A runoff prediction method and device based on data statistical analysis

Abstract

The application relates to the technical field of hydraulic engineering, and discloses a runoff prediction method and device based on data statistical analysis, which comprises the following steps: determining a periodic time variable of a runoff event occurrence time based on a runoff event period and a preset reference time; performing regression training on historical runoff data and the periodic time variable corresponding to the runoff event occurrence time to obtain a probability distribution of a relative deviation; randomly extracting a relative deviation of a to-be-predicted time based on the probability distribution of the relative deviation; determining runoff average trend data of the to-be-predicted time based on the reference time and the runoff event period; and determining runoff data of the to-be-predicted time based on the runoff average trend data of the to-be-predicted time and the relative deviation of the to-be-predicted time. The application realizes the acquisition of runoff prediction of a hydrological station with long-series historical runoff data, effectively reduces the difficulty of runoff prediction data acquisition and transmission, simplifies a runoff prediction process, and improves runoff prediction efficiency.

Description

Technical Field

[0001] This invention relates to the field of water conservancy engineering technology, specifically to a method and apparatus for runoff forecasting based on statistical data analysis. Background Technology

[0002] Runoff refers to the flow of water, such as rainfall, snowmelt, or irrigation water, along the surface or underground under the influence of gravity. Runoff forecasting plays an important role in the planning, design, construction, and operation management of water conservancy projects, and is also of great significance for flood control and disaster reduction, rational allocation of water resources, and ecological environmental protection.

[0003] Related runoff prediction methods rely on short-term rainfall or runoff data for prediction. However, short-term rainfall or runoff data is difficult to obtain, resulting in redundant runoff prediction processes and low prediction efficiency. Summary of the Invention

[0004] In view of this, the present invention provides a runoff forecasting method based on statistical analysis of data to solve the problem that existing forecasting methods are difficult to obtain rainfall or runoff data, resulting in redundant runoff forecasting processes and low forecasting efficiency.

[0005] In a first aspect, the present invention provides a runoff forecasting method based on statistical data analysis, the method comprising:

[0006] Historical runoff data and runoff event cycles of the hydrological stations to be forecasted are obtained, and the periodic time variables of the time of runoff event occurrence are determined based on the runoff event cycle and the preset reference time.

[0007] Regression training was performed on historical runoff data and the periodic time variables corresponding to the time of runoff events to obtain the probability distribution of the relative deviation;

[0008] The relative deviation of the time to be predicted is randomly selected based on the probability distribution of the relative deviation.

[0009] The average trend data of runoff at the time to be predicted are determined based on the baseline time and the runoff event cycle;

[0010] The runoff data for the forecast time is determined based on the average trend data of runoff at the forecast time and the relative deviation at the forecast time.

[0011] The runoff forecasting method based on statistical data analysis provided in this embodiment acquires historical runoff data and runoff event cycles of the hydrological station to be forecasted. Based on the runoff event cycle and a preset reference time, it determines the periodic time variable of the runoff event occurrence time. Then, based on the periodic time variable corresponding to the historical runoff data and the runoff event occurrence time, it determines the relative deviation of the forecast time. Based on the reference time and runoff event cycle, it determines the average runoff trend data for the forecast time. Finally, based on the average runoff trend data and the relative deviation of the forecast time, it determines the runoff data for the forecast time. In the process of forecasting runoff at future times for the hydrological station, it reduces reliance on short-term rainfall data of the watershed or short-term runoff data of upstream stations, enabling the acquisition of runoff forecasts for hydrological stations with long-term historical runoff data. This effectively reduces the difficulty of acquiring and transmitting runoff forecast data, while also reducing reliance on real-time runoff data, simplifying the runoff forecasting process and improving runoff forecasting efficiency.

[0012] In one optional implementation, regression training is performed on historical runoff data and periodic time variables corresponding to the time of runoff events to obtain a probability distribution of the relative deviation, including:

[0013] Obtain the functional relationship between the average runoff trend and periodic time variables. Based on the periodic time variables corresponding to the time of runoff events, use the functional relationship between the average runoff trend and periodic time variables to determine the average runoff trend prediction data.

[0014] The relative deviation is determined based on average runoff trend forecast data and historical runoff data;

[0015] The standard deviation of the relative deviation is calculated based on the relative deviation, and the probability distribution of the relative deviation is obtained based on the standard deviation of the relative deviation.

[0016] The runoff forecasting method based on statistical analysis provided in this embodiment determines the average runoff trend forecast data by utilizing the functional relationship between the average runoff trend and periodic time variables. Based on the average runoff trend forecast data and historical runoff data, the relative deviation is determined, realizing the accurate calculation of the relative deviation between the average runoff trend forecast data and historical runoff data. The probability distribution of the relative deviation is obtained through the standard deviation of the relative deviation. The influence of random errors on the forecast results is considered in the runoff forecasting process, making the runoff forecast results more accurate.

[0017] In an optional implementation, before obtaining the functional relationship between the runoff average trend and the periodic time variable, and determining the runoff average trend prediction data based on the periodic time variable corresponding to the time of the runoff event using the functional relationship between the runoff average trend and the periodic time variable, the method further includes:

[0018] The amount of historical runoff data is obtained, and the number of decision trees and the search range of pre-pruning parameters are determined based on the amount of historical runoff data, and a parameter grid is established.

