Reservoir flood risk intelligent regulation and control method and system based on rainfall prediction

Abstract

The invention discloses a reservoir flood risk intelligent regulation and control method and system based on rainfall prediction, and particularly relates to the technical field of intelligent regulation and control, and the method comprises the steps: collecting the actual precipitation amount of each upstream river region and the predicted precipitation amount in weather forecast, and calculating the precipitation amount error of each region; and according to the precipitation amount of each region and the proportion of flowing into the river channel, evaluating an error score, and weighting the error score of each region by adopting a superior sequence diagram method. And then, integrating the error scores through a weighted average method to obtain an overall rainfall error state. And determining a predicted time range by using a fuzzy inference rule in combination with the overall error coefficient and the rainfall duration. Furthermore, the standard deviation and the mean value of the errors are calculated, and a future rainfall adjustment coefficient is calculated by using a logistic regression algorithm. And finally, calculating future precipitation according to the adjustment coefficient, the weather forecast data and the prediction time range. According to the method, the future precipitation can be accurately predicted, the risk alarm precision is improved, and the flood prevention and forecasting accuracy is optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent regulation, and more specifically, to an intelligent regulation method and system for reservoir flood risk based on rainfall prediction. Background Art

[0002] The background art of the intelligent regulation method and system for reservoir flood risk based on rainfall prediction is to improve the dispatching and management capabilities of reservoirs in the face of sudden heavy rain and floods, and ensure the safety of reservoirs and the flood control capabilities of downstream areas. To achieve this goal, with the progress of technology in recent years, the level of intelligence and precision in reservoir management and flood dispatching has been significantly improved.

[0003] Reservoirs play a crucial role in flood regulation, mainly by regulating water flow through means such as water storage, flood discharge, and dispatching. However, sudden heavy rainfall often increases the load on reservoirs, and without scientific early warning and reasonable dispatching, it is very likely to cause catastrophic consequences such as dam breaches and floods. Therefore, implementing intelligent regulation based on rainfall prediction can effectively improve the flood dispatching capabilities of reservoirs and reduce flood risks.

[0004] In the prior art, the consideration of the prediction range is lacking in the process of rainfall prediction, which easily leads to inaccurate prediction, unable to accurately predict the precipitation amount, resulting in deviations in risk alarms, and is not conducive to flood prevention.

[0005] To solve the above defects, a technical solution is provided now. Summary of the Invention

[0006] In order to overcome the above defects of the prior art, embodiments of the present invention provide an intelligent regulation method and system for reservoir flood risk based on rainfall prediction to solve the problems raised in the above background art.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] An intelligent regulation method for reservoir flood risk based on rainfall prediction, comprising the following steps:

[0009] Step S1: Collect the precipitation amounts in each upper stream river area and the predicted precipitation amounts in the weather forecast, and calculate the precipitation error in each area;

[0010] Step S2: Determine the error score by determining the precipitation amount in each area and the proportion of the precipitation flowing into the river in each area, assign weights to the error scores of each area according to the preference ranking method; perform weighted averaging on each area to obtain the overall error state;

[0011] Step S3: Use fuzzy rules for fuzzy inference based on the overall error coefficient and precipitation duration to determine the prediction time range;

[0012] Step S4: Determine the standard deviation of the overall error and the mean of the errors, calculate the future precipitation adjustment coefficient using the logistic regression algorithm, and determine the future precipitation amount according to the precipitation adjustment coefficient, the weather forecast data, and the prediction time range.

[0013] In a preferred embodiment, in step S1, collect the precipitation amounts in each upstream river channel area and the predicted precipitation amounts in the weather forecast. For each time point, calculate the forecast error: the error is equal to the actual precipitation amount minus the forecast precipitation amount.

[0014] In a preferred embodiment, in step S2, calculate the part of the precipitation in the basin converted into runoff by using the runoff formula; the commonly used formula is: Q = C * P * A; where: Q is the runoff volume flowing into the river channel; C is the runoff coefficient; P is the precipitation amount; A is the basin area.

