Disposal method for events influencing cascade reservoir dispatching operation

Abstract

The invention relates to the technical field of reservoir dispatching, in particular to a method for handling events influencing cascade reservoir dispatching operation, which comprises the following steps: acquiring real-time monitoring data of a cascade reservoir to be detected; performing data preprocessing on the real-time monitoring data of the to-be-detected cascade reservoir to obtain preprocessed data; inputting the preprocessed data into a trained key factor determination model, and obtaining a key data type and corresponding data; determining whether an abnormal event occurs or not and the type of the abnormal event according to the key data type and the corresponding data; when an abnormal event occurs, inputting the type of the abnormal event into a preset intelligent diagnosis system, and obtaining a plurality of exception handling schemes; and SWOT analysis is carried out on the various exception handling schemes to obtain a final handling result. According to the method, the safety and reliability of cascade reservoir dispatching operation are improved, and the cascade reservoir operation management level is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of reservoir dispatching, and in particular to a method for handling events that affect the dispatching operation of cascade reservoirs. Background Art

[0002] In the traditional cascade reservoir operation process, the event handling mode mainly relies on the early warning and manual operation of the on-duty personnel. The on-duty personnel need to monitor the hydrological, meteorological, and electric power data in real time, and issue early warning signals when abnormalities occur. The relevant handling personnel then analyze and calculate, evaluate the best handling method, and notify the relevant departments to perform corresponding operations and adjustments. However, this model has many limitations: first, long-term manual duty not only consumes human resources, but also causes fatigue of the on-duty personnel, which in turn affects work efficiency and accuracy; second, the delay of manual operation makes the process from the occurrence of the event to the completion of the handling longer, especially when handling emergencies, this delay may lead to the loss of the handling opportunity; in addition, human factors may lead to false alarms or missed alarms, and different judgment standards of different personnel may also cause inconsistent handling methods, affecting the handling effect.

[0003] Therefore, there is an urgent need for a method to deal with events that affect the scheduling and operation of cascade reservoirs. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present invention provides a method for handling events that affect the scheduling and operation of cascade reservoirs, which solves the technical problems in the prior art that the event handling process is complicated, involves multiple departments and personnel, and the handling process is long, resulting in untimely handling and low efficiency, and human operation is prone to omissions and errors, which increases the risk of event handling.

[0006] (II) Technical solution

[0007] In order to achieve the above object, the main technical solutions adopted by the present invention include:

[0008] In a first aspect, an embodiment of the present invention provides a method for handling events that affect the scheduling and operation of cascade reservoirs, including:

[0009] S10, obtaining real-time monitoring data of the cascade reservoirs to be detected; performing data preprocessing on the real-time monitoring data of the cascade reservoirs to be detected to obtain preprocessed data;

[0010] The real-time monitoring data includes: hydrological data, meteorological data, power data and engineering safety data;

[0011] S20, inputting the preprocessed data into the trained key factor determination model to obtain key data types and corresponding data; and determining whether an abnormal event occurs and the type of abnormal event according to the key data types and corresponding data;

[0012] The key factor determination model includes: a LASSO regression sub-model and a tree sub-model;

[0013] S30, when an abnormal event occurs, the type of the abnormal event is input into a pre-set intelligent diagnosis system to obtain a variety of abnormal handling solutions;

[0014] S40: Conduct a SWOT analysis on the various abnormality handling solutions to obtain a final handling result.

[0015] Optionally, in S10, the real-time monitoring data of the cascade reservoir to be detected is preprocessed, and obtaining the preprocessed data includes:

[0016] S11, cleaning the real-time monitoring data of the cascade reservoirs to be detected to obtain cleaned data;

[0017] The data cleaning includes: missing value processing and outlier monitoring;

[0018] S12, inputting the cleaned data into a Kalman filter, performing real-time denoising and smoothing processing, and obtaining pre-processed data.

[0019] Optionally, the S10 further includes:

[0020] Use Page-Hinkley detection to evaluate the time series characteristics of the real-time monitoring data in real time and determine whether the change trend of the real-time monitoring data is greater than a pre-set threshold;

[0021] If the change trend of the real-time monitoring data is greater than a pre-set threshold, increasing the sampling frequency of the real-time monitoring data;

[0022] If the change trend of the real-time monitoring data is less than or equal to a preset threshold, the sampling frequency of the real-time monitoring data is reduced.