[0019] Set cross-validation parameters and evaluation metrics, and determine the optimal number of decision trees and pre-pruning parameters based on the parameter grid, cross-validation folds, and evaluation metrics;

[0020] Based on the optimal number of decision trees and pre-pruning parameters, the random forest algorithm is used to obtain the functional relationship between the average trend of runoff and the periodic time variable.

[0021] The runoff forecasting method based on statistical data analysis provided in this embodiment determines the optimal number of decision trees and pre-pruning parameters through parameter grids, cross-validation folds, and evaluation indicators. This effectively avoids the contradiction between runoff forecasting accuracy and computational cost when the amount of historical runoff data is large. It achieves the selection of parameters for the random forest algorithm, and then, based on the optimal number of decision trees and pre-pruning parameters, uses the random forest algorithm to obtain the functional relationship between the runoff average trend and periodic time variables. This achieves an accurate description of the functional relationship between the runoff average trend and periodic time variables, simplifying the runoff forecasting process.

[0022] In one optional implementation, the average trend data of runoff at the time to be predicted is determined based on a reference time and the runoff event cycle, including:

[0023] The periodic time variables for the time to be predicted are calculated based on the reference time and the runoff event cycle;

[0024] The average trend data of runoff at the time to be predicted are determined based on the periodic time variables at the time to be predicted.

[0025] The runoff forecasting method based on statistical data analysis provided in this embodiment calculates the periodic time variables and the average runoff trend data at the time to be forecasted, thereby accurately calculating the average trend of runoff changes and laying the foundation for subsequent runoff forecasting.

[0026] In one optional implementation, the periodic time variable of the runoff event occurrence time is determined based on the runoff event cycle and a preset reference time. The calculation formula for the periodic time variable of the runoff event occurrence time is as follows:

[0027]

[0028] Where, τ i The periodic time variable representing the time of occurrence of a runoff event, where frac represents the fractional part, T represents the period of the runoff event, t0 represents the reference time, and t i Indicates the time when the runoff event occurred.

[0029] In one optional implementation, the runoff data for the forecast time is determined based on the average runoff trend data for the forecast time and the relative deviation of the forecast time. The calculation formula for the runoff data for the forecast time is as follows:

[0030]

[0031] Among them, Q j This represents runoff data for the time to be predicted. r represents the average trend data of runoff at the time to be predicted. j This indicates the relative deviation of the time to be predicted.

[0032] Secondly, the present invention provides a runoff forecasting device based on statistical data analysis, the device comprising:

[0033] The first determining module is used to acquire historical runoff data and runoff event cycles of the hydrological stations to be forecasted, and to determine the periodic time variable of the time of runoff event occurrence based on the runoff event cycle and the preset reference time.

[0034] The regression training module is used to perform regression training on historical runoff data and the periodic time variables corresponding to the time of runoff events to obtain the probability distribution of the relative deviation.

[0035] The extraction module is used to randomly extract the relative deviation of the time to be predicted based on the probability distribution of the relative deviation;

[0036] The second determination module is used to determine the average trend data of runoff at the time to be predicted based on the reference time and the runoff event cycle;

[0037] The third determination module is used to determine the runoff data for the forecast time based on the average runoff trend data and the relative deviation of the forecast time.

[0038] Thirdly, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the runoff forecasting method based on data statistical analysis described in the first aspect or any corresponding embodiment.

[0039] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to execute the runoff forecasting method based on data statistical analysis described in the first aspect or any corresponding embodiment thereof.

[0040] Fifthly, the present invention provides a computer program product, including computer instructions for causing a computer to execute the runoff forecasting method based on data statistical analysis described in the first aspect or any corresponding embodiment. Attached Figure Description

[0041] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0042] Figure 1 This is a flowchart illustrating a runoff forecasting method based on statistical data analysis according to an embodiment of the present invention.

[0043] Figure 2 This is a flowchart illustrating another runoff forecasting method based on statistical data analysis according to an embodiment of the present invention;

[0044] Figure 3 This is a flowchart illustrating another runoff forecasting method based on data statistical analysis according to an embodiment of the present invention;

[0045] Figure 4 This is a flowchart illustrating another runoff forecasting method based on data statistical analysis according to an embodiment of the present invention;

[0046] Figure 5 This is a schematic diagram of the daily measured runoff machine learning results from 2003 to 2020 at hydrological station A according to an embodiment of the present invention;

[0047] Figure 6 This is a probability (frequency) distribution diagram of the deviation of measured runoff at hydrological station A relative to its average trend according to an embodiment of the present invention.

[0048] Figure 7 The daily runoff forecast results for hydrological station A in 2021 are based on an embodiment of the present invention.

[0049] Figure 8 This is a structural block diagram of a runoff forecasting device based on data statistical analysis according to an embodiment of the present invention;

[0050] Figure 9 This is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0052] Relevant runoff prediction methods include:

[0053] The first method is based on a rainfall-runoff hydrological model. It uses rainfall, evaporation and infiltration data of the watershed within the forecast period to calculate runoff generation and confluence through the hydrological model, thereby realizing runoff forecasting at the stations. This method has a clear concept, but the forecast period is short and it cannot obtain and transmit relevant data with a longer forecast period and higher accuracy in a timely manner, making it difficult to improve the accuracy of runoff forecasts for more than 3 days in the future.

