A reservoir water level reverse prediction method based on osmotic pressure data fusion

Abstract

The application discloses a reservoir water level reverse prediction method based on osmotic pressure data fusion, first, the osmotic pressure data and the water level data are preprocessed, data alignment and missing value filling are carried out; then, the correlation between the osmotic pressure gauge and the reservoir water level is calculated, and the fused osmotic pressure data is formed by weighting different osmotic pressure gauge data according to the correlation; finally, the fused data is used to train the gradient boosting regression model to predict the reservoir water level, and the mean square error is used as an evaluation index to measure the performance of the prediction model. The application provides accurate reservoir water level prediction, has strong adaptability and reliability, and is especially suitable for the safety management and disaster prevention and reduction of the reservoir, and provides a scientific basis for the reservoir management.

Description

TECHNICAL FIELD

[0001] The present application relates to reservoir water level prediction and data processing technology, in particular to a reservoir water level regression prediction method based on multi-penetration pressure gauge data fusion. BACKGROUND

[0002] As an important water conservancy infrastructure, reservoirs play a crucial role in flood control, irrigation, water resource allocation, and ecological protection. The safety and stable operation of reservoirs are directly related to social and economic development and the safety of people's lives and property. Therefore, accurate prediction of reservoir water level changes is crucial for reservoir management. Traditionally, the prediction of reservoir water level mainly relies on meteorological data and hydrological models. However, due to the uncertainty of weather prediction and the complexity of hydrological models, the prediction results often have large errors.

[0003] In recent years, as an effective tool for monitoring seepage and water level changes in reservoirs, the penetration pressure gauge has gradually been applied to reservoir water level prediction. The penetration pressure gauge can accurately reflect the seepage state inside the reservoir dam body, thereby providing important early data support for water level changes. However, due to the large number of penetration pressure gauges, high data dimension, and possible noise, how to effectively fuse the data of multiple penetration pressure gauges for accurate prediction has become a key problem in current technology.

[0004] In the prior art, most methods rely on single penetration pressure gauge data or simple multi-point data weighted average. This approach may ignore the spatial correlation and time sequence characteristics between different penetration pressure gauges, thereby affecting the accuracy of water level prediction. To address this issue, the fusion of multiple penetration pressure gauge data and the accurate prediction of water level through regression analysis method have become a hot research topic. To achieve this goal, a series of key technical problems such as interpolation processing, anomaly detection, weighted fusion, and regression modeling of penetration pressure gauge data need to be solved. SUMMARY

[0005] The purpose of the present application is to provide a reservoir water level regression prediction method based on multi-penetration pressure gauge data fusion, which can improve the accuracy and reliability of water level prediction through reasonable weighting mechanism and regression analysis method.

[0006] The technical solution adopted by the present application to solve its technical problems is:

[0007] A reservoir water level inverse prediction method based on penetration pressure data fusion, comprising the following steps:

[0008] Step S1: Read the pressure data of multiple penetration pressure gauges and reservoir water level data, and align the data and fill in the missing values according to time;

[0009] Step S2: Calculate the correlation between each osmometer and the reservoir water level, analyze the correlation degree of pressure data and water level data of different osmometers;

[0010] Step S3: According to the calculated correlation, weight the data of each osmometer to form weighted pressure data;

[0011] Step S4: Fuse the weighted pressure data to generate single fused osmotic pressure data as the input feature of the regression model;

[0012] Step S5: Use gradient boosting regression model to train the model using fused osmotic pressure data to predict the reservoir water level;

[0013] Step S6: Evaluate the model performance, use mean square error as evaluation index, and optimize the model parameters;

[0014] Step S7: Apply the trained model to real-time data prediction to provide accurate reservoir water level warning information.

[0015] Further techniques of the application:

[0016] Preferably, the correlation between each osmometer and the reservoir water level is calculated by using Pearson correlation coefficient, the value range is [-1, 1], and the calculation method is as follows:

[0017]

[0018] Wherein, x i and y i represent the i-th value in the osmotic pressure data and the water level data respectively, and respectively represent the mean value of osmotic pressure data and water level data, and n is the length of the data set.

