Runoff ensemble forecasting method based on weight real-time correction machine learning model

Abstract

The invention discloses a runoff ensemble forecasting method based on a weight real-time correction machine learning model, and the method comprises the steps: collecting and researching the hydrometeorological data in a drainage basin, dividing the hydrometeorological data into a training set, a verification set and a test set, building a machine learning runoff forecasting set, and carrying out the real-time correction of the weight. Obtaining the dynamic weight of each model in the machine learning runoff forecast set by using a dynamic Bayesian algorithm; then classifying runoff simulation values obtained by the machine learning model based on runoff measured values, constructing a category prediction model, outputting a prediction category, dividing a training set according to the category, extracting a weight sequence, calibrating a weight correction value according to the category so as to adjust the sum of weights, and verifying an effect on a verification set; and finally, dividing the test set based on the test set prediction category, extracting a weight sequence, and generating a final runoff prediction result in combination with the correction weight. The influence of the overall deviation of the model set on final prediction is effectively reduced, so that the basin runoff prediction precision is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of watershed hydrological forecasting, and in particular to a runoff ensemble forecasting method based on a weighted real-time corrected machine learning model. Background Art

[0002] Runoff prediction is a core issue in hydrological research and is of great significance to water resources management, flood prevention and disaster reduction, and ecological protection. However, the runoff process is affected by a variety of natural and human factors, showing significant nonlinearity and complexity, which brings great challenges to the prediction work. Traditional physical mechanism models are limited by complex simulation processes and large computing requirements. Therefore, data-driven machine learning models have become a research hotspot in the field of runoff prediction in recent years.

[0003] In order to improve the accuracy of runoff prediction, ensemble forecasting technology is widely used in the field of hydrology. Ensemble forecasting can effectively reduce the uncertainty of a single model and improve the overall prediction performance by integrating the prediction results of multiple models. Common model ensemble methods include simple arithmetic mean, weighted mean, and dynamic weighted mean based on optimization algorithms. However, traditional methods usually assign weights based on the overall performance of a single model in the historical period, and fail to fully consider the specific performance of the model at different time steps. For example, when all models show systematic overestimation or underestimation at a certain time step, existing methods cannot adjust the weights for this situation, resulting in deviations in the prediction results. Summary of the invention

[0004] The object of the present invention is to provide a runoff ensemble forecasting method based on a real-time weighted correction machine learning model, which can effectively reduce the impact of the overall deviation of the model ensemble on the final runoff prediction effect by classifying the runoff prediction results of all models and adopting corresponding weight correction strategies according to the overall performance of the prediction results of each model in different time periods, thereby improving the accuracy of basin runoff forecasting.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: A runoff ensemble forecasting method based on a weighted real-time modified machine learning model, the steps of which include: The hydrological and meteorological data in the study basin are collected and divided into a training set, a validation set and a test set, and a machine learning runoff prediction model is trained using multiple machine learning models, and a machine learning runoff forecast set is constructed based on all the machine learning runoff prediction models obtained through training; the hydrological and meteorological data include measured runoff and meteorological characteristic data; Use all machine learning runoff forecasting models to predict runoff and obtain corresponding simulated runoff data; According to the simulated runoff data of all machine learning runoff forecast models and their corresponding measured runoff data, the dynamic Bayesian algorithm is used to estimate the weight sequence of each machine learning runoff prediction model in the machine learning runoff forecast model set; Classifying the runoff simulation value predicted by the machine learning runoff prediction model according to the measured runoff data to obtain the category of the runoff simulation value; the category of the runoff simulation value includes category 0, category 1 and category 2; The machine learning runoff forecast set is used to predict the runoff simulation values ​​corresponding to the training set and the validation set. The runoff simulation values ​​and corresponding category labels in the training set and the validation set are used as target variables to train and construct a category prediction model. The constructed category prediction model is used to predict the category labels of the runoff simulation values ​​in the test set. Based on the runoff simulation values ​​in the training set and their corresponding category labels, the training set is divided into three category subsets: category 0 subset, category 1 subset and category 2 subset. The weight sequence corresponding to each category subset extracted according to the time label is corrected, and the weight correction value corresponding to each category subset is verified and determined using the validation set. According to the category label results of the runoff simulation values ​​in the test set predicted by the category prediction model, the test set is divided into three category subsets: category 0 subset, category 1 subset and category 2 subset. The weight sequence of the corresponding category subset extracted in combination with the time label is corrected according to the weight correction value corresponding to each category subset, and the corrected weight sequence of each category subset is used for integrated calculation to generate the final runoff prediction result.