[0054] The second method is the rainfall-runoff regression method, which forecasts by establishing a regression relationship between rainfall and runoff. It omits the runoff calculation process in the first method and relies only on rainfall data within the forecast period, but it is difficult to obtain and transmit relevant data with a long forecast period and high accuracy in a timely manner.

[0055] The third method is the corresponding flow method. This method is based on real-time measured runoff at upstream stations and calculates the runoff at downstream stations to be predicted by combining water surface wave propagation. It requires that the distance between upstream and downstream stations be short and that the water inflow between upstream and downstream can be negligible. Although it does not rely on rainfall data, it is difficult to obtain runoff data from upstream stations in a short period of time, resulting in low accuracy of runoff forecast.

[0056] In long-term water conservancy practice, hydrological stations across various regions have accumulated a considerable amount of historical runoff data; however, this historical runoff data has not been fully utilized in runoff forecasting methods. If in-depth analysis can be conducted based on the historical runoff data of local hydrological stations, and new forecasting methods can be proposed accordingly, it is hoped that the difficulties currently faced by runoff forecasting methods in acquiring and transmitting basic data can be resolved, the reliance on real-time data can be reduced, the forecasting process can be simplified, and the efficiency of runoff forecasting can be improved.

[0057] To address the aforementioned technical problems, this invention provides a runoff forecasting method based on statistical data analysis. This method overcomes the shortcomings of other runoff forecasting methods that rely on short-term rainfall or runoff data. Given the long-term historical runoff data of the hydrological station to be forecasted and the timing of runoff events, it effectively forecasts the future runoff of the hydrological station. The runoff forecasting process not only incorporates the average trend of runoff changes but also random errors.

[0058] This invention provides a runoff forecasting method based on statistical data analysis. It should be noted that the execution entity of this method can be a device for runoff forecasting based on statistical data analysis. This runoff forecasting can be implemented as part or all of an electronic device through software, hardware, or a combination of both. The electronic device can be a server or a terminal. In this embodiment, the server can be a single server or a server cluster composed of multiple servers. The terminal can be a smartphone, personal computer, tablet computer, or other intelligent hardware device such as a smart robot. The following method embodiments all use an electronic device as the execution entity for illustration.

[0059] According to an embodiment of the present invention, a method for runoff forecasting based on statistical data analysis is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0060] This embodiment provides a runoff forecasting method based on statistical data analysis, which can be used in the aforementioned electronic equipment. Figure 1 This is a flowchart of a runoff forecasting method based on data statistical analysis according to an embodiment of the present invention, such as... Figure 1 As shown, the process includes the following steps:

[0061] Step S101: Obtain historical runoff data and runoff event cycles of the hydrological stations to be forecasted, and determine the periodic time variable of the time of runoff event occurrence based on the runoff event cycle and the preset reference time.

[0062] Specifically, most hydrological stations exhibit an annual cycle. For these stations, the runoff event cycle is "annual." When hydrological data is statistically analyzed according to the natural year, since the number of days in a normal year and a leap year are 365 and 366 days respectively, if the runoff event cycle is "annual," it is necessary to address the inconsistency in the number of days in normal and leap years. Furthermore, considering the use of long-series runoff data (i.e., historical runoff data from the hydrological stations to be predicted) as the basic data, which typically includes data from several normal and leap years periodically, unifying the processing of normal and leap year data can not only simplify data processing but also help to bridge errors.

[0063] Furthermore, since leap years occur every 4 years, the formula for calculating the runoff event cycle T (in days) is as follows:

[0064] T = (365 * 3 + 366) / 4 = 365.25 (1)

[0065] Therefore, the period T of the runoff event is determined to be 365.25 days.

[0066] Furthermore, with reference time t0 selected as the quantification benchmark, the calculation formula for the periodic time variable of the runoff event occurrence time for each data point in the historical runoff data of the hydrological station to be predicted is as follows:

[0067]

[0068] Where, τ i Indicates the time t when the runoff event occurs. i The periodic time variable, frac represents the fractional part, T represents the runoff event period, t0 represents the reference time, t i Indicates the time when the runoff event occurred.

[0069] Furthermore, the above formula (1) quantifies the historical runoff data Q. i (i = 1, 2, ..., m) and the time of occurrence t i The periodic relative relationship between them, the reference time t0 only provides a quantitative reference, and the historical runoff data Q i With the time of occurrence t i The periodic relative relationship between them does not have an impact; therefore, the reference time t0 can be arbitrarily selected.

[0070] Step S102: Perform regression training on the historical runoff data and the periodic time variables corresponding to the time of runoff events to obtain the probability distribution of the relative deviation.

[0071] Step S103: Randomly select the relative deviation of the time to be predicted based on the probability distribution of the relative deviation.

