Flood control scheduling method for cascade reservoirs

Abstract

The invention discloses a flood control scheduling method for cascade reservoirs, and relates to the technical field of reservoir scheduling. Comprising the following steps: acquiring characteristic parameters of a target flood and a plurality of case floods, and selecting a typical flood most matched with the target flood according to the characteristic parameters of the target flood and the characteristic parameters of the case floods; traversing the operation data of the cascade reservoirs under the typical flood period by period from the first period; if the gate opening degrees of the cascade reservoirs in the adjacent time periods change, integrating the average reservoir inflow of the cascade reservoirs in the next time period, the current reservoir water level and the gate opening degrees into flood control knowledge, and storing the flood control knowledge into a flood control knowledge base; determining a scheduling scheme of the target flood cascade reservoir according to flood control knowledge of typical flood in the flood control knowledge base; and summarizing the flood control scheduling schemes of all the time periods to obtain flood control scheduling schemes of the cascade reservoir in all the time periods under the target flood. According to the invention, efficient deduction can be realized, and the flood control scheduling decision period is remarkably shortened.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of reservoir regulation, in particular to a flood control regulation method of cascade reservoirs. BACKGROUND

[0002] Formulating a flood control regulation scheme of cascade reservoirs is a complex and multi-level process, which needs to comprehensively consider multiple factors and targets. First, cascade reservoirs are usually equipped with multiple flood discharge facilities, such as middle holes, surface holes, spillways and flood discharge tunnels, and the selection of the opening degree of different gates also varies, and the reasonable operation of the flood discharge facilities is the key to flood control regulation; second, the discharge capacity of the gate is usually not linear, which depends not only on the opening degree of itself, but also on the current water level of the reservoir; third, frequent gate operation not only increases the operation difficulty, but also may cause operation errors, in order to reduce the complexity and error risk of operation, the gate operation should be reduced as much as possible; in addition, after the formulation of the flood control regulation scheme, the reservoir needs to be checked and confirmed according to the regulations before the gate operation is performed, and the downstream relevant units need to be notified in advance, so the operation command of the gate must be issued in advance.

[0003] The traditional flood control regulation method mainly relies on manual decision-making, which needs to be repeated by manual calculation, and is time-consuming and prone to errors. Especially in the case of extreme weather events leading to large changes in flood peaks, the traditional manual calculation method is not up to the task. SUMMARY

[0004] Therefore, it is necessary to provide a flood control regulation method of cascade reservoirs aiming at the above technical problems.

[0005] The present application provides a flood control regulation method of cascade reservoirs, comprising: obtaining the characteristic parameters of the target flood and the multiple case floods, determining the matching degrees of the target flood and the case floods according to the characteristic parameters of the target flood and the characteristic parameters of the case floods, and taking the case flood with the highest matching degree as a typical flood; starting from the first period, traversing the operation data of the cascade reservoir under the typical flood period by period, judging whether the gate opening degree of the adjacent period of the cascade reservoir changes; if yes, determining the average inflow of the cascade reservoir in the future period, and integrating the average inflow of the cascade reservoir in the future period, the reservoir water level and the gate opening degree into the flood control knowledge and storing it into the flood control knowledge base; if not, continue to traverse the operation data of the cascade reservoir in the next period under the typical flood; and The average inflow of the cascade reservoir under the target flood in a time period is obtained, and a most matched flood control knowledge is obtained from the flood control knowledge base according to the average inflow of the cascade reservoir under the target flood in a time period; a flood control scheduling scheme of a current time period is obtained by performing flood control scheduling on the target flood according to the most matched flood control knowledge; whether the total number of time periods for which the flood control scheduling scheme has been obtained is equal to the total number of time periods of the typical flood is determined; if yes, all time periods of the target flood are determined, and the flood control scheduling schemes of each time period are summarized to obtain the flood control scheduling scheme of the cascade reservoir under the target flood in all time periods; if no, the average inflow of the cascade reservoir under the target flood in a next time period is determined.