[0006] According to the above technical solution, the categories of runoff simulation values ​​include: Class 0: In a time step, the simulated runoff value is larger or smaller than the measured value; Category 1: at one time step, the runoff simulation values ​​of all models are greater than the measured values; Category 2: At one time step, the runoff simulation values ​​of all models are smaller than the measured values.

[0007] According to the above technical solution, the category prediction model: ; In the formula, Represents the category label of the runoff simulation value predicted by the category prediction model; Represents the training category prediction model, which includes support vector classification SVC, random forest RF, extreme gradient boosting XGBoost, etc. Represents the first machine learning runoff prediction model at time The runoff simulation value of Represents the second machine learning runoff prediction model at time The simulated runoff value, Indicates Machine learning runoff prediction model in time The runoff simulation value.

[0008] According to the above technical solution, the weight sequence correction of the category subset includes: weight correction of category 1 category subset and weight correction of category 2 category subset, while the weight sequence corresponding to category 0 category subset is not corrected; The weight correction formula of the category 1 subset is: ; The 2-category category subset weight correction formula is: ; In the formula, , Indicates the first Machine learning runoff prediction model in time The modified weight sequence of Indicates the first Machine learning runoff prediction model in time The weight sequence of Indicates the weight correction value of category 1 subset; , Indicates the first Machine learning runoff prediction model in time The modified weight sequence of Indicates the first Machine learning runoff prediction model in time The weight sequence of Indicates the weight correction value of 2-category category subset.

[0009] According to the above technical solution, the weight correction value corresponding to the category 1 subset is It is determined using the Bayesian optimization method. The range is between -0.5 and 0; The weight correction values ​​corresponding to the two category subsets Using Bayesian optimization method, The range is between 0 and 0.5.

[0010] According to the above technical solution, the integrated calculation formula of the weight sequence after correction of each category subset includes: ; ; ; In the formula, It represents the weight sequence integration result corresponding to the 0-class category subset. Indicates the corrected weight sequence integration corresponding to the 1-category category subset, Indicates the corrected weight sequence integration corresponding to the 2-category category subset, Represents the weight sequence corresponding to the 0-category subset of the test set, represents the modified weight sequence corresponding to the 1-category category subset, represents the modified weight sequence corresponding to the 2-category category subset, Indicates the first machine learning runoff prediction model in the 0-category category subset at time and the machine learning runoff prediction model at time The simulated runoff value, Indicates the second machine learning runoff prediction model in the 0-class subset at time The simulated runoff value, Indicates the first Machine learning runoff prediction model in time The runoff simulation value of Represents the first machine learning runoff prediction model in the 1-category category subset at time The simulated runoff value, Represents the second machine learning runoff prediction model in the 1st category subset at time The simulated runoff value, Indicates the first Machine learning runoff prediction model in time The runoff simulation value of Represents the first machine learning runoff prediction model in the 2-category category subset at time The simulated runoff value, Represents the second machine learning runoff prediction model in the 2-category category subset at time The simulated runoff value, Indicates the first Machine learning runoff prediction model in time The simulated runoff value, T represents the transposition, represents the 0-class category subset, Represents a subset of 1 category, Represents a 2-class category subset.

[0011] According to the above technical solution, according to the time tag , and Rearrange to get the final runoff prediction result.

[0012] A storage medium for storing computer executable instructions, which, when executed, implement the above-mentioned technical solution, a runoff ensemble forecasting method based on a real-time weighted correction machine learning model.