[0019] Preferably, in step S3, the weight parameter of each osmotic pressure data is calculated and normalized by using the correlation of osmotic pressure data:

[0020]

[0021] Wherein, w i is the weight of osmotic pressure data i, r i is the correlation of osmotic pressure data i and reservoir water level data, the influence of osmometer with high correlation is enhanced by square transformation (|r i | 2 ), and the sum of all weights is ensured to be 1 by normalization operation.

[0022] Preferably, the weighted data fusion formula in step S4 is as follows:

[0023]

[0024] wherein, is the fused osmotic data, P i is the pressure data of the osmometer i.

[0025] Preferably, step S5 specifically:

[0026] S51, the fused pressure data and water level data W are divided into training set and test set, wherein 80% is used for training and 20% is used for testing;

[0027] S52, model training: the gradient boosting regression model updates the model by fitting the residual step by step, and the update formula of each round is:

[0028]

[0029] wherein, is the prediction result of the m-1th round, is the model prediction result of the mth round;

[0030] S53, model prediction: the prediction process is represented as:

[0031]

[0032] wherein, is the predicted water level of the ith sample, is the final prediction result of the ith sample, obtained by weighted sum of M trees, is the initial prediction value, set as the mean value of water level W here, η=0.1 is the learning rate, controlling the contribution of each tree to the final prediction result, and M=100 is the total number of trees.

[0033] The present application has the following beneficial effects:

[0034] Through efficient data preprocessing, feature extraction and weighted fusion, the gradient boosting regression model is used for water level prediction, which significantly improves the accuracy and stability of reservoir water level prediction. By weighting and fusing multiple osmometer data, the reliability of the data is ensured, and the influence of single osmometer failure or deviation on the prediction result is reduced. This method not only accurately captures the complex relationship between osmometer data and reservoir water level changes, but also has strong robustness and adaptability, can handle dynamic changes in different environments, and ensure the stability and real-time performance of the prediction results. It is especially suitable for reservoir water level monitoring, flood warning and disaster prevention and reduction, etc., and provides scientific decision basis for reservoir safety management, and improves the efficiency and safety of water resources scheduling and management. BRIEF DESCRIPTION OF DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments description. Obviously, the drawings described below are only some embodiments of the present application, and all other drawings obtained by those of ordinary skill in the art without creative effort based on these drawings are within the scope of the present application.

[0036] Figure 1 is a flowchart of a reservoir water level reverse prediction method based on osmotic pressure data fusion disclosed by the embodiments of the present application. DETAILED DESCRIPTION

[0037] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the following will further describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application but not all. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of the present application.

[0038] In combination with Figure 1 the present application provides a reservoir water level reverse prediction method based on osmotic pressure data fusion, and specific embodiments include:

[0039] Step S1: reading pressure data of multiple osmometers and reservoir water level data and aligning data and filling missing values according to time.

[0040] S11, extracting pressure data of multiple osmometers related to the reservoir and water level data of the reservoir from the data platform, and ensuring consistent time range.

[0041] Data set = {pressure data 渗压计1 , pressure data 渗压计2 ……, water level data

[0042] Among them, the pressure data represents the measurement value of each osmometer in the time period, and the water level data represents the water level of the reservoir at the corresponding time point.

[0043] S12, aligning the time stamp of each data set to ensure that the pressure data of all osmometers and the water level data of the reservoir have corresponding records at the same time point;

[0044] The following interpolation formula is used to fill in the missing data:

[0045]

[0046] Here P 渗压计j (t k ) is the osmometer data filled after interpolation

[0047] S13, merge the pressure data and water level data of all osmometers according to the timestamp to form a unified data table, each row records multiple osmometer data and water level data at a certain time point. The final data form is as follows:

[0048] ​ P 渗压计1 (t1) … P 渗压计m (t1) [W(t1) <!-- 3 -->]]> [ t2 ] P 渗压计1 (t2) … P 渗压计m (t2) [W(t2)] … … … … … [cat n ]] P 渗压计1 (t n )]]> … P 渗压计m (t n )]]> [W(t n )]]>

[0049] wherein m represents the number of reservoir osmometers, t i represents the timestamp, P 渗压计j (t i ) represents the pressure value of osmometer j at time t i , and W(t i ) represents the water level value of the reservoir at the current time.