[0006] Optionally, the average inflow of the cascade reservoir in a future time period, the reservoir water level and the gate opening are integrated into the flood control knowledge and stored in the flood control knowledge base, and specifically comprising: The time corresponding to the first flood control knowledge is taken as a starting data point, and the operation data of the cascade reservoir under the typical flood is traversed time by time, and whether the gate opening of adjacent time periods changes is determined; if yes, the average inflow of the cascade reservoir in a next time period is determined according to the typical flood data, and a flood control knowledge is constructed according to the average inflow of the cascade reservoir in the next time period, the current reservoir water level and the gate opening and stored in the flood control knowledge base; if no, the operation data of the cascade reservoir under the next time period is continuously traversed; The storage structure of the flood control knowledge base is determined based on the following formula: Flood control knowledge i : average inflow of the cascade reservoir in a next time period, reservoir water level, gate opening; Wherein, i Indicates the serial number of the flood control knowledge.

[0007] Optionally, when all time periods of the target flood are determined, the flood control scheduling schemes of each time period are summarized to obtain the flood control scheduling scheme of the cascade reservoir under the target flood in all time periods, and specifically comprising: A current flood control knowledge with the most matched average inflow is obtained from the knowledge base according to the average inflow of the cascade reservoir under the target flood in a time period; when the cascade reservoir is opened, the current flood control knowledge is searched from the cascade reservoir opening time period, and the current flood control knowledge includes the gate opening corresponding to the water level and the inflow; The water abandonment time period of the cascade reservoir is determined, and the water abandonment time period is a time period in which the cascade reservoir must be discharged when the water level exceeds the limit only for power generation without opening; the flood control scheduling scheme of a current time period is determined by performing flood control scheduling on the target flood based on the most matched current flood control knowledge; determining whether the total number of the current period is equal to the total number of periods of the target flood; if yes, all periods of the target flood are determined, and the flood control scheduling scheme of each period is summarized to obtain the flood control scheduling scheme of the cascade reservoirs in all periods under the target flood, the flood control scheduling scheme including the discharge flow, water level process, power generation flow, flood discharge flow and gate opening; if no, the average inflow of the cascade reservoirs in the next period under the target flood is determined.

[0008] Optionally, based on the most matched flood control knowledge, the flood control scheduling of the target flood is performed to determine the flood control scheduling scheme of the current period, specifically including: If the cascade reservoir does not discharge, and when the actual water level difference of the reservoir is less than a set threshold or exceeds the water level of the matched knowledge in the knowledge base, the matched gate opening is not all 0, and the current period is greater than or equal to 4 periods from the starting period or is within 3 periods from the water abandonment period, the discharge is started, the gate opening is set to the gate opening recommended by the flood control knowledge, and the future operating state of the cascade reservoir is predicted; If the cascade reservoir does not discharge, and the gate opening recommended by the flood control knowledge is all 0, the discharge is not performed, the gate opening is set to 0, and the future operating state of the cascade reservoir is predicted; If the cascade reservoir is discharging, the last gate opening time is found, and whether the current gate opening is maintained is determined according to the gate opening time interval; If the cascade reservoir is discharging, the last gate opening time is found, and whether the current gate opening is maintained is determined according to the gate opening time interval; If the discharge continues, whether the gate opening is changed is determined, if the water level difference from the water level of the flood control knowledge is less than a set threshold, and compared with maintaining the original gate opening, the difference between the discharge flow and the inflow corresponding to the flood control knowledge is less than a set threshold, the gate opening is updated, and the current operating state of the cascade reservoir is determined according to the gate opening and the output.

[0009] Optionally, the current operating state of the cascade reservoir is determined according to the gate opening and the output, specifically including: The discharge flow is determined according to the gate opening and the water level; The tail water level is determined according to the output and the discharge flow, and the power generation flow is obtained; The reservoir capacity is determined through the water balance equation according to the discharge flow, the power generation flow and the inflow, so as to update the reservoir water level; The current operating state of the cascade reservoir is determined according to the updated reservoir water level and the gate opening.

[0010] Optionally, if the cascade reservoir is releasing flood, find the last time of gate operation, determine whether to keep the current gate opening according to the gate operation interval, including: Traverse the time sequence, find the first non-zero opening period, record the first opening time and gate opening; if no non-zero opening period is found, the same opening is not kept, then determine whether to stop releasing flood; If the current period is within the initial gate operation interval after the first opening, the same opening is kept, and the future operation state of the cascade reservoir is predicted; The last gate operation time is initialized as the first opening time plus the initial gate operation interval; from the current period, find the time point of the last gate opening change, and update the last gate operation time; Determine the interval between the current period and the last gate operation time, if less than the gate operation interval, keep the same opening, and predict the future operation state of the cascade reservoir; if more than the gate operation interval, do not keep the same opening, and determine whether to stop releasing flood.