[0013] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: the traditional multi-model averaging method usually determines the weights based on the overall performance of the runoff simulation of a single model in the historical period and sets the sum of the model weights to 1. However, in the runoff simulation, when the simulation results of all models appear to be higher or lower than the measured values ​​as a whole, the runoff prediction results cannot be corrected by the model averaging method. The present invention classifies the prediction results of all machine learning runoff prediction models so that each type of data can adopt different weight correction strategies according to its characteristics, effectively reducing the poor final runoff prediction results caused by the overall deviation of the machine learning runoff forecast ensemble, thereby improving the accuracy of basin runoff forecast. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 It is a step diagram of a runoff ensemble forecasting method based on a real-time weighted correction machine learning model of the present invention. DETAILED DESCRIPTION

[0015] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0016] The present invention divides the model runoff simulation results into three categories, and then constructs a classification model for the purpose of predicting category labels. According to the runoff simulation values ​​of different categories, appropriate weight correction values ​​are calibrated, and integrated calculations are performed using the corrected weights to generate the final runoff prediction results. The specific steps are as follows: Figure 1 )include: S1. Collect hydrological and meteorological data such as precipitation, temperature, potential evapotranspiration and runoff in the study basin, divide the hydrological and meteorological data into training set, validation set and test set in a ratio of 7:1.5:1.5, and use all machine learning runoff prediction models obtained through training to construct a machine learning runoff forecast set. The specific steps include: S101. Based on the hydrological and meteorological data of the training set, a random forest (RF) model was established to select appropriate prediction factors as the input of the machine learning runoff prediction model.

[0017] S102. Based on the prediction factor data selected in the training set, multiple machine learning models such as long short-term memory network LSTM, gated recurrent unit GRU, support vector machine SVM, etc. are selected for training to obtain multiple corresponding machine learning runoff prediction models, and a machine learning runoff prediction set is constructed based on all the trained machine learning runoff prediction models. The machine learning runoff prediction set puts all the trained machine learning runoff prediction models in one set.

[0018] S103. Based on the test set and the constructed machine learning runoff forecast set, basin runoff simulation is performed, and the Nash Efficiency Coefficient (NSE) is used to evaluate the performance of a single machine learning runoff prediction model.

[0019] S2. Based on the simulated runoff data and measured runoff data of multiple machine learning runoff forecasting models, the dynamic Bayesian algorithm is used to estimate the weight sequence of each machine learning runoff prediction model in the machine learning runoff forecasting model set. Specifically: S201, performing Box-Cox transformation on the simulated runoff data and measured runoff data obtained by each machine learning runoff prediction model to ensure that they satisfy the normal distribution; S202, using the dynamic Bayesian algorithm to estimate the weight sequence of each machine learning runoff prediction model based on the simulated runoff data and the measured runoff data transformed by Box-Cox; S3. Classify the runoff simulation values ​​predicted by the machine learning runoff prediction model in the machine learning runoff forecast set according to the measured runoff data to obtain the categories of the runoff simulation values. Specifically, the categories of the runoff simulation values ​​are divided into three categories: Class 0, that is, in a time step, the simulated runoff value is larger or smaller than the measured value; Category 1, that is, at one time step, the runoff simulation values ​​of all models are greater than the measured values; Category 2, that is, at one time step, the runoff simulation values ​​of all models are smaller than the measured values.

[0020] The formula for classifying runoff simulation values ​​is as follows: ; In the formula, Represents the simulated runoff value at time The category label of Representative Machine learning runoff prediction model in time The runoff simulation value, represent The measured value of runoff over time, Represents the number of machine learning runoff prediction models.

[0021] S4. A category prediction model is constructed based on the runoff simulation values ​​in the training set and the validation set and the corresponding category labels as target variables for training, and the category labels of the runoff simulation values ​​in the test set are predicted using the constructed category prediction model.

[0022] Among them, the training category prediction model: ; In the formula, Represents the category label of the runoff simulation value predicted by the category prediction model; Represents the training category prediction model, which includes support vector classification SVC, random forest RF, extreme gradient boosting XGBoost, etc. Represents the first machine learning runoff prediction model at time The runoff simulation value of Represents the second machine learning runoff prediction model at time The runoff simulation value, Indicates Machine learning runoff prediction model in time The runoff simulation value.

[0023] S5. Based on the runoff simulation values ​​and their category labels in the training set, the training set is divided into three subsets, and the weight sequence corresponding to each subset is extracted according to the time label. The weight correction values ​​are calibrated for different category subsets to adjust the sum of weights, and the effect is verified on the validation set to determine the weight correction values ​​corresponding to each category subset.