[0072] Specifically, based on pseudo-random number generation, random sampling can be performed using the Ziggurat algorithm (a method for generating random numbers) based on the accept-rejection strategy, or the Box-Muller algorithm (a method for generating standard normally distributed random numbers) based on the inverse transform, or the Marsaglia polar coordinate algorithm (a method for generating random numbers with a specific probability distribution) based on both the inverse transform and the accept-rejection strategy, and a modified Box-Muller algorithm, to obtain the relative deviation of the time to be predicted.

[0073] Furthermore, since the Permuted Congruential Generator (PCG) series of algorithms has shown advantages over other algorithms in terms of simplicity, speed, storage saving, good statistical effect and unpredictability, the Permuted Congruential Generator series of algorithms is used to generate pseudo-random numbers.

[0074] Step S104: Determine the average trend data of runoff at the time to be predicted based on the reference time and the runoff event cycle.

[0075] Step S105: Determine the runoff data for the forecast time based on the average runoff trend data for the forecast time and the relative deviation of the forecast time.

[0076] Specifically, the formula for calculating runoff data at the time to be predicted is as follows:

[0077]

[0078] Among them, Q j This represents runoff data for the time to be predicted. r represents the average trend data of runoff at the time to be predicted. j This indicates the relative deviation of the time to be predicted.

[0079] The runoff forecasting method based on statistical data analysis provided in this embodiment acquires historical runoff data and runoff event cycles of the hydrological station to be forecasted. Based on the runoff event cycle and a preset reference time, it determines the periodic time variable of the runoff event occurrence time. Then, based on the periodic time variable corresponding to the historical runoff data and the runoff event occurrence time, it determines the relative deviation of the forecast time. Based on the reference time and runoff event cycle, it determines the average runoff trend data for the forecast time. Finally, based on the average runoff trend data and the relative deviation of the forecast time, it determines the runoff data for the forecast time. In the process of forecasting future runoff at the hydrological station, this method reduces reliance on short-term rainfall data of the watershed or short-term runoff data of upstream stations. It enables the acquisition of runoff forecasts for hydrological stations with long-term historical runoff data, effectively reducing the difficulty of acquiring and transmitting runoff forecast data. Simultaneously, it reduces reliance on real-time runoff data, simplifies the runoff forecasting process, and improves runoff forecasting efficiency.

[0080] This embodiment provides a runoff forecasting method based on statistical data analysis, which can be used in the aforementioned electronic equipment. Figure 2 This is a flowchart of a runoff forecasting method based on data statistical analysis according to an embodiment of the present invention, such as... Figure 2 As shown, the process includes the following steps:

[0081] Step S201: Obtain historical runoff data and runoff event cycles for the hydrological stations to be forecasted, and determine the periodic time variable of the runoff event occurrence time based on the runoff event cycle and a preset reference time. For details, please refer to [link to relevant documentation]. Figure 1 Step S101 of the illustrated embodiment will not be described again here.

[0082] Step S202: Perform regression training on the historical runoff data and the periodic time variables corresponding to the time of runoff events to obtain the probability distribution of the relative deviation.

[0083] Specifically, step S202 includes:

[0084] Step S2021: Obtain the functional relationship between the average runoff trend and the periodic time variable. Based on the periodic time variable corresponding to the time of runoff event, determine the average runoff trend prediction data using the functional relationship between the average runoff trend and the periodic time variable.

[0085] Specifically, due to the complexity of runoff events, the average trend of runoff is not a simple periodic functional relationship, and therefore cannot be represented by a known function or a combination thereof. The continuous development and widespread application of machine learning technology provides new means to solve this problem. Machine learning is used to obtain the functional relationship between the average runoff trend and periodic time variables. Among machine learning methods, the random forest algorithm has advantages such as being unaffected by the actual function that the data should satisfy, strong generalization ability, and low difficulty in parameter tuning, making it suitable for application under various hydrological conditions. Therefore, the random forest algorithm is used to obtain the functional relationship between the average runoff trend and periodic time variables.

[0086] Furthermore, the random forest algorithm contains parameters such as the number of decision trees, maximum tree depth, minimum tree split value, and minimum number of samples per leaf node (the latter three are pre-pruning parameters, and usually only one of them needs to be controlled). These parameters are generally selected based on a balance between forecast accuracy and computational cost requirements. However, for cases with a large volume of historical runoff data, the contradiction between forecast accuracy and computational cost requirements may be more prominent, making it necessary to select the optimal parameters within the allowable range. Therefore, forecast accuracy and computational cost requirements can be used as evaluation factors, and a grid search method combined with cross-validation can be used to optimize the relevant parameters of the random forest algorithm.

[0087] Furthermore, the process of determining the functional relationship between the average runoff trend and the periodic time variable is as follows: Obtain the amount of historical runoff data; based on the amount of historical runoff data, determine the number of decision trees and the search range of pre-pruning parameters, and establish a parameter grid; set cross-validation parameters and evaluation indicators; based on the parameter grid, cross-validation folds, and evaluation indicators, determine the optimal number of decision trees and pre-pruning parameters; based on the optimal number of decision trees and pre-pruning parameters, use the random forest algorithm to obtain the functional relationship between the average runoff trend and the periodic time variable.