[0011] Optionally, the future operation state of the cascade reservoir is predicted, including: According to the power generation flow and gate opening of the cascade reservoir in the current period, determine the reservoir water level in the next period; If the reservoir water level in the next period does not exceed the difference between the highest limit water level and the reserved margin, determine the current operation of the cascade reservoir according to the gate opening and output in the current period; If the reservoir water level in the next period exceeds the difference between the highest limit water level and the reserved margin, determine the power generation flow according to the non-water rejection mode, and determine the flood discharge flow according to the reservoir balance principle to obtain the intermediate gate opening; return the calculation time to the previous gate operation time, take the intermediate gate opening as the gate opening in the next period, and determine the future operation state of the cascade reservoir according to the gate opening and output in the next period.

[0012] The above-mentioned cascade reservoir flood control scheduling method provided by the embodiment of the application has the following advantages compared with the prior art: The application selects a typical flood through a flood characteristic parameter matching mechanism, extracts knowledge related to flood control in the typical flood for storage, and constructs a flood control knowledge base; determines the flood control scheduling scheme of the cascade reservoir in all periods under the target flood according to the flood control knowledge base, can solve the problem of long time-consuming and easy to make mistakes caused by repeated trial calculation by artificial, realize efficient deduction, and significantly shorten the flood control scheduling decision-making cycle.

[0013] In addition, in the case of rapid flood change, the optimal scheduling parameters are directly obtained by matching the flood control knowledge base, which can avoid the subjective bias of artificial experience, and the time-consuming of artificial decision-making is reduced from several hours to minutes. This not only significantly reduces the scheduling risk in extreme flood events, but also optimizes the construction of an intelligent decision-making system through continuous learning of the knowledge base, providing reliable technical support for flood control in the basin. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 A flowchart of a flood control scheduling method of a cascade reservoir provided in an embodiment; Figure 2 A flowchart of a generated solution of a flood control scheduling method of a cascade reservoir provided in an embodiment; Figure 3 A flowchart of a judgment of maintaining gate opening of a flood control scheduling method of a cascade reservoir provided in an embodiment; Figure 4 A schematic diagram of a cascade reservoir of a flood control scheduling method of a cascade reservoir provided in an embodiment; Figure 5 A target flood flow process diagram of a flood control scheduling method of a cascade reservoir provided in an embodiment; Figure 6 A reservoir A operation diagram under a typical flood of a flood control scheduling method of a cascade reservoir provided in an embodiment; Figure 7 A gate opening diagram of reservoir A under a typical flood of a flood control scheduling method of a cascade reservoir provided in an embodiment; Figure 8 A reservoir A operation diagram under a target flood of a flood control scheduling method of a cascade reservoir provided in an embodiment; Figure 9 A gate opening diagram of reservoir A under a target flood of a flood control scheduling method of a cascade reservoir provided in an embodiment; Figure 10 A reservoir B operation diagram under a typical flood of a flood control scheduling method of a cascade reservoir provided in an embodiment; Figure 11 A gate opening diagram of reservoir B under a typical flood of a flood control scheduling method of a cascade reservoir provided in an embodiment; Figure 12 A reservoir B operation diagram under a target flood of a flood control scheduling method of a cascade reservoir provided in an embodiment; Figure 13 A gate opening diagram of reservoir B under a target flood of a flood control scheduling method of a cascade reservoir provided in an embodiment. DETAILED DESCRIPTION

[0015] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0016] (1) In the research on reservoir optimization operation, the operation objectives are mainly based on reducing peak flow, staggering peak flow, etc., and using outflow and water level as decision variables to study the flood control strategy of the reservoir. It is less detailed to the opening and closing and adjustment of the gate, so it is often difficult to apply in actual application.

[0017] (2) The flood control scheduling of cascade reservoirs is solved using traditional optimization algorithms. The decision variables involve many objects such as reservoirs, gates, openings, time, etc. Dynamic programming is prone to fall into the curse of dimensionality, and intelligent optimization algorithms have the problem of "premature convergence". Therefore, existing research cannot fully consider the available opening of the gates.

[0018] (3) In existing studies, the power generation flow is mostly taken as the maximum reference flow of the turbine. However, in reality, after the power station is generating electricity at full load, the power generation flow is greatly affected by the reservoir water level and tailwater level. When the reservoir discharges flood water, the large change in tailwater level will lead to an increase in the power generation flow. When the reservoir water level increases rapidly during flood control, the power generation flow will decrease.