[0015] In a preferred embodiment, after determining the total precipitation amount in the basin and the precipitation amount flowing into the river channel (i.e., the runoff volume), the proportion of precipitation flowing into the river channel can be calculated: the proportion is equal to the precipitation amount flowing into the river channel divided by the total precipitation amount multiplied by 100%.

[0016] In a preferred embodiment, in step S2, the error score of each region is equal to the normalized precipitation amount of each region plus the proportion of precipitation flowing into the river channel in each region; after calculating the error scores of each region, assign weights to the error scores of each region respectively according to the preference ranking method, determine the weight values of the errors in each region, and perform weighted average calculation according to the weight values of the errors in each region to obtain the overall error coefficient of all regions.

[0017] In a preferred embodiment, in step S3, set the overall error coefficient and the precipitation duration as input variables, divide them into different fuzzy sets respectively, define the prediction time range as the output variable, and divide it into a fuzzy set; formulate a set of fuzzy rules to describe the influence of different input variables on the output variable, and the definition of the rules is obtained based on professional knowledge and data analysis; perform fuzzy reasoning according to the fuzzy rules to determine the prediction time range.

[0018] In a preferred embodiment, in step S4, determine the mean and standard deviation of the prediction time range and the overall error coefficient; specifically, the calculation formula for the mean of the overall error coefficient is as follows: e i is the error of the i-th measurement value; n is the number of measurements; a is the mean of the overall error coefficient; the calculation formula for the standard deviation of the overall error coefficient is as follows: where e i is the error of the i-th measurement value, a is the mean of the overall error coefficient, n is the number of measurements, and b is the standard deviation of the overall error coefficient.

[0019] In a preferred embodiment, the future precipitation adjustment coefficient is calculated according to the standard deviation and mean of the overall error coefficient using the following formula: P = 1 - e -(aα+bβ) ; P represents the precipitation adjustment coefficient, a is the mean of the overall error coefficient, and b is the standard deviation of the overall error coefficient.

[0020] In a preferred embodiment, in step S4, according to the precipitation adjustment coefficient and the prediction time range, the predicted precipitation for a future period of the weather forecast is adjusted to generate the predicted precipitation. If the error is positive, it is adjusted positively; if the error is negative, it is adjusted negatively.

[0021] In a preferred embodiment, it includes: a data collection and preprocessing module, an error scoring and weighting module, a fuzzy inference and prediction time range module, a statistical analysis and regression model module, and a future precipitation prediction module;

[0022] The data collection and preprocessing module collects the precipitation, the predicted precipitation of the weather forecast, and the precipitation duration in each upstream river area, and calculates the precipitation error in each area; transmits the precipitation to the error scoring and weighting module, transmits the precipitation duration to the fuzzy inference and prediction time range module, and transmits the predicted precipitation of the weather forecast to the future precipitation prediction module;

[0023] The error scoring and weighting module calculates the error score in each area according to the actual precipitation in each area and the proportion of the precipitation flowing into the river, assigns weights to the error scores using the preference ranking organization method, and performs a weighted average on the scores of each area to obtain the overall error state; transmits the overall error state to the fuzzy inference and prediction time range module and the statistical analysis and regression model module;

[0024] The fuzzy inference and prediction time range module determines the prediction time range based on the overall error coefficient and the precipitation duration using fuzzy inference rules; transmits the prediction time range to the future precipitation prediction module;

[0025] The statistical analysis and regression model module calculates the standard deviation and mean of the error, and adjusts the prediction coefficient of the future precipitation using the logistic regression algorithm; transmits the prediction coefficient to the future precipitation prediction module;

[0026] The future precipitation prediction module combines the precipitation adjustment coefficient, the weather forecast data, and the prediction time range to predict the precipitation for a future period.

