A method for predicting events affecting cascade reservoir operation

Abstract

The present application relates to a kind of prediction methods of the event that influences cascade reservoir operation, comprising: obtaining the real-time monitoring data of the cascade reservoir to be detected, the real-time monitoring data of the cascade reservoir to be detected is input into water conservancy professional model, obtains the prediction data corresponding to the real-time monitoring data of the cascade reservoir to be detected;Obtain the prediction data of meteorology, the prediction data of meteorology and the prediction data corresponding to the real-time monitoring data of the cascade reservoir to be detected are input into trained data analysis model, obtain monitoring result;The monitoring result includes: key factor, event label and event occurrence probability;The data analysis model includes: key factor determination unit, event classification determination unit and event occurrence probability determination unit;Based on the monitoring result and pre-set early warning rule, event early warning is carried out.The present application can improve the safety and reliability of reservoir operation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of reservoir operation, and particularly relates to a method for predicting events affecting the operation of cascade reservoirs. BACKGROUND

[0002] During the operation of cascade reservoirs, reservoir management departments often face challenges from various unexpected events, such as natural disasters, equipment failures, etc. The occurrence of these events is often unpredictable, and the existing handling methods mainly rely on post-event recording and archiving, lacking systematic management and analysis means. This passive response not only cannot effectively prevent the occurrence of events, but also may lead to a lag in scheduling decisions, increasing the risk and uncertainty of reservoir operation.

[0003] Currently, the prediction of some events mainly depends on time factors, such as making predictions based on seasonal changes or historical data. However, this prediction method has low accuracy and cannot accurately predict the specific time and impact range of unexpected events, thus failing to provide strong support for arranging emergency preparedness work.

[0004] Therefore, there is an urgent need for a method for predicting events affecting the operation of cascade reservoirs. SUMMARY

[0005] (I) Technical problems to be solved

[0006] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present application provides a method for predicting events affecting the operation of cascade reservoirs, which solves the technical problem that the prior art cannot effectively prevent the occurrence of events and may lead to a lag in scheduling decisions, increasing the risk and uncertainty of the operation of cascade reservoirs.

[0007] (II) Technical solutions

[0008] In order to achieve the above-mentioned purposes, the main technical solutions adopted by the present application include:

[0009] The present application provides a method for predicting events affecting the operation of cascade reservoirs, comprising:

[0010] S100, acquiring real-time monitoring data of a cascade reservoir to be detected, inputting the real-time monitoring data of the cascade reservoir to be detected into a water conservancy professional model, and acquiring prediction data corresponding to the real-time monitoring data of the cascade reservoir to be detected;

[0011] The real-time monitoring data includes hydrological data, power data, and engineering safety monitoring data; and the water conservancy professional model includes a hydrological and hydrodynamic prediction sub-model, a business scenario prediction sub-model, a power market electricity load prediction sub-model, and a wind and light pumped storage prediction and dispatching sub-model.

[0012] S200, acquire the prediction data of the weather, input the prediction data of the weather and the prediction data corresponding to the real-time monitoring data of the cascade reservoir to be detected into the trained data analysis model, and acquire a monitoring result;

[0013] The monitoring result includes a key factor, an event label, and an event occurrence probability; and the data analysis model includes a key factor determination unit, an event classification determination unit, and an event occurrence probability determination unit.

[0014] S300, perform event early warning based on the monitoring result and a preset early warning rule.

[0015] Optionally, the S100 further includes the following steps:

[0016] S000, collect historical event data, train the data analysis model by using the historical event data, and acquire the trained data analysis model.

[0017] Optionally, the historical event data includes:

[0018] event type, event occurrence time, event occurrence location, event influence range, event control measure, weather data, hydrological data, power data, and engineering safety monitoring data corresponding to the event occurrence.

[0019] Optionally, the S000 specifically includes:

[0020] S010, classify the historical event data according to event type, event object, and event type name, and assign a label to each historical event data based on the classification result;

[0021] S020, input each data in the classified historical event data into a corresponding data analysis model, learn by using a cross-entropy function and an Adam optimizer, a learning rate is 0.001, until a cross-entropy regression loss function is minimized, and acquire the trained data analysis model.