[0019] In one embodiment, a flood control scheduling method for cascade reservoirs is provided, the method comprising: S1. Select the typical flood that best matches the target flood from historical cases; The characteristic parameters of the target flood and multiple case floods are obtained, and the matching degree between the target flood and each case flood is determined according to the characteristic parameters of the target flood and the characteristic parameters of each case flood, and the case flood with the highest matching degree is taken as the typical flood.

[0020] S2. Knowledge extraction stage: determine the storage structure of flood control knowledge, create a flood control knowledge base, and automatically extract and store flood control knowledge from typical floods.

[0021] Starting from the first period, the operation data of the cascade reservoirs under typical floods are traversed period by period to determine whether the gate openings of the cascade reservoirs in adjacent periods have changed. If so, the average inflow of the cascade reservoirs in the next period is determined, and the average inflow, reservoir water level and gate opening of the cascade reservoirs in the next period are integrated into flood control knowledge and stored in the flood control knowledge base; if not, the operation data of the cascade reservoirs under typical floods in the next period are continued to be traversed.

[0022] S3. Knowledge application stage: Utilize the relevant knowledge contained in the knowledge base, combine it with the reservoir operation status under the target flood, and make decision interventions when necessary to intelligently generate reservoir flood control scheduling plans.

[0023] The average inflow of the target flood under the cascade reservoir in a time period is obtained, and the most matched flood control knowledge is obtained from the flood control knowledge base according to the average inflow of the target flood under the cascade reservoir in a time period. The flood control scheduling of the target flood is performed according to the most matched flood control knowledge, and the flood control scheduling scheme of the current time period is obtained. It is judged whether the total number of time periods for which the flood control scheduling scheme has been obtained is equal to the total number of time periods of the typical flood; if yes, all time periods of the target flood are determined, and the flood control scheduling schemes of each time period are summarized to obtain the flood control scheduling scheme of the target flood under the cascade reservoir in all time periods; if no, the average inflow of the target flood under the cascade reservoir in the next time period is determined.

[0024] Further, the S1 selects the most matched typical case from the historical cases according to the target flood, and there are various implementation methods, and the commonly used methods include matching based on feature similarity, matching based on text similarity, matching based on rules, machine learning model and multi-model hybrid method. Here, the patent takes the matching based on feature similarity as an example to introduce the steps of selecting the typical flood, which specifically includes,

[0025] S11, selecting feature information for case matching, such as flood peak flow, 12h flood volume, 1 day (24h) flood volume, 2 days (48h) flood volume, total process flood volume, average output, flood discharge, initial water level, etc. If it is a multi-peak flood, the flood peak and flood volume are the flood with the largest flood peak flow; S12, calculating the feature parameters of the target flood according to the selected feature information; S13, calculating the feature parameters of each typical case in the knowledge base according to the selected feature information; S14, calculating the matching degree of the target flood and each case flood based on the feature parameters; S15, sorting the matching degrees of each case flood to find the case flood with the highest matching degree as the typical flood; Further, the S2 automatically extracts and stores the flood control knowledge from the typical flood, specifically including, S21, determining the storage structure of the flood control knowledge base. The storage structure is:

[0026] Flood control knowledge i : average inflow of the next time period of the cascade reservoir, water level of the reservoir, gate opening; Wherein, i indicates the serial number of the flood control knowledge. The case flood and the target flood are both data records formed at an interval of 1 hour.

[0027] S22, traverse the operation data of the cascade reservoir under typical flood (such as reservoir inflow and outflow, water level, gate opening, etc.). The starting data point is the time corresponding to the first data, and the calculation is started from the first period and recursively backward, to detect whether the gate opening changes in the adjacent period. If it changes, jump to step S23; otherwise, continue to traverse the next period until all periods of the case flood are detected.

[0028] The specific process of traversal is as follows: the flood data in flood control is in segments, and the interval between periods is generally one hour. The length of the future period for calculating the average inflow is generally three hours. Starting from the first period, the calculation is pushed back to the next period after each period is detected, and the calculation is continued until the last period. In this process, the average inflow in the future period (such as three hours) is calculated continuously. If the remaining period is less than three hours, the average inflow of the remaining period is taken as the average inflow in the future period.