[0088] Furthermore, the specific steps for determining the optimal number of decision trees and pre-pruning parameters are as follows: define the search range of parameters and establish a parameter grid; set the number of folds K for cross-validation (5 folds or 10 folds); select evaluation metrics (such as prediction accuracy, computational cost, or a combination of both); perform grid search; sort the grid search results and find the parameter combination with the best average evaluation metric; select the parameter combination with the optimal evaluation metric from the sorted results as the optimal parameters, which include the optimal number of decision trees and pre-pruning parameters; the specific steps of the grid search are as follows:

[0089] a) For each set of parameters in the parameter grid: For each fold in the K-fold, train the random forest model using the K-1 fold data, validate the model on the remaining 1 fold and calculate the evaluation metric; calculate the average of the K evaluation metrics.

[0090] b) Record each parameter combination and its corresponding average evaluation index.

[0091] Furthermore, machine learning was used to perform regression training on the periodic time variable series at the time of runoff events and historical runoff data series to obtain the functional relationship between the runoff mean trend and the periodic time variable, and the runoff mean trend. The function relating to the periodic time variable τ is shown below:

[0092]

[0093] Step S2022: Determine the relative deviation based on the average runoff trend prediction data and historical runoff data.

[0094] Specifically, each periodic time variable τ i Substituting into formula (4) above, we obtain the average runoff trend prediction data. Furthermore, each historical runoff data Q i With corresponding average runoff trend forecast data The relative deviation r between i The specific formula is shown below;

[0095]

[0096] Step S2023: Calculate the standard deviation of the relative deviation based on the relative deviation, and obtain the probability distribution of the relative deviation based on the standard deviation of the relative deviation.

[0097] Specifically, the relative deviation r i Consider it as a random sample of the population relative deviation r, assuming that the population relative deviation r follows a normal distribution with a mean of zero. i Estimate the population standard deviation σ of variable r. The formula for calculating the standard deviation σ is as follows:

[0098]

[0099] Where m is the number of runoff data points.

[0100] Furthermore, the probability distribution of the relative deviation is determined by the standard deviation of the relative deviation. The formula for calculating the probability distribution of the relative deviation is as follows:

[0101]

[0102] Where ξ represents the threshold of relative deviation, P r This represents the probability that the relative deviation is less than the threshold ξ, where x is the integral variable.

[0103] Step S203: Randomly select the relative deviation of the time to be predicted based on the probability distribution of the relative deviation. For details, please refer to [link to relevant documentation]. Figure 1 Step S103 of the illustrated embodiment will not be described again here.

[0104] Step S204: Determine the average runoff trend data for the time to be predicted based on the baseline time and runoff event cycle. For details, please refer to [link to relevant documentation]. Figure 1 Step S104 of the illustrated embodiment will not be described again here.

[0105] Step S205: Determine the runoff data for the forecast time based on the average runoff trend data and the relative deviation between the forecast time and the forecast time. For details, please refer to [link to relevant documentation]. Figure 1 Step S105 of the illustrated embodiment will not be described again here.

[0106] The runoff forecasting method based on statistical analysis provided in this embodiment determines the average runoff trend forecast data by utilizing the functional relationship between the average runoff trend and periodic time variables. Based on the average runoff trend forecast data and historical runoff data, the relative deviation is determined, realizing the accurate calculation of the relative deviation between the average runoff trend forecast data and historical runoff data. The probability distribution of the relative deviation is obtained through the standard deviation of the relative deviation. The influence of random errors on the forecast results is considered in the runoff forecasting process, making the runoff forecast results more accurate.

[0107] This embodiment provides a runoff forecasting method based on statistical data analysis, which can be used in the aforementioned electronic equipment. Figure 3 This is a flowchart of a runoff forecasting method based on data statistical analysis according to an embodiment of the present invention, such as... Figure 3 As shown, the process includes the following steps:

[0108] Step S301: Obtain historical runoff data and runoff event cycles for the hydrological stations to be forecasted, and determine the periodic time variable of the runoff event occurrence time based on the runoff event cycle and a preset reference time. For details, please refer to [link to relevant documentation]. Figure 2 Step S201 of the illustrated embodiment will not be described again here.

[0109] Step S302 involves performing regression training on the historical runoff data and the periodic time variables corresponding to the time of runoff events to obtain the probability distribution of the relative deviation. For details, please refer to [link to relevant documentation]. Figure 2 Step S202 of the illustrated embodiment will not be described again here.

[0110] Step S303: Randomly select the relative deviation of the time to be predicted based on the probability distribution of the relative deviation. For details, please refer to [link to relevant documentation]. Figure 2 Step S203 of the illustrated embodiment will not be described again here.

[0111] Step S304: Determine the average trend data of runoff at the time to be predicted based on the reference time and the runoff event cycle.

[0112] Specifically, step S304 includes:

[0113] Step S3041: Calculate the periodic time variable of the time to be predicted based on the reference time and the runoff event cycle.

[0114] Specifically, the time to be predicted t j Substituting (j = 1, 2, ..., n; n is the number of times to be predicted) into the above formula (2), we obtain the time to be predicted t. j The corresponding periodic time variable τ j .

[0115] Step S3042: Determine the average trend data of runoff at the time to be predicted based on the periodic time variable of the time to be predicted.