[0022] Optionally, the S200 includes:

[0023] S210, input the prediction data of the weather and the prediction data corresponding to the real-time monitoring data of the cascade reservoir to be detected into the trained key factor determination unit, and acquire corresponding key factors;

[0024] S220, input the corresponding key factors into the event classification determination unit, and acquire an event label; the event label includes event type, event object, and event type name.

[0025] S230. Input the corresponding key factors and event labels into the event occurrence probability determination unit to obtain the event occurrence probability.

[0026] Optionally, in S210, the corresponding key factor is:

[0027]

[0028] Among them, Key Factors are the key factors, X n For each type of predicted data, For each type of predicted data, the regression coefficient estimate is... denoted as the standard error for each prediction; T represents the confidence level.

[0029] Optionally, in S220,

[0030] The event classification determination unit is a convolutional neural network model; the event classification determination unit includes: an input layer, a hidden layer, and an output layer; wherein, the input layer inputs the key factors corresponding to the predicted data; the hidden layer includes multiple neurons, used to obtain the event labels corresponding to the predicted data through the sigmoid activation function; the output layer outputs the event labels corresponding to the predicted data.

[0031] Optionally, in S230,

[0032] The event occurrence probability unit includes multiple event occurrence probability sub-units;

[0033] Specifically, based on the event tag, a corresponding event occurrence probability subunit is obtained; the event occurrence probability subunit is:

[0034]

[0035] Where β0 is the intercept term constant, β i For the i-th key factor K i The coefficient, γ i For the i-th key factor K i The coefficient of the quadratic term, δ ij Let be the interaction coefficient between the i-th and j-th key factors, u be the random effects constant, and K be the coefficient of the interaction term. i For the i-th key factor, K j Let Y be the j-th key factor, m be the total number of key factors, and P(Y = 1 | K1, K2, ..., K) be the key factor. m ) represents the probability of the event occurring.

[0036] Optionally, the method further includes:

[0037] The periodic early warning is set according to the periodic events of the dry season, the wet season and the normal water period; and the periodic early warning is determined according to the start and end time of the dry season, the wet season and the normal water period.

[0038] (III) Beneficial Effects

[0039] The beneficial effects of the present application are that the prediction method for events affecting cascade reservoir operation can predict and warn in advance of possible events, provide sufficient events for emergency preparation for management personnel, reduce the impact of events on reservoir operation, and improve the safety and reliability of cascade reservoir operation, compared with the prior art; at the same time, the subsequent model is trained by using historical data, which provides a systematic and scientific event identification and prediction method for the reservoir management department, and has important practical value and popularization significance. BRIEF DESCRIPTION OF DRAWINGS

[0040] Figure 1 The flowchart of the prediction method for events affecting cascade reservoir operation of the present application embodiment 1 is shown.

[0041] Figure 2 The schematic diagram of different time components of task execution in the periodic early warning in the present application embodiment 2. DETAILED DESCRIPTION

[0042] In order to better explain the present application, the following will be described in detail by specific embodiments in combination with the drawings.

[0043] The prediction method for events affecting cascade reservoir operation proposed in the present application embodiment solves the technical problems that the prior art cannot effectively prevent the occurrence of events, and may also cause the lag of scheduling decision, increase the risk and uncertainty of cascade reservoir operation, improves the safety and reliability of cascade reservoir operation, and provides a systematic and scientific event identification and prediction method for the reservoir management department, which has important use value and popularization significance.

[0044] In order to better understand the above technical solutions, the following will describe the exemplary embodiments of the present application in more detail with reference to the drawings. Although the exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application 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, more thorough understanding of the present application and to convey the complete scope of the present application to those skilled in the art.

[0045] DETAILED DESCRIPTION

[0046] Embodiment 1

[0047] Referring to Figure 1 A prediction method for events affecting cascade reservoir operation in an embodiment of the present application comprises:

[0048] Step S100, real-time monitoring data of a cascade reservoir to be detected is acquired, the real-time monitoring data of the cascade reservoir to be detected is input into a water conservancy professional model, and prediction data corresponding to the real-time monitoring data of the cascade reservoir to be detected is acquired;

[0049] The real-time monitoring data includes hydrological data, power data, and engineering safety monitoring data; and the water conservancy professional model includes a hydrological and hydrodynamic prediction sub-model, a business scenario forecasting sub-model, a power market electricity load prediction sub-model, and a wind and light pumped storage forecasting and dispatching sub-model.