[0029] S23, if the gate opening changes, record the corresponding period, and extract flood control knowledge according to the gate operation, mainly including the following steps: S231, according to the case flood, calculate the average inflow (actual data) of the typical flood in the future period (such as the future three hours); S232, create a new flood control knowledge, and take the average inflow of the typical flood in the future period (such as the future three hours), the current reservoir water level and the gate opening as the content of the flood control knowledge; S233, store the newly created knowledge in the knowledge base for subsequent analysis and application.

[0030] Further, the S3 generates a flood control scheme for the target flood based on the flood control knowledge base, specifically including, S31, set the outflow to the average value of the case outflow, calculate the average inflow (forecast data) in the future period (such as the future three hours), and use it for subsequent case matching.

[0031] S32, according to the average inflow, obtain the most matched flood control knowledge from the knowledge base; if the reservoir needs to be opened, find the flood control knowledge from the reservoir opening period (the period when the gate changes from completely closed to open); the flood control knowledge is the recommended gate opening under the corresponding water level and inflow.

[0032] S33, calculate the water abandonment period (the period when the water level exceeds the limit and must be discharged for power generation only), based on the obtained most matched flood control knowledge, carry out flood control scheduling for the target flood, and determine the flood control scheduling scheme of the current period; S34, judge whether the current time period number is equal to the target flood time period number, if yes, it means that all time periods have been calculated, end the calculation and turn to S35; otherwise, turn to S31 to continue calculating the remaining time periods.

[0033] S35, output the results, including the calculated results of the outflow, water level process, power generation flow, flood discharge flow, gate opening, etc.

[0034] Further, the specific steps of determining the current time period flood control scheduling scheme based on the obtained most matched flood control knowledge in S33 include, S331, if the reservoir does not discharge flood, and three conditions are met: the reservoir water level is close to (water level difference is less than 0.2m, 0.3m, 1m, etc.) or exceeds the water level of the matched knowledge in the knowledge base, the matched gate opening is not all 0, and the current time period is greater than or equal to 4 time periods from the starting time period or within 3 time periods from the water abandonment time period, if all three conditions are met, start the flood discharge, set the gate opening to the gate opening recommended by the flood control knowledge, and turn to S336; otherwise, turn to S332.

[0035] S332, if the reservoir does not discharge flood, and the gate opening recommended by the flood control knowledge is all 0, it means that there is no need to discharge flood, set the gate opening to 0, and also turn to S336; otherwise, turn to S333.

[0036] S333, if the reservoir is discharging flood, find the last gate opening time, and judge whether it is necessary to keep the current gate opening according to the gate opening time interval.

[0037] S334, judge whether it is necessary to stop the flood discharge, if the reservoir water level is less than the water level corresponding to the flood control knowledge and the corresponding gate opening is all 0, it means that the reservoir can stop the flood discharge, set the gate opening to 0, and turn to S336; if the reservoir does not discharge flood, and the gate opening recommended by the flood control knowledge is all 0, it means that there is no need to discharge flood, set the gate opening to 0, and turn to S336; if it is necessary to continue the flood discharge, turn to S335.

[0038] S335, judge whether it is necessary to change the gate opening, if the water level difference with the water level of the flood control knowledge is small (water level difference is less than 0.2m, 0.3m, 1m, etc.), and the difference between the outflow and the inflow corresponding to the flood control knowledge is smaller than keeping the original gate opening, update the gate opening.

[0039] S336, predict the future operation state of the reservoir, and automatically intervene when necessary to ensure the safe operation of the reservoir during the flood control period.

[0040] S337, according to the gate opening and the output, calculate the current operation of the reservoir. The main method is: according to the gate opening and the water level, calculate the flood discharge; according to the output and the flood discharge, trial the tail water level and get the power generation flow; according to the flood discharge, the power generation flow and the inflow, calculate the reservoir capacity according to the water balance equation, and update the reservoir water level.

[0041] Further, the S333 of finding the last gate operation time to determine whether the current gate opening needs to be maintained specifically includes, S3331, traverse the time sequence to find the first non-zero opening period, record the first opening time and the gate opening; S3332, if no non-zero opening period is found, the same opening does not need to be maintained, and S334 is turned to; S3333, if the current period is within the initial gate operation interval after the first opening (referring to the time interval during which the first opening gate opening needs to be maintained), the same opening needs to be maintained, and S336 is turned to; S3334, initialize the last gate operation time as the first opening time plus the initial opening interval; S3335, traverse from the current period to find the last time point when the gate opening changes, and update the last gate operation time; S3336, calculate the interval between the current period and the last gate operation time, if less than the gate operation interval (the minimum time interval after one gate operation except the first opening during which the opening needs to be maintained), the same opening needs to be maintained, and S336 is turned to; if more than the gate operation interval, the same opening does not need to be maintained, and S334 is turned to.