[0023] Optionally, in S20, inputting the preprocessed data into a trained key factor determination model, and obtaining key data types and corresponding data includes:

[0024] S21, inputting the preprocessed data into the trained LASSO regression sub-model to obtain the first key data type that is initially screened;

[0025] S22. Input the first key data types into the tree sub-model, obtain the feature importance score corresponding to each first key data type, and obtain the first key data types with the top 10% scores as the final key data types.

[0026] Optionally, before S21, the step further includes:

[0027] S21a, collecting historical event data, using the historical event data to train the LASSO regression sub-model, and obtaining a trained LASSO regression sub-model;

[0028] The historical event data includes:

[0029] The hydrological data, meteorological data, power data and engineering safety monitoring data corresponding to the historical events.

[0030] Optionally, the S21a specifically includes:

[0031] Classifying the historical event data according to event types, and assigning labels to each historical event data based on the classification results;

[0032] Each data in the classified historical event data is input into the LASSO regression sub-model, and the cross entropy function and Adam optimizer are used with a learning rate of 0.001 until the cross entropy regression loss function is minimized to obtain the trained LASSO regression sub-model.

[0033] Optionally, in S20, determining whether an abnormal event occurs and the type of the abnormal event according to the key data type and the corresponding data includes:

[0034] Input the key data type and corresponding data into the abnormal event determination model to determine whether an abnormal event occurs and obtain the type of the abnormal event;

[0035] The abnormal event determination model is a convolutional neural network model; the convolutional neural network model includes: an input layer, a hidden layer, and an output layer; wherein the input layer inputs key data types and corresponding data; the hidden layer includes multiple neurons, which are used to determine whether an abnormal event occurs and obtain the type of the abnormal event through a sigmoid activation function; and the output layer outputs the type of the abnormal event.

[0036] Optionally, in S30,

[0037] The intelligent diagnosis system comprises:

[0038] Reservoir safety operation risk assessment models, water resources efficient utilization economic evaluation models, river flooding analysis models and water body ecological environment evaluation models provided by different model providers.

[0039] Optionally, the S40 includes:

[0040] S41. Conduct a SWOT analysis of each disposal option, including identifying its strengths, weaknesses, opportunities and threats;

[0041] S42. Assign points to the strengths, weaknesses, opportunities and threats of each disposal option, where 0 means the factor does not exist or has the least impact, and 5 means the factor has the greatest impact;

[0042] S43. Add up the scores of the advantages, disadvantages, opportunities and threats to obtain the total score of each solution, and select the solution with the highest total score as the final solution.

[0043] Optionally, the method further comprises:

[0044] S50, constructing an event library, and storing the real-time monitoring data, each event type and its corresponding key data type, and the final processing result corresponding to each event type in the event library.

[0045] (III) Beneficial effects

[0046] The beneficial effects of the present invention are: compared with the prior art, the method for handling events that affect the scheduling and operation of cascade reservoirs of the present invention can improve the safety and reliability of the scheduling and operation of cascade reservoirs, continuously improve the operation and management level of cascade reservoirs, and also provide a systematic and scientific event handling plan for reservoir management departments, which has important practical value and promotion significance. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 A schematic flow chart of a method for handling events that affect the scheduling and operation of cascade reservoirs according to an embodiment of the present invention;

[0048] Figure 2 This is a flowchart of a method for handling events that affect the scheduling and operation of cascade reservoirs according to Example 2 of the present invention. DETAILED DESCRIPTION

[0049] In order to better explain the present invention and facilitate understanding, the present invention is described in detail below through specific implementation modes in conjunction with the accompanying drawings.

[0050] The embodiment of the present invention proposes a method for handling events that affect the scheduling and operation of cascade reservoirs. In order to solve the technical problems that the event handling process in the prior art is complicated, involves multiple departments and personnel, and the handling process is long, resulting in untimely handling, low efficiency, and human operation is prone to omissions and errors, which increases the risk of event handling. The trained key factor determination model is used to obtain key data types and corresponding data, and whether an abnormal event occurs and the type of abnormal event are determined based on the key data types and corresponding data. When an abnormal event occurs, the type of abnormal event is input into the pre-set intelligent diagnosis system to obtain a variety of abnormal handling solutions. Thereby improving the efficiency and accuracy of event handling, reducing human errors, and ensuring the safety and reliability of the scheduling and operation of cascade reservoirs.