[0027] The technical effects and advantages of the present invention:

[0028] The present invention first collects the actual precipitation in each upstream river area and the predicted precipitation in the weather forecast, and calculates the precipitation error in each area. Then, it determines the precipitation in each area and the proportion of the precipitation flowing into the river, and evaluates the error score accordingly. The preference ranking organization method (PROMETHEE) is used to assign weights to the error scores of each area. Next, based on the weighted average method, the error scores of each area are integrated to obtain the overall precipitation error state. On this basis, combining the overall error coefficient and the precipitation duration, the prediction time range is determined through fuzzy inference rules. Subsequently, the standard deviation and mean of the overall error are calculated, and the adjustment coefficient of future precipitation is deduced using the logistic regression algorithm. Finally, according to the adjustment coefficient, weather forecast data, and prediction time range, the precipitation in a future period is calculated. It can accurately predict the future precipitation, improve the accuracy of risk alarm, and make the flood prevention more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] For the convenience of those skilled in the art to understand, the present invention will be further described below with reference to the accompanying drawings;

[0030] Figure 1 is a schematic flowchart of the intelligent regulation method for reservoir flood risk based on rainfall prediction of the present invention;

[0031] Figure 2 is a schematic structural diagram of the intelligent regulation system for reservoir flood risk based on rainfall prediction of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0032] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0033] The present invention first collects the actual precipitation in each upstream river area and the predicted precipitation in the weather forecast, and calculates the precipitation error in each area. Then, it determines the precipitation in each area and the proportion of the precipitation flowing into the river, and evaluates the error score accordingly. The preference ranking organization method (PROMETHEE) is used to assign weights to the error scores of each area. Next, based on the weighted average method, the error scores of each area are integrated to obtain the overall precipitation error state. On this basis, combining the overall error coefficient and the precipitation duration, the prediction time range is determined through fuzzy inference rules. Subsequently, the standard deviation and mean of the overall error are calculated, and the adjustment coefficient of future precipitation is deduced using the logistic regression algorithm. Finally, according to the adjustment coefficient, weather forecast data, and prediction time range, the precipitation in a future period is calculated. It can accurately predict the future precipitation, improve the accuracy of risk alarm, and make the flood prevention more accurate.

[0034] Example 1

[0035] The present invention provides an intelligent regulation method for reservoir flood risk based on rainfall prediction as shown in Figure 1 the following, including the steps:

[0036] Step S1: Collect the precipitation in each upper river channel area and the predicted precipitation in the weather forecast, and calculate the precipitation error in each area;

[0037] Step S2: Determine the precipitation in each area and the proportion of precipitation flowing into the river in each area to determine the error score, and assign weights to the error scores of each area according to the preference ranking method; perform weighted averaging on each area to obtain the overall error state;

[0038] Step S3: Use fuzzy rules for fuzzy inference according to the overall error coefficient and precipitation duration to determine the prediction time range;

[0039] Step S4: Determine the standard deviation of the overall error and the mean value of the error, calculate the future precipitation adjustment coefficient using the logistic regression algorithm, and determine the future precipitation according to the precipitation adjustment coefficient, weather forecast data, and prediction time range.

[0040] In step S1, install multiple rain gauges in each area for on-site data collection; common rain gauges include cylinder rain gauges, tipping bucket rain gauges, etc. The rain gauge has functions of automatic recording and wireless transmission, and summarizes and transmits the real-time data into the system; calculate the on-site average precipitation in each area as the actual precipitation in that area.

[0041] Obtain meteorological data such as the predicted precipitation and precipitation duration in each area through a weather station.

[0042] For each time point, calculate the forecast error, that is, the difference between the predicted value and the actual value: the error is equal to the actual precipitation minus the forecast precipitation; this error value can be positive or negative, indicating that the forecast precipitation is too high or too low.