[0042] Further, the S336 of moderately automatically intervening based on the future operation of the reservoir specifically includes, S3361, according to the current power generation flow and the gate opening, trial the reservoir water level of the next period; S3362, if the reservoir water level of the next period does not exceed the highest limit water level-0.1 meters (reserve a certain margin), S337 is turned to; if the reservoir water level of the next period exceeds the highest limit water level-0.1 meters, it indicates that the current extracted knowledge cannot completely cover the situation faced, and S3363 is turned to; S3363, calculate the power generation flow according to the non-rejection water mode; S3364, calculate the flood discharge according to the inflow and outflow balance principle, and get the new gate opening; S3365, return the calculation time to the last gate operation time, set the gate opening to the new opening calculated to ensure that the reservoir operates within the safe water level range, and turn to S337.

[0043] Taking two adjacent reservoirs (referred to as Reservoir A and Reservoir B) as an example, a specific embodiment of the present invention is provided, including: (1) Solution for Reservoir A S1: Select a typical case. Since Reservoir A has a large storage capacity and a wide water level range, the threshold for determining water level proximity in S331 is 1 meter. The time period for calculating the average inflow is three hours.

[0044] The inflow of Reservoir A and Reservoir B during the target flood is as follows. The initial water levels of Reservoir A and Reservoir B are 864.44m and 717.92m respectively. Reservoir A and Reservoir B are matched with rated output of 200MW and 150MW respectively. During the target flood, the inflow of Reservoir A and the flow process of the AB interval (AB interval) are as follows Figure 5 shown.

[0045] The target flood exhibited a three-peak pattern, with flooding primarily occurring in the first half. In the second half, inflows showed a fluctuating, slightly decreasing trend. The primary source of the flood was inflows from Reservoir A, with all three peaks exceeding 1000 m³ / s. Flows in intervals A and B were relatively small, and their changing trends mirrored those of inflows from Reservoir A. The key characteristics of the target flood at Reservoir A are shown in Table 1.

[0046] Table 1 Target flood characteristics of reservoir A From historical operational data, 38 typical cases were extracted for Reservoirs A and B from 2012 to the present. After calculating the matching degree, Case 28, Reservoir A's operating conditions, were found to be most similar to the target flood, with a matching degree of 0.925. Case 28 (from 01:00 on June 16, 2015, to 04:00 on June 26, 2015) was designated as the typical flood. The characteristic parameters of the typical flood are shown in Table 2:

[0047] Table 2 Flood characteristics corresponding to the best matching typical flood (reservoir A) The following illustrates the operation of Reservoir A during a typical flood. The flood process in this case exhibits a bimodal pattern, with a primary peak flow rate of 2,000 m³ / s and a secondary peak flow rate of approximately 1,500 m³ / s. During this period, the reservoir operated in a full-load continuous power generation mode, with the outflow rate varying in steps with the dynamic adjustment of the gate opening, as shown in Figure 1. Figure 6 and Figure 7 shown.

[0048] S2, knowledge extraction stage: create the flood control knowledge base of reservoir A, as shown in Table 3, 18 flood control knowledge are extracted from typical floods, the inflow is the average inflow of the next three hours, and the gate opening degree from left to right represents the left middle hole, the right middle hole, the left surface hole, the middle surface hole and the right surface hole.

[0049] Table 3 Reservoir A flood control knowledge base Enter S3 knowledge application stage, calculate reservoir scheduling scheme from the first period.

[0050] Period 1, S31 calculates that the average inflow in the next three hours is 329 m³ / s, and the reservoir water level is 864.49 m; S32 obtains the most matched knowledge as knowledge 1; enter S33 to determine the flood control scheduling scheme. Since the reservoir has no flood discharge, in S331, the calculation of abandoned water period is 13, so it is necessary to enter the fifth period (interval of 4 hours) to start the flood discharge. Thereafter, the reservoir maintains the state of no flood discharge operation.