[0050] Step S200, prediction data of weather is acquired, the prediction data of the weather and the prediction data corresponding to the real-time monitoring data of the cascade reservoir to be detected are input into a trained data analysis model, and a monitoring result is acquired;

[0051] The monitoring result includes key factors, event labels, and event occurrence probabilities; and the data analysis model includes a key factor determination unit, an event classification determination unit, and an event occurrence probability determination unit.

[0052] Step S300, event warning is performed based on the monitoring result and a pre-set warning rule.

[0053] In a specific implementation process, the hydrological data includes water level, hydrological station monitoring flow, reservoir inflow and outflow, water quality parameters, etc., the power data includes power generation, power generation efficiency, power transmission, load condition, etc., and the engineering safety monitoring data includes displacement data, seepage data, crack data, stress and strain data, vibration data, etc. The above data is acquired through various sensors and monitoring equipment installed in the cascade reservoir basin.

[0054] The prediction data of the weather includes prediction data of temperature, humidity, wind speed, wind direction, rainfall, etc., which is acquired through weather forecasting.

[0055] In this embodiment, the data analysis model including the key factor determination unit, the event classification determination unit, and the event occurrence probability determination unit is acquired by training historical data, and the prediction data corresponding to the real-time monitoring data is input into the data analysis model, so as to predict the events affecting the cascade reservoir operation. The occurrence probability of the event can be predicted in advance, and when the occurrence probability of the event is too high, a corresponding response plan can be quickly formulated, so that the event can be disposed quickly and effectively when it occurs, thereby reducing the impact of the sudden event on the reservoir operation and improving the safety and reliability of the cascade reservoir operation.

[0056] Embodiment 2

[0057] The prediction method of the event affecting the cascade reservoir operation in the embodiment comprises:

[0058] In step S100, real-time monitoring data of the cascade reservoir to be detected is acquired, the real-time monitoring data of the cascade reservoir to be detected is input into a water conservancy professional model, and prediction data corresponding to the real-time monitoring data of the cascade reservoir to be detected is acquired.

[0059] The real-time monitoring data comprises hydrological data, power data and engineering safety monitoring data; and the water conservancy professional model comprises a hydrological and hydrodynamic prediction submodel, a business scenario forecasting submodel, a power market electricity load prediction submodel and a wind and light pumped storage forecasting and dispatching submodel.

[0060] In step S200, prediction data of the weather is acquired, the prediction data of the weather and the prediction data corresponding to the real-time monitoring data of the cascade reservoir to be detected are input into a trained data analysis model, and a monitoring result is acquired.

[0061] The monitoring result comprises a key factor, an event label and an event occurrence probability; and the data analysis model comprises a key factor determination unit, an event classification determination unit and an event occurrence probability determination unit.

[0062] In step S300, event warning is performed based on the monitoring result and a pre-set warning rule.

[0063] In the embodiment, before step S100, the following step is further included:

[0064] In step S000, historical event data is collected, the data analysis model is trained by using the historical event data, and a trained data analysis model is acquired.

[0065] In the specific implementation process, the historical event data comprises:

[0066] an event type, an event occurrence time, an event occurrence location, an event influence range, an event control measure, weather data corresponding to the event occurrence, hydrological data, power data and engineering safety monitoring data.

[0067] In the embodiment, after the historical event data is collected, the historical event data is input into a pre-set database to form a historical event database.

[0068] In the specific implementation process, the event can be traced back according to the selected historical event, the event process change rule can be intuitively displayed by using icons, and after a certain event is selected, weather data, hydrological data, power data and engineering safety monitoring data corresponding to the event period can be viewed.

[0069] In this embodiment, step S000 specifically includes:

[0070] Step S010, classify the historical event data according to the event type, the object to which the event belongs, and the event type name, and assign labels to each historical event data based on the classification results;

[0071] In the specific implementation process, a three-level tree is used to classify events, and the event type is divided into typical events and daily events. The object to which the event belongs is the reservoir where the event occurs, and the event type name is the specific occurrence. After classification, the event type, the object to which the event belongs, and the event type name are all labeled in the form of labels on the corresponding historical event data.