[0043] In step S2, determine the proportion of precipitation flowing into the river in each area, specifically including the following steps:

[0044] A1: Collect precipitation data;

[0045] Meteorological data: First of all, it is necessary to collect the precipitation in this area;

[0046] Spatial distribution: The spatial distribution of precipitation may be uneven, so it is necessary to obtain precipitation data in different areas of the basin (such as upstream, middle reaches, downstream), and even consider the influence of terrain on precipitation (such as mountains, slopes, wind directions, etc.).

[0047] A2: Determine the basin area and river channel location;

[0048] Basin area: It is necessary to clarify the geographical scope of the river basin. This can usually be determined through a Digital Elevation Model (DEM), and DEM data can help identify the boundaries of the basin.

[0049] River channel location: Define the scope of the main river channel and the flow monitoring points.

[0050] A3: Estimate the catchment area of precipitation;

[0051] Based on the watershed (basin boundary) of the basin, determine the area where precipitation flows into the river channel. Not all precipitation flows towards the river channel in all areas. Some precipitation will infiltrate into the ground or disappear through evaporation and transpiration, etc.

[0052] Evaporation and transpiration: These processes will reduce the proportion of precipitation entering the river channel. Hydrological models can be used to estimate these losses.

[0053] A4: Hydrological model analysis;

[0054] Hydrological model: Use hydrological models such as SWAT (Soil and Water Assessment Tool), HEC-HMS (Hydrologic Engineering Center - Hydrologic Modeling System) to simulate the flow process of precipitation in the basin. The model can help calculate the proportion of precipitation converted into runoff and can take into account factors such as groundwater recharge, evaporation and transpiration, etc.

[0055] Runoff coefficient: The proportion of precipitation converted into surface runoff in each area depends on factors such as surface type (such as forest, urban, agricultural land, etc.), soil type, terrain slope, etc. The hydrological model will estimate a runoff coefficient (Runoff Coefficient, CC) for a region based on these factors.

[0056] A5: Calculate the precipitation flowing into the river channel;

[0057] Runoff calculation: Calculate the part of precipitation converted into runoff in the basin by using the runoff formula. The commonly used formula is: Q = C * P * A; where: Q is the runoff volume flowing into the river channel (cubic meters per second or cubic meters per year, etc.); C is the runoff coefficient; P is the precipitation amount (millimeters); A is the basin area (square kilometers or square meters). This can help estimate the total amount of precipitation flowing into the river channel in a certain area.

[0058] A6: Determine the proportion of precipitation flowing into the river channel;

[0059] After obtaining the total precipitation amount in the basin and the precipitation amount flowing into the river channel (i.e., runoff volume), the proportion of precipitation flowing into the river channel can be calculated: the proportion is equal to the precipitation amount flowing into the river channel divided by the total precipitation amount multiplied by 100%;

[0060] The calculated proportion of precipitation flowing into the river channel is the historical proportion of precipitation flowing into the river channel in this area obtained from historical data.

[0061] Determine the precipitation amount and the proportion of precipitation flowing into the river channel in each region, and perform data processing on the precipitation amount and the proportion of precipitation flowing into the river channel in each region using linear normalization; specifically, use the following formula to normalize the data R i to the range of [0, 1]: where R' i is the normalized value, and R min and R max are the minimum and maximum values of the data R i respectively.

[0062] In this embodiment, the area is divided into five regions; specifically, the error scores of calculating each region in this embodiment are exemplified as follows:

[0063] The error score of each region is equal to the normalized precipitation amount in each region plus the proportion of precipitation flowing into the river channel in each region; after calculating the error scores of each region, compare the magnitudes of the error scores of each region. Within a unit time, the larger the precipitation amount or the proportion of precipitation flowing into the river channel in each region, the more important the region is, the more it needs to be observed intensively, and the greater the error consideration weight of the region should be. The unit time can be set according to requirements.

[0064] Calculate the error scores of each region according to the normalized precipitation amount in each region plus the proportion of precipitation flowing into the river channel in each region, and assign weights to the error scores of each region respectively according to the preference ranking method to determine the weight values of the errors in each region.