[0051] Period 5, S31 calculates that the average inflow is 663 m³ / s, and the reservoir water level is 864.90 m; S32 obtains the most matched knowledge as knowledge 1; enter S33 to determine the flood control scheduling scheme. Since the reservoir has no flood discharge, the difference between the reservoir water level 864.90 and the knowledge corresponding water level 867.53 m is large, so the flood discharge is temporarily not performed; after calculation, the future reservoir water level does not exceed the water level limit, so no manual intervention is required; according to the gate opening degree, the reservoir operation is calculated.

[0052] Period 11, S31 calculates that the average inflow is 867 m³ / s, and the reservoir water level is 866.68 m; at this time, the reservoir needs to be opened, S32 obtains the most matched knowledge as knowledge 1; enter S33 to determine the flood control scheduling scheme. Since the reservoir has no flood discharge, in S331, since the reservoir water level 866.68 is close to the knowledge corresponding water level 867.53 m, the flood discharge is performed according to the opening degree [0.0, 1.0, 0.0, 0.0, 0.0]; after calculation, the future reservoir water level does not exceed the water level limit, so no manual intervention is required; according to the gate opening degree, the reservoir operation is calculated.

[0053] Period 12, the last gate opening time is period 11, since the gate has not been opened before, the first gate opening time is period 11, the initial gate opening interval is 2, and this period is not within the initial gate opening interval after the first gate opening, so the gate opening degree of the last period is maintained. Thereafter, period 12, 13 and 14 are discharged according to the opening degree [0.0, 1.0, 0.0, 0.0, 0.0].

[0054] Period 14, after calculation by S336, the future reservoir water level will exceed the water level limit, and manual intervention is needed; enter S336, and the new gate opening obtained by trial calculation is [0.0, 6.0, 0.0, 0.0, 0.0], and the last time of gate movement is period 13, so return to period 13 and discharge according to the new gate opening.

[0055] Periods 13-16 discharge according to the opening of [0.0, 6.0, 0.0, 0.0, 0.0].

[0056] Period 17, S31 calculates that the average inflow is 1059m³ / s, and the reservoir water level is 867.58m; the most matched knowledge is knowledge 6, and S333 judges that the gate opening does not need to be kept. Enter S335, and the discharge corresponding to the gate opening kept is 775 / 936m³ / s, and the discharge corresponding to the gate opening kept according to the flood control knowledge is 903 / 1064m³ / s, so the gate opening recommended by the flood control knowledge better fits the inflow, and the gate opening is changed to [0.0, 6.7, 0.0, 0.0, 0.0]. Thereafter, periods 18-21 discharge according to the opening of [0.0, 6.7, 0.0, 0.0, 0.0].

[0057] Period 22, S31 calculates that the average inflow is 843m³ / s, and the reservoir water level is 867.23m; S32 obtains the most matched knowledge as knowledge 7. S333 judges that the gate opening does not need to be kept, enters S335, and the discharge corresponding to the gate opening kept is 901 / 1062m³ / s, and the discharge corresponding to the gate opening kept according to the flood control knowledge is 623 / 784m³ / s, so the gate opening recommended by the flood control knowledge better fits the inflow, and the gate opening is changed to [0.0, 5.0, 0.0, 0.0, 0.0].

[0058] Period 163, S31 calculates that the average inflow is 241m³ / s, and the reservoir water level is 865.76m, and S32 obtains the most matched knowledge as knowledge 16. S333 judges that the gate opening does not need to be kept, enters S334, and the gate opening according to the flood control knowledge is all 0, and the average inflow and reservoir water level according to the flood control knowledge are 248m³ / s and 867.58m, while the actual average inflow and reservoir water level are 241m³ / s and 865.76m, and both the inflow and the reservoir water level are lower than the flood control knowledge, so the reservoir stops discharging.

[0059] After the above steps, the scheduling scheme of reservoir A under the target flood is obtained after each period is calculated, and the operation of reservoir A is as shown in Figure 8 and Figure 9 .