[0024] That is, based on the corresponding runoff simulation values ​​and their category labels in the training set, they are divided into three category subsets: 0, 1, and 2, and the corresponding weight sequence is extracted according to the time label, combined with the obtained weight correction value and , calculate the average value of the runoff simulation value of the validation set, and use NSE to evaluate the effect of the correction value. The optimal weight correction value can be determined based on the effect of the correction value. and .

[0025] Among them, each category subset extracts the corresponding weight sequence according to the time label, that is, the weight sequence composed of the weights of each machine learning runoff prediction model in the machine learning runoff prediction model set under the corresponding time label.

[0026] Specifically, the weight corresponding to the 0-class category subset No corrections made.

[0027] For the 1st category subset, the simulated values ​​of all machine learning runoff prediction models are larger than the measured values, so the sum of the weights should be reduced. Therefore, a weight correction value for the 1st category subset is set. , which ranges from -0.5 to 0, and Bayesian optimization is used to find the optimal , for the weight sequence (Indicates the first category in the category subset Machine learning runoff prediction model in time The correction formula is as follows: ; In the formula, , Indicates the first Machine learning runoff prediction model in time The modified weight sequence.

[0028] For the 2-category subset, the predicted values ​​of all machine learning runoff prediction models are smaller than the measured values, so the sum of the weights should be increased. Therefore, a 2-category subset weight correction value is set , which ranges from 0 to 0.5, and uses the Bayesian optimization algorithm to find the optimal , for the weight sequence (Indicates the first Machine learning runoff prediction model in time The weight sequence of the correction is as follows: ; In the formula, , Indicates the first Machine learning runoff prediction model in time The modified weight sequence.

[0029] S6. Use the category prediction model to predict the category label results of the runoff simulation values ​​in the test set, and divide the test set into 0 category subsets based on the category prediction results. , 1 category subset and 2-category subsets The weight sequence corresponding to each category subset is extracted by combining the time label, and the weight is adjusted according to the obtained weight correction value to obtain the corrected weight sequence corresponding to the category 0 category subset, the category 1 category subset and the category 2 category subset. , and The weight sequence of each category after correction during the test period is integrated and calculated, and the calculation formula is as follows: ; ; ; In the formula, It represents the weight sequence integration result corresponding to the 0-class category subset. Indicates the corrected weight sequence integration corresponding to the 1-category category subset, It represents the integration of the modified weight sequence corresponding to the 2-category category subset, and T represents transposition.

[0030] Finally, according to the time label , and Rearrange to get the final runoff prediction result.

[0031] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0032] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A runoff ensemble forecasting method based on a weighted real-time modified machine learning model, characterized in that: The steps include: The hydrological and meteorological data in the study basin are collected and divided into a training set, a validation set and a test set, and a machine learning runoff prediction model is trained using multiple machine learning models, and a machine learning runoff forecast set is constructed based on all the machine learning runoff prediction models obtained through training; the hydrological and meteorological data include measured runoff and meteorological characteristic data; Use all machine learning runoff forecasting models to predict runoff and obtain corresponding simulated runoff data; According to the simulated runoff data of all machine learning runoff forecast models and their corresponding measured runoff data, the dynamic Bayesian algorithm is used to estimate the weight sequence of each machine learning runoff prediction model in the machine learning runoff forecast model set; Classifying the runoff simulation value predicted by the machine learning runoff prediction model according to the measured runoff data to obtain the category of the runoff simulation value; the category of the runoff simulation value includes category 0, category 1 and category 2; The machine learning runoff forecast set is used to predict the runoff simulation values ​​corresponding to the training set and the validation set. The runoff simulation values ​​and corresponding category labels in the training set and the validation set are used as target variables to train and construct a category prediction model. The constructed category prediction model is used to predict the category labels of the runoff simulation values ​​in the test set. Based on the runoff simulation values ​​in the training set and their corresponding category labels, the training set is divided into three category subsets: category 0 subset, category 1 subset and category 2 subset. The weight sequence corresponding to each category subset extracted according to the time label is corrected, and the weight correction value corresponding to each category subset is verified and determined using the validation set. According to the category label results of the runoff simulation values ​​in the test set predicted by the category prediction model, the test set is divided into three category subsets: category 0 subset, category 1 subset and category 2 subset. The weight sequence of the corresponding category subset extracted in combination with the time label is corrected according to the weight correction value corresponding to each category subset, and the corrected weight sequence of each category subset is used for integrated calculation to generate the final runoff prediction result.