[0050] Step S2: calculate the correlation between each osmometer and the reservoir water level, and analyze the correlation degree of the pressure data and water level data of different osmometers.

[0051] S21, calculate the linear correlation between the pressure data and water level data of each osmometer by using the Pearson correlation coefficient, which can be used to measure the linear relationship between two variables, and its value range is [-1, 1], and the calculation method is as follows:

[0052]

[0053] wherein x i and y i represent the i-th value in the osmometer data and water level data, respectively, and represent the mean value of the osmometer data and water level data, respectively, and n is the length of the data set.

[0054] Step S3: weight the data of each osmometer according to the calculated correlation to form weighted pressure data. Calculate the weight parameter of each osmometer data using the correlation of the osmometer data and normalize it:

[0055]

[0056] wherein w i is the weight of osmometer data i, r i is the correlation between osmometer data i and reservoir water level data, the influence of osmometers with high correlation is enhanced through square transformation (|r i | 2 ), and the sum of all weights is ensured to be 1 through normalization operation.

[0057] Step S4: fuse the weighted pressure data to generate single weighted fused osmometer data as the input feature of the regression model, and the weighted data fusion formula is as follows:

[0058]

[0059] wherein, is the fused osmotic data, P i is the pressure data of the osmometer i.

[0060] Step S5: using the gradient boosting regression model, using the weighted osmotic data to train the model to predict the reservoir water level;

[0061] S51, the fused pressure data and water level data W are divided into training set and test set, wherein 80% is used for training and 20% is used for testing.

[0062] S52, model training: the gradient boosting regression model updates the model by fitting the residual step by step, and the update formula of each round is:

[0063]

[0064] wherein, is the prediction result of the m-1th round, is the model prediction result of the mth round.

[0065] S53, model prediction: the gradient boosting regression model used in this method, its prediction process is represented as:

[0066]

[0067] wherein, is the predicted water level of the ith sample (model output), is the final prediction result of the ith sample, which is obtained by the weighted sum of M trees, is the initial prediction value, which is set as the mean of the water level W here, η = 0.1 is the learning rate, which controls the contribution of each tree to the final prediction result, M = 100 is the total number of trees, and max_depth = 3 is set in this method to control the maximum depth of the decision tree. Smaller depth will result in simpler tree model, improve the generalization ability of the model, and avoid overfitting.

[0068] Step S6: evaluate the model performance, use mean square error (MSE) as evaluation index, optimize the model parameters to improve the prediction accuracy. Mean square error (MSE) measures the difference between the model prediction value and the true value, the smaller the value represents the better the prediction effect of the model, and its calculation formula is:

[0069]

[0070] wherein y i is the true value (i.e. water level data), and n is the sample number.

[0071] Step S7: In this embodiment, we apply the trained model to real-time data for reservoir water level prediction. By predicting the real-time data of Qianchengmen Reservoir and Jiudouchuan Reservoir in Quancheng City for half a year, we can accurately predict the change of reservoir water level and provide timely warning information of reservoir water level. After model training and optimization, the mean square error (MSE) on the test set reached 0.00059 and 0.00132, respectively. These results show that this method can accurately predict the change of reservoir water level and effectively reflect the performance of the model in practical application. Through the water level prediction provided by this method, real-time monitoring and early warning of reservoir water level can be realized, further improving the ability of reservoir safety management and disaster prevention and reduction.