[0065] Further, the weights are assigned to the error scores of each region respectively according to the preference ranking method as shown in Table 1 below:

[0066]

[0067] Table 1

[0068] Calculate the weighted average of the unit time errors of each monitoring region according to the weight values of the errors in each region to obtain the overall error coefficient of all regions. Calculate the relative weights of each region separately and then perform weighted averaging to make the overall error coefficient more representative and more accurately reflect the error coefficient of the overall region.

[0069] In step S3, the overall error coefficient and precipitation duration are set as input variables and are respectively divided into different fuzzy sets. For example, four ranges are set for the error coefficient, namely "Tiny", "Medium", "Big", and "Huge"; four ranges are set for the precipitation duration, namely "Brief", "Shorter", "Lengthy", and "Infinite".

[0070] The predicted time range is defined as the output variable and is divided into fuzzy sets, such as "Brief", "Shorter", "Lengthy", "Infinite", for the predicted time range. It should be noted that the division of the fuzzy sets can be adjusted according to the actual situation. For example, although four fuzzy sets are taken as examples in this embodiment, in fact, the overall error coefficient, precipitation duration, and predicted time range can be divided into more than four sets according to the actual situation to facilitate better precise adjustment.

[0071] Formulate a set of fuzzy rules to describe the influence of different input variables on the output variable. The definition of the rules is obtained based on professional knowledge and data analysis. For example:

[0072] Mark the overall error coefficient as W, the precipitation duration as C, and the predicted time range as Y_time, then the following can be defined

[0073] Rule 1:IF (W is Tiny) AND (C is Infinite) THEN (T_time is Infinite)

[0074] Rule 2:IF (W is Huge) AND (C is Brief) THEN (T_time is Brief) ...

[0076] Conduct fuzzy inference according to the fuzzy rules to determine the predicted time range.

[0077] Furthermore, according to which fuzzy set the overall error coefficient, precipitation duration, and predicted time range belong to, thresholds can be set for judgment according to the actual situation. For example, when the overall error coefficient exceeds 5 mm, it is marked as "Huge", and when the precipitation duration is higher than 20 hours, it is marked as "Infinte", etc., which will not be elaborated here.

[0078] The predicted time range is the precipitation amount predicted for a period of time in the future. The "Brief", "Shorter", "Lengthy", "Infinite" obtained after fuzzy inference can be converted into corresponding durations, which will not be elaborated here.

[0079] In step S4, determine the mean and standard deviation of the prediction time range and the overall error coefficient;

[0080] Specifically, the calculation formula for the mean of the overall error coefficient is as follows: e i is the error of the i-th measurement value; n is the number of measurements; a is the mean of the overall error coefficient.

[0081] The standard deviation of the error measures the degree of dispersion of the error and reflects the fluctuation of the difference between the measured value and the true value. The larger the standard deviation, the greater the fluctuation of the error, indicating a higher degree of uncertainty in the measurement. The calculation formula for the standard deviation of the overall error coefficient is as follows: where e i is the error of the i-th measurement value, μ is the mean of the overall error coefficient, n is the number of measurements, and b is the standard deviation of the overall error coefficient.

[0082] Calculate the future precipitation adjustment coefficient according to the standard deviation of the overall error coefficient and the mean of the error using the following formula: P = 1 - e -(aα+bβ) ; P represents the precipitation adjustment coefficient, a is the mean of the overall error coefficient. The larger the mean of the overall error coefficient, the larger the precipitation adjustment coefficient, and vice versa; b is the standard deviation of the overall error coefficient. The larger the standard deviation of the overall error coefficient, the larger the precipitation adjustment coefficient, and vice versa. The mean of the error in this formula is the value after removing the sign. Adjust the predicted precipitation for a future period of the weather forecast according to the precipitation adjustment coefficient and the prediction time range to generate the predicted precipitation. If the error is positive, adjust it positively; if the error is negative, adjust it negatively; the future period is the prediction time range.