[0051] In order to better understand the above technical solution, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to enable a clearer and more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0052] Example 1

[0053] See also Figure 1 A method for handling events that affect the operation of cascade reservoirs according to an embodiment of the present invention includes:

[0054] Step S10, obtaining real-time monitoring data of the cascade reservoirs to be detected; performing data preprocessing on the real-time monitoring data of the cascade reservoirs to be detected to obtain preprocessed data;

[0055] Real-time monitoring data include: hydrological data, meteorological data, power data and engineering safety data;

[0056] Step S20: input the preprocessed data into the trained key factor determination model to obtain key data types and corresponding data; and determine whether an abnormal event occurs and the type of abnormal event according to the key data types and corresponding data;

[0057] The key factor determination model includes: LASSO regression sub-model and tree sub-model;

[0058] Step S30: When an abnormal event occurs, the type of the abnormal event is input into a pre-set intelligent diagnosis system to obtain a variety of abnormal handling solutions;

[0059] Step S40: Conduct a SWOT analysis on various abnormality handling solutions to obtain a final handling result.

[0060] In this embodiment, real-time monitoring data of cascade reservoirs are collected through sensors and other monitoring equipment; among them, hydrological data include water level, flow, water quality, storage capacity, inflow and outflow; meteorological data include rainfall, temperature, humidity, wind speed and direction, etc.; power data include power generation, load, voltage, current and power factor, etc.; engineering safety data include structural stress, displacement, leakage, vibration and crack detection, etc.

[0061] Real-time monitoring of the above data is particularly important for ensuring the safe operation of cascade reservoirs, and is also an important basis for specifying disposal plans. Through comprehensive analysis of these data, we can better understand the changes in the cascade reservoirs and their surrounding environment, so as to adopt appropriate disposal methods.

[0062] Through the above steps, the present invention can effectively preprocess the real-time monitoring data of cascade reservoirs, select key factor types, determine event types, generate abnormal handling plans, and finally select them. It can not only improve the efficiency and correctness of event handling and reduce the impact of events on the scheduling and operation of cascade reservoirs, but also improve the safety and reliability of the scheduling and operation of cascade reservoirs, thereby continuously improving the operation and management level of cascade reservoirs.

[0063] Example 2

[0064] See also Figure 2 A method for handling an event affecting the operation of cascade reservoirs according to Embodiment 2 of the present invention comprises:

[0065] Step S10, obtaining real-time monitoring data of the cascade reservoirs to be detected; performing data preprocessing on the real-time monitoring data of the cascade reservoirs to be detected to obtain preprocessed data;

[0066] Real-time monitoring data include: hydrological data, meteorological data, power data and engineering safety data;

[0067] Step S20: input the preprocessed data into the trained key factor determination model to obtain key data types and corresponding data; and determine whether an abnormal event occurs and the type of abnormal event according to the key data types and corresponding data;

[0068] The key factor determination model includes: LASSO regression sub-model and tree sub-model;

[0069] Step S30: When an abnormal event occurs, the type of the abnormal event is input into a pre-set intelligent diagnosis system to obtain a variety of abnormal handling solutions;

[0070] Step S40: Conduct a SWOT analysis on various abnormality handling solutions to obtain a final handling result.

[0071] In this embodiment, in step S10, the real-time monitoring data of the cascade reservoir to be detected is preprocessed, and the preprocessed data is obtained including:

[0072] Step S11, cleaning the real-time monitoring data of the cascade reservoirs to be detected to obtain cleaned data;

[0073] Data cleaning includes: missing value processing and outlier monitoring;

[0074] Step S12: input the cleaned data into the Kalman filter for real-time denoising and smoothing to obtain pre-processed data.

[0075] In the specific implementation process, the real-time monitoring data is first monitored for missing values ​​and filled using interpolation methods such as linear interpolation or polynomial interpolation. After filling, the filled data is monitored for outliers, and the data is grouped through a clustering algorithm (such as K-means), and points far away from the center of the cluster are identified as outliers, or the original data is reconstructed through an autoencoder model. If the reconstruction error is greater than a certain threshold, it is marked as an outlier.