[0072] The present application discloses a reservoir water level reverse prediction method based on osmotic pressure data fusion. The method first reads pressure data from multiple osmotic pressure gauges and aligns the data with the reservoir water level data in time and fills in the missing values to ensure the time sequence consistency and integrity of the data. After data processing, the present application calculates the correlation between each osmotic pressure gauge and the reservoir water level data, obtains the weight of different osmotic pressure gauges, and weights the osmotic pressure gauge data according to the correlation coefficient to form the fused pressure data. Then, the gradient boosting regression model is used to train the fused pressure data, and the water level data is used as the target variable to predict the water level. The method uses mean square error (MSE) as the evaluation index, and further improves the prediction accuracy through model optimization to ensure the robustness and accuracy of the model.

[0073] The reservoir water level prediction method of the present application integrates data missing value processing, feature weighting, regression modeling and model optimization, significantly improves the accuracy and stability of reservoir water level prediction, and is especially suitable for reservoir safety monitoring, flood control and drainage and other fields. Through accurate prediction of the reservoir water level, the present application can provide scientific basis for reservoir management and realize real-time warning of the reservoir water level, and enhance the ability of disaster prevention and reduction.

[0074] The above content is only an example and description of the structure of the present application. Those skilled in the art can make various modifications or supplements to the described specific embodiments or use similar ways to replace them, as long as they do not deviate from the structure of the invention or exceed the scope defined by the present claims, which shall belong to the protection scope of the present application.

Claims

1. A reservoir water level reverse prediction method based on osmotic pressure data fusion, characterized in that, Comprising the following steps: Step S1: read the pressure data of multiple osmometers and reservoir water level data, and align the data and fill in the missing values according to time; Step S2: calculate the correlation between each osmometer and the reservoir water level, and analyze the correlation degree of the pressure data and the water level data of different osmometers; Step S3: weight the data of each osmometer according to the calculated correlation to form weighted pressure data; Step S3, specifically, the weight parameters of each osmometer data are calculated and normalized using the correlation of osmometer data: where w i is the weight of the i-th seepage data, m represents the number of reservoir seepage gauges, r i is the correlation of the i-th seepage data and reservoir water level data, the influence of seepage gauges with high correlation is enhanced by squaring transformation (|r i | 2 ), and the sum of all weights is ensured to be 1 by normalization operation; Step S4: fuse the weighted pressure data to generate single fused osmometer data as the input feature of the regression model; Step S4, the weighted data fusion formula is as follows: wherein, is the fused osmotic data, P i is the pressure data of the i-th osmometer; Step S5: use the gradient boosting regression model to train the model using the fused osmometer data to predict the reservoir water level; Step S6: evaluate the model performance, use mean square error as evaluation index, optimize model parameters; Step S7: apply the trained model to real-time data prediction to provide accurate reservoir water level warning information.

2. The reservoir water level inverse prediction method based on osmotic pressure data fusion according to claim 1, characterized in that, By using the Pearson correlation coefficient, the correlation between each osmometer and the reservoir water level is calculated, and the value range is [-1, 1], the calculation method is as follows: where x i and y i denote the i-th value of the osmolarity data and the water level data, respectively, and denote the mean of the osmolarity data and the water level data, respectively, and n is the length of the data set.

3. The reservoir water level inverse prediction method based on osmotic pressure data fusion according to claim 1, characterized in that, Step S5 specifically: S51, dividing the fused pressure data and water level data W into a training set and a test set, with 80% for training and 20% for testing; S52, model training: the gradient boosting regression model updates the model by fitting the residual error step by step, and the update formula of each round is: wherein, is the prediction result of the m-1th round, is the model prediction result of the mth round, η = 0.1 is the learning rate, and controls the contribution of each tree to the final prediction result. S53, model prediction: the prediction process is represented as: where, is the predicted water level for the i-th sample, is the final prediction result for the i-th sample, obtained by the weighted sum of M trees, is the initial prediction value, which is set as the mean of water levels, η = 0.1 is the learning rate, controls the contribution of each tree to the final prediction result, and M = 100 is the total number of trees.

Images