[0072] For example, assume that the reservoir downstream door and window vibration occurs, at this time, first determine the event type as a typical event, then determine the object to which the event belongs, that is, the location where the event occurs, and finally determine the event type name as door and window vibration. At this time, the meteorological data, hydrological data, power data, and engineering safety monitoring data corresponding to the event are all labeled with the labels of "typical data, a reservoir (event location), downstream door and window vibration".

[0073] Step S020, input each data in the classified historical event data into the corresponding data analysis model, use cross-entropy function and Adam optimizer, learning rate is 0.001, until the cross-entropy regression loss function is minimized, and obtain the trained data analysis model.

[0074] In this embodiment, step S200 includes:

[0075] Step S210, input the prediction data of the meteorological prediction data and the real-time monitoring data of the to-be-detected cascade reservoir into the trained key factor determination unit, and obtain the corresponding key factor;

[0076] Step S220, input the corresponding key factor into the event classification determination unit to obtain the event label;

[0077] The event label includes: event type, object to which the event belongs, and event type name;

[0078] Step S230, input the corresponding key factor and event label into the event occurrence probability determination unit to obtain the event occurrence probability.

[0079] In this embodiment, in step S210, the corresponding key factor is:

[0080]

[0081] Where, Key Factors is the key factor, Xn for each prediction data, for each prediction data, for each prediction data; T is the confidence level.

[0082] for each prediction data,

[0083]

[0084] for each prediction data, for each prediction data, σ 2 is the residual variance, X n is each real-time prediction data, X n is the sample mean corresponding to each prediction data.

[0085] For example, for each prediction data X n , its regression coefficient estimate and standard error If in the prediction data, the regression coefficient estimate of water level is the standard error is and the set confidence level T is 1.96 (corresponding to a 95% confidence interval), then At this time, the water level is the key factor.

[0086] In step S220,

[0087] The event classification determination unit is a convolutional neural network model; the event classification determination unit includes an input layer, a hidden layer, and an output layer; wherein the input layer inputs the key factors corresponding to the prediction data; the hidden layer includes a plurality of neurons, which are used to obtain the event label corresponding to the prediction data through a sigmoid activation function; and the output layer outputs the event label corresponding to the prediction data.

[0088] In this embodiment, in step S230, the event occurrence probability unit includes a plurality of event occurrence probability sub-units;

[0089] wherein the corresponding event occurrence probability sub-unit is obtained according to the event label; the event occurrence probability sub-unit is:

[0090]

[0091] wherein β0is the constant term, β i is the coefficient of the i-th key factor K i , γ i is the quadratic term coefficient of the i-th key factor K i , δ ij is the interaction term coefficient between the i-th and j-th key factors, and u is a random effect constant, Ki K1 is the i-th key factor j Kj is the j-th key factor, m is the total number of key factors, P(Y=1|K1, K2, …, K m m) is the probability of event occurrence.

[0092] In this embodiment, the pre-set early warning rule is:

[0093] When the probability of event occurrence is greater than or equal to 80%, first-level early warning is performed; when the probability of event occurrence is greater than or equal to 50% and less than 80%, second-level early warning is performed; when the probability of event occurrence is less than 50%, no early warning is performed; the first-level early warning is sound and light early warning, and the second-level early warning is sound early warning.

[0094] For example, it is assumed that the key factors of an event are water level change rate, rainfall of catchment area hydrological station and upstream inflow through the trained key factor determination unit, and at this time, the corresponding water level change rate is 0.03 m / h, the upstream rainfall is 30 mm / h, and the upstream inflow is 120 m 3 / s, that is, K1=30 mm / h, K2=30 mm / h and K3=120 m 3 / s.

[0095] At the same time, the following parameters have been obtained through historical data analysis and model training:

[0096] The intercept constant β0 is-4, the first-order term coefficients are β1=0.03, β2=0.03 and β3=0.0001, the second-order term coefficients are γ1=0.0003, γ2=0.0003 and γ3=0.000002, the interaction coefficients are δ 12 =0.00003, δ 13 =0.0000002 and δ 23 =0.0000002, and the random effect constant u can be ignored or set to 0;

[0097] The above parameters are substituted into the formula to obtain the final probability of event occurrence, which is 79%, and at this time, second-level early warning is performed.

[0098] In the specific implementation process, the harm degree of the event also needs to be considered.