[0076] After data cleaning, the cleaned data is input into the Kalman filter; specifically, through the state equation and observation equation of the Kalman filter, combined with the dynamic characteristics and observation values ​​of the system, state prediction is first performed, and then the state estimate is updated using the Kalman gain, thereby achieving real-time denoising and smoothing of the original data.

[0077] This method can effectively remove noise interference in the data, and can also optimize the time series characteristics of the data through a smoothing algorithm, ultimately obtaining high-quality preprocessed data, providing a reliable basis for subsequent key factor analysis and event type determination.

[0078] In the specific implementation process, step S10 also includes:

[0079] Use Page-Hinkley detection to evaluate the time series characteristics of real-time monitoring data in real time and determine whether the change trend of real-time monitoring data is greater than the threshold set in advance;

[0080] If the change trend of the real-time monitoring data is greater than the pre-set threshold, increase the sampling frequency of the real-time monitoring data;

[0081] If the change trend of the real-time monitoring data is less than or equal to the threshold set in advance, the sampling frequency of the real-time monitoring data is reduced.

[0082] Specifically, the Page-Hinkley detection algorithm is used to dynamically evaluate whether the change trend of the monitoring data exceeds the preset threshold. If the test results show that the change trend of the data exceeds this threshold, it indicates that there are significant changes or abnormalities in the current data. At this time, the monitoring device will automatically increase the sampling frequency of the real-time monitoring data to capture more details and changes; on the contrary, if the change trend of the data is less than or equal to the preset threshold, that is, when the data changes are relatively stable, the sampling frequency will be reduced to reduce unnecessary data redundancy and processing burden. This method can optimize data processing efficiency while ensuring data quality, ensuring that important change information can be captured in time at critical moments.

[0083] For example, suppose that the current water level of the reservoir is relatively stable and does not change much. At this time, data is collected at a frequency of once per minute. As time goes by, it suddenly starts to rain, causing the water level to rise rapidly. The Page-Hinkley detection algorithm monitors this change trend in real time and exceeds the preset threshold. It automatically increases the sampling frequency to once every 30 seconds to collect data more frequently to ensure that the rapid rise in water level can be captured in time. When the rain stops, the water level gradually stabilizes, and the data change trend is evaluated again. If it is lower than the threshold at this time, the sampling frequency is restored to once a minute.

[0084] Through the above method, it is possible to capture these changes in time when the water level changes significantly, and reduce the frequency of data collection when the water level is relatively stable, thereby ensuring data quality while optimizing resource utilization. This not only improves the efficiency of data collection, but also reduces unnecessary data processing burden.

[0085] In step S20, the preprocessed data is input into the trained key factor determination model to obtain key data types and corresponding data including:

[0086] Step S21, input the preprocessed data into the trained LASSO regression sub-model to obtain the first key data type of preliminary screening;

[0087] Step S22: input the first key data type into the tree sub-model, obtain the feature importance score corresponding to each first key data type, and obtain the first key data types with the top 10% scores as the final key data types.

[0088] The LASSO regression submodel is a linear regression model that compresses unimportant feature coefficients to zero by adding a λ regularization term to the loss function, thereby achieving feature selection. In this embodiment, the LASSO regression submodel is used to screen out preliminary key data types from the preprocessed data.

[0089] A tree sub-model refers to a collection of decision tree models, such as random forests or gradient boosted trees. These models are able to assign an importance score to each key data type through multiple iterations and ensemble learning. The higher the score, the greater the importance of the data type in predicting the event type. Through the tree sub-model, the first key data type with the top 10% score is selected as the final key data type.

[0090] For example, suppose the LASSO regression sub-model selects the following features: water level, rainfall, and power generation. These features are then fed into the tree sub-model and trained to obtain feature importance scores. Suppose the final score ranking is: water level v rainfall > power generation > water temperature.

[0091] Finally, the top 10% of features are selected as the final key data type based on the scores, and the final key data type is "water level".

[0092] In this embodiment, before step S21, the following steps are also included:

[0093] Step S21a, collecting historical event data, using the historical event data to train the LASSO regression sub-model, and obtaining a trained LASSO regression sub-model;

[0094] Historical event data includes:

[0095] The hydrological data, meteorological data, power data and engineering safety monitoring data corresponding to the historical events.