[0083] Embodiment 2

[0084] The present invention provides an intelligent regulation system for reservoir flood risk based on rainfall prediction as Figure 2 shown, including: a data acquisition and preprocessing module, an error scoring and weighting module, a fuzzy reasoning and prediction time range module, a statistical analysis and regression model module, and a future precipitation prediction module;

[0085] The data acquisition and preprocessing module collects the precipitation, the predicted precipitation of the weather forecast, and the precipitation duration in each upstream river area, and calculates the precipitation error in each area; transmits the precipitation to the error scoring and weighting module, transmits the precipitation duration to the fuzzy reasoning and prediction time range module, and transmits the predicted precipitation of the weather forecast to the future precipitation prediction module;

[0086] The error scoring and weighting module calculates the error score for each region based on the actual precipitation in each region and the proportion of precipitation flowing into the river channel, assigns weights to the error scores using the preference ranking organization method, performs a weighted average on the scores of each region to obtain the overall error status, and transmits the overall error status to the fuzzy inference and prediction time range module and the statistical analysis and regression model module.

[0087] The fuzzy inference and prediction time range module determines the prediction time range using fuzzy inference rules based on the overall error coefficient and precipitation duration, and transmits the prediction time range to the future precipitation prediction module.

[0088] The statistical analysis and regression model module calculates the standard deviation and mean of the error, and adjusts the prediction coefficient of future precipitation using the logistic regression algorithm, and transmits the prediction coefficient to the future precipitation prediction module.

[0089] The future precipitation prediction module predicts the precipitation for a future period by combining the precipitation adjustment coefficient, weather forecast data, and the prediction time range.

[0090] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0091] In several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed mutual coupling or direct coupling or communication connection can be through some interfaces, and the indirect coupling or communication connection of the device or unit can be in electrical, mechanical, or other forms.

[0092] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place, or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0093] In addition, each functional unit in various embodiments of the present application may be integrated into one processing unit, may exist physically alone for each unit, or two or more units may be integrated into one unit.

[0094] As described above, it is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed in the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. An intelligent control method for reservoir flood risk based on rainfall prediction, characterized in that: The following steps are involved: Step S1: Collect the precipitation in each upstream river area and the expected precipitation in the weather forecast, and calculate the precipitation error in each area; Step S2: Determine the precipitation in each area and the proportion of precipitation in each area flowing into the river to determine the error score, assign weights to the error scores of each area according to the priority diagram method; perform weighted average on each area to obtain the overall error status; Step S3: using fuzzy rules to perform fuzzy reasoning based on the overall error coefficient and the precipitation duration to determine the prediction time range; Step S4: Determine the standard deviation of the overall error and the mean of the error, calculate the future precipitation adjustment coefficient using a logistic regression algorithm, and determine the future precipitation according to the precipitation adjustment coefficient, weather forecast data, and the forecast time range.

2. The intelligent control method for reservoir flood risk based on rainfall prediction according to claim 1 is characterized by: In step S1, the precipitation in each upstream river channel area and the predicted precipitation in the weather forecast are collected, and for each time point, the forecast error is calculated: the error is equal to the actual precipitation minus the predicted precipitation.

3. The intelligent control method for reservoir flood risk based on rainfall prediction according to claim 1 is characterized by: In step S2, the portion of precipitation in the basin converted into runoff is calculated by using the runoff formula; the commonly used formula is: Q = C*P*A; where: Q is the runoff amount flowing into the river; C is the runoff coefficient; P is the precipitation; and A is the basin area.

4. The intelligent control method for reservoir flood risk based on rainfall prediction according to claim 3 is characterized by: After determining the total precipitation in the basin and the total precipitation flowing into the river (i.e. runoff), the proportion of precipitation flowing into the river can be calculated: the proportion is equal to the precipitation flowing into the river divided by the total precipitation multiplied by the percentage.