[0116] Specifically, the time to be predicted t j The corresponding periodic time variable τ j Substituting into formula (4) above, we obtain the time τ to be predicted. j Average runoff trend data

[0117] Step S305: Determine the runoff data for the forecast time based on the average runoff trend data and the relative deviation of the forecast time; for details, please refer to [link to relevant documentation]. Figure 2 Step S205 of the illustrated embodiment will not be described again here.

[0118] The runoff forecasting method based on statistical data analysis provided in this embodiment calculates the periodic time variables and the average runoff trend data at the time to be forecasted, thereby accurately calculating the average trend of runoff changes and laying the foundation for subsequent runoff forecasting.

[0119] The following specific embodiment illustrates the steps of a runoff forecasting method based on statistical data analysis.

[0120] Example 1:

[0121] The daily runoff forecast for 2021 is selected from hydrological station A in a certain area. The historical runoff data used in the forecast are the daily measured runoff data of this station from 2003 to 2020. Figure 4 As shown, the specific steps of the runoff forecasting method based on statistical data analysis are as follows:

[0122] Step 1: Determine the runoff event cycle of hydrological station A as "year". Since the period from 2003 to 2020 includes both common years and leap years, the two are treated the same, and the "year" is calculated as 365.25 days.

[0123] Step 2: Select 00:00:00:00 on January 1, 2001 as the reference time t0, and for each historical runoff data Q of the daily measured runoff data of hydrological station A from 2003 to 2020... i To simplify the calculation, we assume that the historical runoff data occurred at time t. i All times are 12:00:00:00 on the same day, and then t is calculated. i The corresponding periodic time variable τ i As shown in Table 1 below:

[0124] Table 1:

[0125] i <![CDATA[Q i ]]> <![CDATA[t i ]]> <![CDATA[τ i ]]> 1 3050 2003-01-01 12:00:00 0.000000000 2 3040 2003-01-02 12:00:00 0.002737851 3 2810 2003-01-03 12:00:00 0.005475702 … … … … 6575 4980 2020-12-31 12:00:00 0.998631075

[0126] Step 3: Using the periodic time variable τ in Table 1 above... i The sequence is used as a training feature, and the historical runoff data Q i The sequence is used as the training target, and the random forest algorithm is employed to train τ. i Sequence and Q i Machine learning regression training was performed on the sequence, with 100 decision trees, maximum tree depth set to fully expanded, minimum tree split value of 2, and minimum leaf node sample size of 1, to obtain the average trend of runoff. The functional relationship between the variable and the periodic time variable τ. For ease of presentation and comparison, this relationship is transformed into... The functional relationship between the actual time t and the actual time t, such as Figure 5 As shown by the solid line, Figure 5 The hollow dots in the diagram represent measured runoff data Q. i The situation.

[0127] Step 4: Based on each τ in Table 1 above i get Series, daily relative deviation of runoff r from 2003 to 2020 i The calculation results are shown in Table 2 below:

[0128] Table 2:

[0129]

[0130] Step 5: Calculate each Q i With the corresponding The relative deviation r between i , get r i Series, daily relative deviation of runoff r from 2003 to 2020 i The calculation results are shown in Table 3 below:

[0131] Table 3:

[0132] i <![CDATA[r i ]]> 1 -0.1874 2 -0.1614 3 -0.2221 … … 6575 0.2272

[0133] Step 6: Place r i The sequence is considered a random sample of the population of variable r, which is assumed to follow a normal distribution with a mean of zero. i The standard deviation σ of the population of variable r is estimated to be 0.225 according to equation (4), thus determining the probability (frequency) distribution of the population of variable r. Figure 6 As shown by the curve in the middle. Figure 6 r was also displayed i The probability (frequency) histogram of the sequence as a random sample of the population of variable r basically conforms to a normal distribution with a mean of zero.

[0134] Step 7: Forecast the daily runoff at hydrological station A in 2021. For simplicity, the forecast time will still be set to 12:00:00:00 noon on the current day based on historical data. Calculate the forecast time t. j The corresponding periodic time variable τ j The periodic time variable of daily runoff in 2021, τ j The calculation results are shown in Table 4 below:

[0135] Table 4:

[0136] j <![CDATA[t j ]]> <![CDATA[τ j ]]> 1 2021-01-01 12:00:00 0.001368925 2 2021-01-02 12:00:00 0.004106776 3 2021-01-03 12:00:00 0.006844627 … … … 365 2021-12-31 12:00:00 0.997946612

[0137] Step 8: Calculate τ j Corresponding Daily runoff average trend in 2021 The forecast results are shown in Table 5 below:

[0138] Table 5:

[0139]

[0140] Step 9: Based on the probability (frequency) distribution, conduct n (=365) random samplings using the Ziggurat algorithm. The pseudo-random number generation uses the PCG-64 algorithm from the PCG series, to obtain r. j The daily relative deviation of runoff in 2021, r j The forecast results are shown in Table 6 below:

[0141] Table 6:

[0142] j <![CDATA[r j ]]> 1 0.0548 2 0.2552 3 -0.2101 … … 365 0.1412

[0143] Step 10: Predict time t j runoff Q j Daily runoff Q in 2021 j The forecast results are shown in Table 7 below:

[0144] Table 7:

[0145] j <![CDATA[Q j ]]> 1 3892.5 2 4615.9 3 2825.4 … … 365 4437.1

[0146] Figure 7 This is a comparison between the daily runoff of hydrological station A in 2021, which was predicted according to the above steps, and the actual measured daily runoff.