[0096] In this embodiment, the historical event data is classified according to the event type, and a label is assigned to each historical event data based on the classification result;

[0097] Each data in the classified historical event data is input into the LASSO regression sub-model, and the cross entropy function and Adam optimizer are used with a learning rate of 0.001 until the cross entropy regression loss function is minimized to obtain the trained LASSO regression sub-model.

[0098] In this embodiment, in step S20, determining whether an abnormal event occurs and the type of the abnormal event according to the key data type and the corresponding data includes:

[0099] Input key data types and corresponding data into the abnormal event determination model to determine whether an abnormal event occurs and obtain the type of the abnormal event;

[0100] The abnormal event determination model is a convolutional neural network model; the convolutional neural network model includes: an input layer, a hidden layer, and an output layer; wherein the input layer inputs key data types and corresponding data; the hidden layer includes multiple neurons, which are used to determine whether an abnormal event occurs and obtain the type of abnormal event through a sigmoid activation function; and the output layer outputs the type of abnormal event.

[0101] In step S30,

[0102] The intelligent diagnosis system comprises:

[0103] Reservoir safety operation risk assessment models, water resources efficient utilization economic evaluation models, river flooding analysis models and water body ecological environment evaluation models provided by different model providers.

[0104] In this embodiment, step S40 includes:

[0105] Step S41, conduct a SWOT analysis on each disposal plan, including determining its strengths, weaknesses, opportunities and threats;

[0106] Step S42, assigning points to the advantages, disadvantages, opportunities and threats of each disposal solution, where 0 means that the factor does not exist or has the smallest impact, and 5 means that the factor has the greatest impact;

[0107] Step S43: Add the scores of the advantages, disadvantages, opportunities and threats to obtain the total score of each solution, and select the solution with the highest total score as the final solution.

[0108] Specifically, a method for handling events that affect the scheduling and operation of cascade reservoirs in this embodiment also includes:

[0109] Step S50: construct an event library, and save the real-time monitoring data, each event type and its corresponding key data type, and the final processing result corresponding to each event type into the event library.

[0110] In the specific implementation process, the real-time monitoring data, each event type and its corresponding key data type, and the final disposal results corresponding to each event type are collected and sorted. By saving the real-time monitoring data and the obtained results data in the event library, unified management and analysis of events are achieved, which can provide reference and support for future scheduling operations.

[0111] The present embodiment provides a method for handling events that affect the operation of cascade reservoirs. By organizing and summarizing data, a systematic event library is constructed to achieve unified management and analysis of events. By constructing a model solution and simulating the solution, it is ensured that a quick response can be made when an event occurs, thereby reducing the impact of the event on the operation of the reservoir. At the same time, the present invention improves the accuracy of the event handling results through SWOT analysis, and solidifies the business scenario model and platform capabilities of the event corresponding to the results, providing reference and support for future operation. Through the above method, not only the safety and reliability of the operation of cascade reservoirs are improved, but also the operation and management level of cascade reservoirs is improved, which has important practical value and promotion significance.

[0112] In the description of the present invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0113] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0114] In the present invention, unless otherwise clearly specified and limited, when a first feature is “on” or “below” a second feature, it may be that the first and second features are in direct contact, or that the first and second features are in indirect contact through an intermediate medium. Moreover, when a first feature is “above”, “above” or “above” a second feature, it may be that the first feature is directly above or obliquely above the second feature, or it may simply mean that the first feature is higher in level than the second feature. When a first feature is “below”, “below” or “below” a second feature, it may be that the first feature is directly below or obliquely below the second feature, or it may simply mean that the first feature is lower in level than the second feature.

[0115] In the description of this specification, the description of the terms "one embodiment", "some embodiments", "embodiment", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0116] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may alter, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A method for handling events that affect the operation of cascade reservoirs, characterized in that: include: S10, obtaining real-time monitoring data of the cascade reservoirs to be tested; Performing data preprocessing on the real-time monitoring data of the cascade reservoirs to be detected to obtain preprocessed data; The real-time monitoring data includes: hydrological data, meteorological data, power data and engineering safety data; S20, inputting the preprocessed data into the trained key factor determination model to obtain key data types and corresponding data; and determining whether an abnormal event occurs and the type of abnormal event according to the key data types and corresponding data; The key factor determination model includes: a LASSO regression sub-model and a tree sub-model; S30, when an abnormal event occurs, the type of the abnormal event is input into a pre-set intelligent diagnosis system to obtain a variety of abnormal handling solutions; S40: Conduct a SWOT analysis on the various abnormality handling solutions to obtain a final handling result.