[0099] If it is determined that the probability of event occurrence is greater than 80%, the harm degree of the event is considered; in specific application, the harm degree of the event is obtained by multiplying the scores of the possibility of event occurrence, exposure frequency and severity. The harm degree of the event is divided into four levels, which are red early warning, orange early warning, yellow early warning and blue early warning, and correspond to different color flash early warnings.

[0100] In the specific implementation process, the method of the present application further comprises:

[0101] The periodic early warning is set according to the periodic events of the dry season, the wet season and the normal season; and the periodic early warning is determined according to the start and end time of the dry season, the wet season and the normal season.

[0102] The time early warning in the embodiment is configured according to the specification of the cron expression, and the frequency of the timing task execution. A standard cron expression contains at most 6 fields, and the fields represent different time components of the execution, such as Figure 2 as shown.

[0103] Each field has a set of allowed values, and the values indicate the time unit to be executed. A single value, a series of values or a wildcard can be used to specify. For example, the allowed values of the minute field are 0-59, the allowed values of the hour field are 0-23, the allowed values of the day field are 1-31, and the allowed values of the month field are 1-12.

[0104] In addition, in the cron expression, "*" represents all possible values, such as setting "*" on the minute field to trigger every minute.

[0105] "," can be used to separate multiple values, such as setting "1, 3, 4, 7" on the hour field to trigger at 1 o'clock, 3 o'clock, 4 o'clock and 7 o'clock respectively.

[0106] "-" can be used to represent a range, such as setting "5-10" on the day field to trigger every day from 5th to 10th.

[0107] " / " can be used to represent a step, such as setting "0 / 15" on the minute field to trigger every 15 minutes, and "* / 15" to trigger every 15 minutes from 0.

[0108] "?" can only be used on the date and week fields, and indicates that the value is not specified, and can be left blank, which is actually equivalent to a placeholder.

[0109] "#" can only be used on the week field, and indicates the number of the week, such as "6#3" indicating the third Friday (6 represents Friday, and 3 represents the third).

[0110] For example, the official flood season of a reservoir is from May 1st to October 31st every year, and at 8 o'clock in the morning on May 1st every year, the staff can be reminded by sound and light early warning, short message, email and other ways that the flood season has begun; during the flood season, the staff is reminded every day at 8 o'clock in the morning to check the safety state and monitoring data of the reservoir by the above-mentioned ways; at 8 o'clock in the morning on October 31st every year, the staff is reminded again by the above-mentioned ways that the flood season is over, and the work achievements during the flood season are summarized.

[0111] Through setting the periodic early warning working mechanism, the necessary preventive measures and preparation of the staff at the key time node can be ensured, potential problems can be found and solved in time, and the probability of accidents can be reduced, so that the safety and operation efficiency of the cascade reservoirs can be effectively improved.

[0112] The prediction method for events affecting the cascade reservoir operation of the embodiment can realize unified management and analysis of events, and can significantly provide a scientific data analysis model for event identification and prediction, so that the staff can be provided with sufficient event emergency preparation before the event occurs, the influence of the event on the reservoir operation is reduced, and the prediction method has important use value and popularization significance.

[0113] In the description of the present application, it should be understood that the terms "first", "second" are only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features limited by "first", "second" can be explicitly or implicitly included one or more of the features. In the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise specifically limited.

[0114] In the present application, unless otherwise specifically defined and limited, the terms "mounting", "connection", "connection", "fixing" and the like should be understood in a broad sense, for example, it can be fixed connection, or detachable connection, or integral; it can be mechanical connection, or electrical connection; it can be directly connected, or indirectly connected through intermediate medium; it can be the internal communication of two elements or the interaction relationship between two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0115] In the present application, unless otherwise specifically defined and limited, the first feature is "on" or "under" the second feature, which can be direct contact between the first and second features, or indirect contact between the first and second features through intermediate medium. Moreover, the first feature is "above", "over" and "on" the second feature, which can be directly above or obliquely above the first feature, or only indicates that the horizontal height of the first feature is higher than that of the second feature. The first feature is "below", "under" and "under" the second feature, which can be directly below or obliquely below the first feature, or only indicates that the horizontal height of the first feature is lower than that of the second feature.

[0116] In the description of the specification, the description of the terms "one embodiment", "some embodiments", "an embodiment", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are contained in at least one embodiment or example of the present application. In the specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples. In addition, the person skilled in the art can combine and combine the different embodiments or examples described in the specification and the features of the different embodiments or examples without contradiction.