[0147] This embodiment also provides a runoff forecasting method apparatus based on data statistical analysis. This apparatus is used to implement the above embodiments and preferred embodiments, and details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0148] This embodiment provides a method and apparatus for runoff forecasting based on statistical data analysis, such as... Figure 8 As shown, it includes:

[0149] The first determining module 801 is used to acquire historical runoff data and runoff event cycles of the hydrological stations to be forecasted, and to determine the periodic time variable of the time of runoff event occurrence based on the runoff event cycle and the preset reference time.

[0150] The regression training module 802 is used to perform regression training on historical runoff data and the periodic time variables corresponding to the time of runoff events to obtain the probability distribution of the relative deviation.

[0151] The extraction module 803 is used to randomly extract the relative deviation of the time to be predicted based on the probability distribution of the relative deviation.

[0152] The second determining module 804 is used to determine the average trend data of runoff at the time to be predicted based on the reference time and the runoff event cycle.

[0153] The third determining module 805 is used to determine the runoff data for the forecast time based on the average runoff trend data and the relative deviation of the forecast time.

[0154] In some alternative implementations, the regression training module 802 includes:

[0155] The acquisition unit is used to acquire the functional relationship between historical runoff data and periodic time variables. Based on the periodic time variables corresponding to the time of runoff event, the average trend prediction data of runoff is determined by using the functional relationship between historical runoff data and periodic time variables.

[0156] The first determining unit is used to determine the relative deviation based on the average runoff trend prediction data and historical runoff data.

[0157] The first calculation unit is used to calculate the standard deviation of the relative deviation based on the relative deviation, and to obtain the probability distribution of the relative deviation based on the standard deviation of the relative deviation.

[0158] In some alternative implementations, the regression training module 802 further includes:

[0159] Sub-units are established to acquire the amount of historical runoff data. Based on the amount of historical runoff data, the number of decision trees and the search range of pre-pruning parameters are determined, and a parameter grid is established.

[0160] Determine the sub-units for setting cross-validation parameters and evaluation metrics. Based on the parameter grid, cross-validation fold number, and evaluation metrics, determine the optimal number of decision trees and pre-pruning parameters.

[0161] The computational subunit is used to obtain the functional relationship between the average trend of runoff and the periodic time variable based on the optimal number of decision trees and pre-pruning parameters using the random forest algorithm.

[0162] In some alternative implementations, the second determining module 804 includes:

[0163] The second calculation unit is used to calculate the periodic time variables of the time to be predicted based on the reference time and the runoff event cycle.

[0164] The second determining unit is used to determine the average trend data of runoff at the time to be predicted based on the periodic time variable of the time to be predicted.

[0165] In some optional implementations, the calculation formula for the periodic time variable of the runoff event occurrence time in the first determining module 801 is as follows:

[0166]

[0167] Where, τ i The periodic time variable representing the time of occurrence of a runoff event, where frac represents the fractional part, T represents the period of the runoff event, t0 represents the reference time, and t i Indicates the time when the runoff event occurred.

[0168] In some optional implementations, the calculation formula for the runoff data at the time to be predicted in the third determining module 805 is as follows:

[0169]

[0170] Among them, Q j This represents runoff data for the time to be predicted. r represents the average trend data of runoff at the time to be predicted. j This indicates the relative deviation of the time to be predicted.

[0171] Further functional descriptions of the above modules and units are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0172] In this embodiment, a method and apparatus for runoff forecasting based on statistical data analysis is presented in the form of functional units. Here, a unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above-mentioned functions.

[0173] This invention also provides a computer device having the above-described runoff forecasting device based on data statistical analysis.

[0174] Please see Figure 9 , Figure 9 This is a schematic diagram of the structure of a computer device provided in an optional embodiment of the present invention, such as... Figure 9As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system). Figure 9 Take a processor 10 as an example.

[0175] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0176] The memory 20 stores instructions executable by at least one processor 10 to cause at least one processor 10 to perform the method shown in the above embodiments.

[0177] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0178] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0179] The computer device also includes an input device 30 and an output device 40. The processor 10, memory 20, input device 30, and output device 40 can be connected via a bus or other means. Figure 9 Taking the example of a connection between China and Israel via a bus.

[0180] Input device 30 can receive input numerical or character information, and generate key signal inputs related to user settings and function control of the computer device, such as a touchscreen, keypad, mouse, trackpad, touchpad, joystick, one or more mouse buttons, trackball, joystick, etc. Output device 40 may include display devices, auxiliary lighting devices (e.g., LEDs), and haptic feedback devices (e.g., vibration motors). The aforementioned display devices include, but are not limited to, liquid crystal displays, light-emitting diodes, displays, and plasma displays. In some alternative embodiments, the display device may be a touchscreen.

[0181] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code, which, when accessed and executed by the computer, processor, or hardware, implements the methods shown in the above embodiments.