2. The solution for handling events affecting the operation of cascade reservoirs according to claim 1 is characterized in that: In S10, the real-time monitoring data of the cascade reservoir to be detected is preprocessed, and the preprocessed data is obtained including: S11, cleaning the real-time monitoring data of the cascade reservoirs to be detected to obtain cleaned data; The data cleaning includes: missing value processing and outlier monitoring; S12, inputting the cleaned data into a Kalman filter, performing real-time denoising and smoothing processing, and obtaining pre-processed data.

3. The method for handling events affecting the operation of cascade reservoirs according to claim 1 is characterized in that: The S10 further includes: Use Page-Hinkley detection to evaluate the time series characteristics of the real-time monitoring data in real time and determine whether the change trend of the real-time monitoring data is greater than a threshold set in advance; If the change trend of the real-time monitoring data is greater than a pre-set threshold, increasing the sampling frequency of the real-time monitoring data; If the change trend of the real-time monitoring data is less than or equal to a preset threshold, the sampling frequency of the real-time monitoring data is reduced.

4. The method for handling events affecting the operation of cascade reservoirs according to claim 1, characterized in that: In S20, the pre-processed data is input into the trained key factor determination model, and the key data type and corresponding data are obtained, including: S21, inputting the preprocessed data into the trained LASSO regression sub-model to obtain the first key data type that is initially screened; S22. Input the first key data types into the tree sub-model, obtain the feature importance score corresponding to each first key data type, and obtain the first key data types with the top 10% scores as the final key data types.

5. The method for handling events affecting the operation of cascade reservoirs according to claim 4, characterized in that: The S21 also includes: S21a, collecting historical event data, using the historical event data to train the LASSO regression sub-model, and obtaining a trained LASSO regression sub-model; The historical event data includes: The hydrological data, meteorological data, power data and engineering safety monitoring data corresponding to the historical events.

6. The method for handling events affecting the operation of cascade reservoirs according to claim 5, characterized in that: The S21a specifically includes: Classifying the historical event data according to event types, and assigning labels to each historical event data based on the classification results; Each data in the classified historical event data is input into the LASSO regression sub-model, and the cross entropy function and Adam optimizer are used with a learning rate of 0.001 until the cross entropy regression loss function is minimized to obtain the trained LASSO regression sub-model.

7. The method for handling events affecting the operation of cascade reservoirs according to claim 1, characterized in that: In S20, determining whether an abnormal event occurs and the type of the abnormal event according to the key data type and the corresponding data includes: Input the key data type and corresponding data into the abnormal event determination model to determine whether an abnormal event occurs and obtain the type of the abnormal event; The abnormal event determination model is a convolutional neural network model; the convolutional neural network model includes: an input layer, a hidden layer, and an output layer; wherein the input layer inputs key data types and corresponding data; the hidden layer includes multiple neurons, which are used to determine whether an abnormal event occurs and obtain the type of the abnormal event through a sigmoid activation function; and the output layer outputs the type of the abnormal event.

8. The method for handling events affecting the operation of cascade reservoirs according to claim 1, characterized in that: In the S30, The intelligent diagnosis system comprises: Reservoir safety operation risk assessment models, water resources efficient utilization economic evaluation models, river flooding analysis models and water body ecological environment evaluation models provided by different model providers.

9. The method for handling events affecting the operation of cascade reservoirs according to claim 1, characterized in that: The S40 includes: S41. Conduct a SWOT analysis of each disposal option, including identifying its strengths, weaknesses, opportunities and threats; S42. Assign points to the strengths, weaknesses, opportunities and threats of each disposal option, where 0 means the factor does not exist or has the least impact, and 5 means the factor has the greatest impact; S43. Add up the scores of the advantages, disadvantages, opportunities and threats to obtain the total score of each solution, and select the solution with the highest total score as the final solution.

10. The solution for handling events affecting the operation of cascade reservoirs according to claim 1, characterized in that: The method further comprises: S50, constructing an event library, and storing the real-time monitoring data, each event type and its corresponding key data type, and the final processing result corresponding to each event type in the event library.

Images