |
| CN110969282A (en) * | 2019-10-17 | 2020-04-07 | 天津大学 | Runoff stability prediction method based on LSTM composite network |
| CN111507505A (en) * | 2020-03-20 | 2020-08-07 | 苏州丰华声赫智能科技有限公司 | Method for constructing reservoir daily input prediction model |
| CN111598724A (en) * | 2020-05-19 | 2020-08-28 | 四川革什扎水电开发有限责任公司 | Time-interval integration method for day-ahead prediction of warehousing flow of small and medium-sized reservoirs |
| US20200380618A1 (en) * | 2019-05-29 | 2020-12-03 | General Electric Company | Systems and methods for enhanced power system model validation |
| CN112348236A (en) * | 2020-10-23 | 2021-02-09 | 浙江八达电子仪表有限公司 | Abnormal daily load demand prediction system and method for intelligent power consumption monitoring terminal |
| CN112541839A (en) * | 2020-12-23 | 2021-03-23 | 四川大汇大数据服务有限公司 | Reservoir storage flow prediction method based on neural differential equation |
-
2021
- 2021-03-30 CN CN202110341420.6A patent/CN112989705A/en active Pending
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104462861A (en) * | 2014-12-31 | 2015-03-25 | 武汉大学 | Reservoir regulation decision-making method based on reservoir regulation rule synthesis |
| CN105139093A (en) * | 2015-09-07 | 2015-12-09 | 河海大学 | Method for forecasting flood based on Boosting algorithm and support vector machine |
| CN105894138A (en) * | 2016-06-28 | 2016-08-24 | 山东大学 | Optimum weighted composite prediction method for shipment amount of manufacturing industry |
| CN107885951A (en) * | 2017-11-27 | 2018-04-06 | 河海大学 | A kind of Time series hydrological forecasting method based on built-up pattern |
| CN108875161A (en) * | 2018-05-31 | 2018-11-23 | 长江勘测规划设计研究有限责任公司 | Flow grade prediction technique based on convolutional neural networks deep learning |
| CN109840587A (en) * | 2019-01-04 | 2019-06-04 | 长江勘测规划设计研究有限责任公司 | Reservoir reservoir inflow prediction technique based on deep learning |
| US20200380618A1 (en) * | 2019-05-29 | 2020-12-03 | General Electric Company | Systems and methods for enhanced power system model validation |
| CN110348622A (en) * | 2019-07-02 | 2019-10-18 | 创新奇智(成都)科技有限公司 | A kind of Time Series Forecasting Methods based on machine learning, system and electronic equipment |
| CN110969282A (en) * | 2019-10-17 | 2020-04-07 | 天津大学 | Runoff stability prediction method based on LSTM composite network |
| CN111507505A (en) * | 2020-03-20 | 2020-08-07 | 苏州丰华声赫智能科技有限公司 | Method for constructing reservoir daily input prediction model |
| CN111598724A (en) * | 2020-05-19 | 2020-08-28 | 四川革什扎水电开发有限责任公司 | Time-interval integration method for day-ahead prediction of warehousing flow of small and medium-sized reservoirs |
| CN112348236A (en) * | 2020-10-23 | 2021-02-09 | 浙江八达电子仪表有限公司 | Abnormal daily load demand prediction system and method for intelligent power consumption monitoring terminal |
| CN112541839A (en) * | 2020-12-23 | 2021-03-23 | 四川大汇大数据服务有限公司 | Reservoir storage flow prediction method based on neural differential equation |
Non-Patent Citations (2)
| Title |
|---|
| 段召辉, 李承军: "日径流的组合预测模型", 水利水运工程学报, no. 03 * |
| 郭华;陈勇;马耀光;: "组合灰色预测模型在入库流量预测中的应用", 干旱地区农业研究, no. 03 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Jiang et al. | Very short-term wind speed forecasting with Bayesian structural break model | |
| CN113298297B (en) | Wind power output power prediction method based on isolated forest and WGAN network | |
| CN111178585A (en) | Fault reporting amount prediction method based on multi-algorithm model fusion | |
| US20230324860A1 (en) | Estimating Energy Consumption for a Building Using Dilated Convolutional Neural Networks | |
| CN111242351A (en) | Tropical cyclone track prediction method based on self-encoder and GRU neural network | |
| Xiyun et al. | Wind power probability interval prediction based on bootstrap quantile regression method | |
| Hennayake et al. | Machine learning based weather prediction model for short term weather prediction in Sri Lanka | |
| Kosana et al. | Hybrid wind speed prediction framework using data pre-processing strategy based autoencoder network | |
| Li et al. | Numerical weather prediction correction strategy for short-term wind power forecasting based on bidirectional gated recurrent unit and XGBoost | |
| CN113361737A (en) | Abnormity early warning method and system for photovoltaic module | |
| CN114548493A (en) | Method and system for predicting current overload of electric energy meter | |
| Usman et al. | Data analytics for short term price and load forecasting in smart grids using enhanced recurrent neural network | |
| McMillan et al. | Flow forecasting for leakage burst prediction in water distribution systems using long short-term memory neural networks and Kalman filtering | |
| CN111697560B (en) | Method and system for predicting load of power system based on LSTM | |
| Raja et al. | Drought prediction and validation for desert region using machine learning methods | |
| Wei et al. | Uncertainty quantification analysis of wind power: A data-driven monitoring-forecasting framework | |
| CN112989705A (en) | Method and device for predicting reservoir entry flow value, electronic device and medium | |
| Dufek et al. | Data-driven symbolic ensemble models for wind speed forecasting through evolutionary algorithms | |
| Ming-guang et al. | Short-term load combined forecasting method based on BPNN and LS-SVM | |
| CN104462215A (en) | Scientific and technical literature quoting number predicting method based on time sequence | |
| Chinforoush et al. | A novel method for forecasting surface wind speed using wind-direction based on hierarchical markov model | |
| Riquelme-Dominguez et al. | A machine learning-based methodology for short-term kinetic energy forecasting with real-time application: Nordic Power System case | |
| CN113779892A (en) | Wind speed and wind direction prediction method | |
| CN113256018B (en) | Wind power ultra-short term probability prediction method based on conditional quantile regression model | |
| Xiao et al. | Research on fault-environment association rules of distribution network based on improved Apriori algorithm |
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 |