2. A runoff ensemble forecasting method based on a weighted real-time corrected machine learning model according to claim 1, characterized in that: The categories of runoff simulation values ​​include: Class 0: In a time step, the simulated runoff value is larger or smaller than the measured value; Category 1: at one time step, the runoff simulation values ​​of all models are greater than the measured values; Category 2: At one time step, the runoff simulation values ​​of all models are smaller than the measured values.

3. The runoff ensemble forecasting method based on the weighted real-time modified machine learning model according to claim 1 is characterized in that: The category prediction model: ； In the formula, Represents the category label of the runoff simulation value predicted by the category prediction model; Represents a training category prediction model; Represents the first machine learning runoff prediction model at time The runoff simulation value of Represents the second machine learning runoff prediction model at time The runoff simulation value, Indicates Machine learning runoff prediction model in time The runoff simulation value.

4. The method for runoff ensemble forecasting based on a weighted real-time corrected machine learning model according to claim 1, characterized in that: The weight sequence correction of the category subsets includes: weight correction of category 1 category subsets and weight correction of category 2 category subsets, while the weight sequence corresponding to category 0 category subsets is not corrected; The weight correction formula of the category 1 subset is: ； The 2-category category subset weight correction formula is: ； In the formula, , Indicates the first Machine learning runoff prediction model in time The modified weight sequence of Indicates the first Machine learning runoff prediction model in time The weight sequence of Indicates the weight correction value of category 1 subset; , Indicates the first Machine learning runoff prediction model in time The modified weight sequence of Indicates the first Machine learning runoff prediction model in time The weight sequence of Indicates the weight correction value of 2-category category subset.

5. The method for runoff ensemble forecasting based on a weighted real-time corrected machine learning model according to claim 4, characterized in that: The weight correction value corresponding to the category 1 subset It is determined using the Bayesian optimization method. The range is between -0.5 and 0; The weight correction values ​​corresponding to the two category subsets It is determined using the Bayesian optimization method. The range is between 0 and 0.

5.

6. The method for runoff ensemble forecasting based on a weighted real-time corrected machine learning model according to claim 1, characterized in that: The integrated calculation formula of the weight sequence after correction of each category subset includes: ； ； ； In the formula, It represents the weight sequence integration result corresponding to the 0-class category subset. Indicates the corrected weight sequence integration corresponding to the 1-category category subset, It indicates the corrected weight sequence integration corresponding to the 2-category category subsets, Represents the weight sequence corresponding to the 0-category subset of the test set, represents the modified weight sequence corresponding to the 1-category category subset, represents the modified weight sequence corresponding to the 2-category category subset, Indicates the first machine learning runoff prediction model in the 0-class subset at time The simulated runoff value, Indicates the second machine learning runoff prediction model in the 0-class subset at time The simulated runoff value, Indicates the first Machine learning runoff prediction model in time The runoff simulation value of Represents the first machine learning runoff prediction model in the 1-category category subset at time The simulated runoff value, Represents the second machine learning runoff prediction model in the 1st category subset at time The simulated runoff value, Indicates the first Machine learning runoff prediction model in time The runoff simulation value of Represents the first machine learning runoff prediction model in the 2-category category subset at time The simulated runoff value, Represents the second machine learning runoff prediction model in the 2-category category subset at time The simulated runoff value, Indicates the first Machine learning runoff prediction model in time The simulated runoff value, T represents the transposition, represents the 0-category subset, Represents a subset of 1 category, Represents a 2-class category subset.

7. The method for runoff ensemble forecasting based on a weighted real-time corrected machine learning model according to claim 6, characterized in that: According to the time label , and Rearrange to get the final runoff prediction result.

8. A storage medium for storing computer executable instructions, characterized in that: When executed, the computer executable instructions implement a runoff ensemble forecasting method based on a weighted real-time corrected machine learning model as described in any one of claims 1 to 7.

Images