[0182] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0183] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A runoff forecasting method based on statistical data analysis, characterized in that, The method includes: Historical runoff data and runoff event cycles of the hydrological stations to be forecasted are obtained. Based on the runoff event cycles and a preset reference time, a periodic time variable for the occurrence time of runoff events is determined. The calculation formula for the periodic time variable for the occurrence time of runoff events is as follows: in, A periodic time variable representing the moment when a runoff event occurs. Indicates the decimal part. Indicates the runoff event cycle, Indicates the reference time. Indicates the time when the runoff event occurred; Regression training is performed on the historical runoff data and the periodic time variables corresponding to the time of the runoff event to obtain the probability distribution of the relative deviation; The relative deviation of the time to be predicted is randomly selected based on the probability distribution of the relative deviation. The average trend data of runoff at the time to be predicted is determined based on the reference time and the runoff event cycle. The runoff data for the forecast time is determined based on the average runoff trend data for the forecast time and the relative deviation of the forecast time; the calculation formula for the runoff data for the forecast time is as follows: in, This represents runoff data for the time to be predicted. This represents the average trend data of runoff at the time to be predicted. Indicates the relative deviation of the time to be predicted; The step of performing regression training on the historical runoff data and the periodic time variables corresponding to the time of the runoff event to obtain the probability distribution of the relative deviation includes: Obtain the functional relationship between the average runoff trend and the periodic time variable; based on the periodic time variable corresponding to the time of the runoff event, use the functional relationship between the average runoff trend and the periodic time variable to determine the average runoff trend prediction data. The relative deviation is determined based on the average runoff trend prediction data and the historical runoff data; The standard deviation of the relative deviation is calculated based on the relative deviation, and the probability distribution of the relative deviation is obtained based on the standard deviation of the relative deviation.

2. The method according to claim 1, characterized in that, Before obtaining the functional relationship between the average runoff trend and the periodic time variable, and determining the average runoff trend prediction data based on the periodic time variable corresponding to the time of the runoff event using the functional relationship between the average runoff trend and the periodic time variable, the method further includes: The amount of historical runoff data is obtained, and the number of decision trees and the search range of pre-pruning parameters are determined based on the amount of historical runoff data, and a parameter grid is established. Set cross-validation parameters and evaluation metrics, and determine the optimal number of decision trees and pre-pruning parameters based on the parameter grid, the number of cross-validation folds, and the evaluation metrics; Based on the optimal number of decision trees and the pre-pruning parameters, the random forest algorithm is used to obtain the functional relationship between the average runoff trend and the periodic time variable.

3. The method according to claim 1, characterized in that, The method for determining the average runoff trend data for the forecast time based on the reference time and the runoff event cycle includes: The periodic time variable of the time to be predicted is calculated based on the reference time and the runoff event cycle; The average trend data of runoff at the time to be predicted are determined based on the periodic time variables of the time to be predicted.

4. A runoff forecasting device based on statistical data analysis, characterized in that, The device includes: The first determining module is used to acquire historical runoff data and runoff event cycles of the hydrological stations to be forecasted, and to determine the periodic time variable of the runoff event occurrence time based on the runoff event cycle and a preset reference time; the calculation formula for the periodic time variable of the runoff event occurrence time is as follows: in, A periodic time variable representing the moment when a runoff event occurs. Indicates the decimal part. Indicates the runoff event cycle, Indicates the reference time. Indicates the time when the runoff event occurred; The regression training module is used to perform regression training on the historical runoff data and the periodic time variables corresponding to the time of the runoff event to obtain the probability distribution of the relative deviation. The regression training module includes: The acquisition unit is used to acquire the functional relationship between historical runoff data and periodic time variables. Based on the periodic time variables corresponding to the time of runoff event, the average trend prediction data of runoff is determined by using the functional relationship between historical runoff data and periodic time variables. The determination unit is used to determine the relative deviation based on runoff average trend prediction data and historical runoff data; The calculation unit is used to calculate the standard deviation of the relative deviation based on the relative deviation, and to obtain the probability distribution of the relative deviation based on the standard deviation of the relative deviation; The extraction module is used to randomly extract the relative deviation of the time to be predicted based on the probability distribution of the relative deviation; The second determining module is used to determine the average trend data of runoff at the time to be predicted based on the reference time and the runoff event cycle; The third determining module is used to determine the runoff data for the forecast time based on the average runoff trend data for the forecast time and the relative deviation of the forecast time; the calculation formula for the runoff data for the forecast time is as follows: in, This represents runoff data for the time to be predicted. This represents the average trend data of runoff at the time to be predicted. This indicates the relative deviation of the time to be predicted.

5. A computer device, characterized in that, include: The system includes a memory and a processor, which are interconnected. The memory stores computer instructions, and the processor executes the computer instructions to perform the runoff forecasting method based on data statistical analysis as described in any one of claims 1 to 3.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to execute the runoff forecasting method based on data statistical analysis as described in any one of claims 1 to 3.

7. A computer program product, characterized in that, Includes computer instructions for causing a computer to execute the runoff forecasting method based on data statistical analysis as described in any one of claims 1 to 3.

Images