5. The intelligent control method for reservoir flood risk based on rainfall prediction according to claim 1 is characterized by: In step S2, the error score of each area is equal to the normalized precipitation in each area plus the proportion of precipitation in each area flowing into the river. After the error score of each area is calculated, the error score of each area is weighted according to the priority diagram method to determine the weight value of the error of each area. The weighted average calculation is performed according to the weight value of the error of each area to obtain the overall error coefficient of all areas.

6. The intelligent control method for reservoir flood risk based on rainfall prediction according to claim 1 is characterized by: In step S3, the overall error coefficient and precipitation duration are set as input variables, which are divided into different fuzzy sets respectively; the prediction time range is defined as the output variable and divided into fuzzy sets; a set of fuzzy rules are formulated to describe the influence of different input variables on the output variables, and the definition of the rules is based on professional knowledge and data analysis; fuzzy reasoning is performed according to the fuzzy rules to determine the prediction time range.

7. The intelligent control method for reservoir flood risk based on rainfall prediction according to claim 1 is characterized by: In step S4, the mean and standard deviation of the prediction time range and the overall error coefficient are determined; specifically, the calculation formula of the mean of the overall error coefficient is as follows: e i is the error of the ith measurement value; n is the number of measurements; a is the mean of the overall error coefficient; the calculation formula for the standard deviation of the overall error coefficient is as follows: where e i is the error of the ith measurement value, a is the mean of the overall error coefficient, n is the number of measurements, and b is the standard deviation of the overall error coefficient.

8. The intelligent control method for reservoir flood risk based on rainfall prediction according to claim 7 is characterized by: The future precipitation adjustment coefficient is calculated based on the overall error coefficient standard deviation and the overall error coefficient mean using the following formula: P = 1-e -(aα+bβ) ; P represents the precipitation adjustment coefficient, a is the mean of the overall error coefficient, and b is the standard deviation of the overall error coefficient.

9. The intelligent control method for reservoir flood risk based on rainfall prediction according to claim 1 is characterized by: In step S4, the forecast precipitation for a period of time in the weather forecast is adjusted according to the precipitation adjustment coefficient and the forecast time range to generate the forecast precipitation. If the error is positive, a positive adjustment is made; if the error is negative, a negative adjustment is made.

10. A reservoir flood risk intelligent control system based on rainfall prediction, used to implement the reservoir flood risk intelligent control method based on rainfall prediction according to any one of claims 1 to 9, characterized in that: include: Data collection preprocessing module, error scoring and weighting module, fuzzy reasoning and prediction time range module, statistical analysis and regression model module, future precipitation prediction module; The data collection and preprocessing module collects the precipitation in each upstream river area and the expected precipitation and precipitation duration in the weather forecast, and calculates the precipitation error in each area; transmits the precipitation to the error scoring and weighting module, transmits the precipitation duration to the fuzzy reasoning and prediction time range module, and transmits the expected precipitation in the weather forecast to the future precipitation prediction module; The error scoring and weighting module calculates the error score of each area according to the actual precipitation in each area and the proportion of precipitation flowing into the river, uses the priority diagram method to assign weights to the error scores, and performs weighted average on the scores of each area to obtain the overall error state; the overall error state is transmitted to the fuzzy reasoning and prediction time range module and the statistical analysis and regression model module; The fuzzy reasoning and prediction time range module uses fuzzy reasoning rules to determine the prediction time range based on the overall error coefficient and precipitation duration; and transmits the prediction time range to the future precipitation prediction module; The statistical analysis and regression model module calculates the standard deviation and mean of the error, and uses the logistic regression algorithm to adjust the prediction coefficient of future precipitation; the prediction coefficient is transmitted to the future precipitation prediction module; The future precipitation prediction module combines the precipitation adjustment coefficient, weather forecast data and prediction time range to predict the precipitation in the future.

Images