[0117] Although the embodiments of the present application have been shown and described above, it is understood that the above-described embodiments are exemplary and are not to be construed as limiting the present application, and the person skilled in the art can modify, modify, replace and modify the above-described embodiments within the scope of the present application.

Claims

1. A method for predicting events affecting the operation and scheduling of cascade reservoirs, characterized in that, include: S100. Obtain real-time monitoring data of the cascade reservoirs to be tested, input the real-time monitoring data of the cascade reservoirs to be tested into the water conservancy professional model, and obtain the predicted data corresponding to the real-time monitoring data of the cascade reservoirs to be tested. The real-time monitoring data includes: hydrological data, power data, and engineering safety monitoring data; the water conservancy professional model includes: hydrological and hydrodynamic prediction sub-model, operational scenario forecasting sub-model, power market load forecasting sub-model, and wind-solar-pumped storage forecasting and scheduling sub-model. S200. Obtain meteorological forecast data, input the meteorological forecast data and the forecast data corresponding to the real-time monitoring data of the cascade reservoirs to be monitored into the trained data analysis model, and obtain the monitoring results; the data analysis model includes: a key factor determination unit, an event classification determination unit, and an event occurrence probability determination unit; S200 includes: S210. Input the predicted meteorological data and the predicted data corresponding to the real-time monitoring data of the cascade reservoirs to be tested into the trained key factor determination unit to obtain the corresponding key factors. In S210, the corresponding key factor is: ； Among them, Key Factors are the key factors, X n For each type of predicted data, For each type of predicted data, the regression coefficient estimate is... Here, represents the standard error of each predicted data point; T represents the confidence level. S220. Input the corresponding key factors into the event classification and determination unit to obtain event tags; the event tags include: event type, event object, and event type name; The event classification determination unit is a convolutional neural network model; the event classification determination unit includes: an input layer, a hidden layer, and an output layer; wherein, the input layer inputs the key factors corresponding to the predicted data; the hidden layer includes multiple neurons, used to obtain the event labels corresponding to the predicted data through the sigmoid activation function; the output layer outputs the event labels corresponding to the predicted data; S230. Input the corresponding key factors and event tags into the event occurrence probability determination unit to obtain the event occurrence probability; The event occurrence probability determination unit includes multiple event occurrence probability determination subunits; Specifically, based on the event tag, a corresponding event occurrence probability determination subunit is obtained; the event occurrence probability determination subunit is: ； Where β0 is the intercept term constant, β i For the i-th key factor K i The coefficient, γ i For the i-th key factor K i The coefficient of the quadratic term, Let be the interaction coefficient between the i-th and j-th key factors, u be the random effects constant, and K be the coefficient of the interaction term. i For the i-th key factor, K j Let m be the j-th key factor, and m be the total number of key factors. The probability of an event occurring; S300. Based on the monitoring results and the pre-set early warning rules, issue an event warning.

2. The method for predicting events affecting the operation and scheduling of cascade reservoirs according to claim 1, characterized in that, The process preceding S100 also includes: S000. Collect historical event data, use the historical event data to train the data analysis model, and obtain the trained data analysis model.

3. The method for predicting events affecting the operation and scheduling of cascade reservoirs according to claim 2, characterized in that, The historical event data includes: Event type, event time, event location, event impact range, event control measures, meteorological data, hydrological data, power data, and engineering safety monitoring data at the time of the event.

4. The method for predicting events affecting the operation and scheduling of cascade reservoirs according to claim 2, characterized in that, The S000 specifically includes: S010. Classify the historical event data according to event type, event belonging object, and event type name, and assign labels to each historical event data based on the classification results; S020. Input each data point in the classified historical event data into the corresponding data analysis model, use the cross-entropy function and Adam optimizer with a learning rate of 0.001 until the cross-entropy regression loss function is minimized, and obtain the trained data analysis model.

5. The method for predicting events affecting the operation and scheduling of cascade reservoirs according to claim 1, characterized in that, The method further includes: Regular early warnings are set up based on the periodic events of dry season, wet season, and normal water season; the regular early warnings are determined according to the start and end times of the dry season, wet season, and normal water season.

Images