[0060] From Figure 8and Figure 9 It can be seen that the inflow of Reservoir A fluctuates greatly, and gradually falls after experiencing three peaks. Because Reservoir A always maintains full load power generation, the power generation flow is relatively stable. The outflow of Reservoir A changes in steps, which is closely related to the gate opening and basically consistent with the trend of the inflow, which makes the reservoir maintain a high water level without exceeding the water level limit and without the phenomenon of "artificial flood". In terms of water level, due to the fact that Reservoir A did not open the gate to discharge flood in the early stage, the reservoir stored flood, and the water level rose rapidly; in the middle stage, the gate opening was dynamically adjusted according to the inflow, and the water level remained relatively stable; in the later stage, the water level gradually decreased, and the reservoir closed the gate and stopped discharging flood. As can be seen, the reservoir effectively controls the water level by precisely adjusting the gate opening, achieves the goal of flood control, and the dispatching effect is ideal.

[0061] (2) Reservoir B solution According to the flow process of Reservoir A, the target flood process of Reservoir B is obtained. Since the reservoir capacity of Reservoir B is small, the threshold for judging the proximity of the water level in S331 is 0.3 m. The time period length for calculating the average inflow is three hours.

[0062] Enter S1, select a typical case. The characteristics of the target flood of Reservoir B are shown in Table 4:

[0063] Table 4 Characteristics of the target flood of Reservoir B From the historical operation data, 38 typical cases of Reservoir A and Reservoir B from 2012 to now are extracted. After calculating the matching degree, it is found that the operation of Reservoir B in case 28 is closest to the target flood, with a matching degree of 0.86. Case 28 (from 01:00 on June 16, 2015 to 04:00 on June 26, 2015) is taken as a typical flood. The characteristic parameters of the typical flood are shown in Table 5:

[0064] Table 5 Characteristics of the best matching typical flood (Reservoir B) The best matching case of Reservoir B is consistent with the time range of the case of Reservoir A, and is affected by the connection of upstream and downstream water conservancy, the flood process of Reservoir B changes with the change of the outflow of Reservoir A, and the case flood mainly presents a double-peak characteristic: the main flood peak appears in the early stage of the dispatching period, and the flow reaches 2000 m³ / s; the second flood peak appears in the middle period, but the flow is relatively small, about 1000 m³ / s. The operation of Reservoir B in the typical flood is shown in Figure 10 and Figure 11 .

[0065] S2, knowledge extraction stage: create the flood control knowledge base of reservoir B, as shown in Table 6, 19 pieces of flood control knowledge are extracted from the typical flood, the inflow is the average inflow of the next three hours, and the gate opening degree is from left to right, indicating the left, middle and right table holes.

[0066] Table 6 Reservoir B flood control knowledge base Enter S3 knowledge application stage, calculate reservoir scheduling scheme from the first period.

[0067] Period 1, S31 calculates the average inflow of the next three hours as 224 m³ / s, the reservoir water level is 717.92 m, and S32 obtains the most matched knowledge as knowledge 1; enter S33 to determine the flood control scheduling scheme. Since the reservoir has no flood discharge, S331 calculates the abandoned water period as 8, so it needs to enter the fourth period before starting the flood discharge.

[0068] Period 5, S31 calculates the average inflow as 245 m³ / s, the reservoir water level is 718.19 m, and S32 finds the most matched knowledge as knowledge 1 according to the reservoir gate opening; enter S33 to determine the flood control scheduling scheme. Since the reservoir has no flood discharge, S331 calculates the abandoned water period as 8, so the reservoir needs to discharge water in the fifth period, at this time, the reservoir water level 718.19 m is already higher than the knowledge corresponding water level 717.61, so the reservoir starts to discharge water according to the gate opening degree [0.125, 0.0, 0.0]; after calculation, the future reservoir water level does not exceed the water level limit, so no manual intervention is needed; according to the gate opening degree, calculate the reservoir operation.

[0069] Period 6, enter S333 to judge whether the gate opening degree needs to be kept, the last gate opening time is period 5, since the gate has not been opened before, the first gate opening time is period 5, and the initial gate opening interval is 2, this period is not within the initial gate opening interval after the first gate opening, so the gate opening degree of the last period is kept.

[0070] Period 9, S31 calculates the average inflow as 453 m³ / s, the reservoir water level is 716.48 m, and S32 obtains the most matched knowledge as knowledge 11. The first gate opening time is period 5, and the current period is outside the initial gate opening interval after the first gate opening; therefore, initialize the last gate opening time as the first gate opening time (period 5) plus 2 (initial gate opening interval) = period 7, since the gate opening degree has not changed before, the last gate opening time remains unchanged; the interval between the current period and the last gate opening time is 2 hours, which is less than or equal to the gate opening interval, so the same opening degree needs to be kept.

[0071] Period 10, S31 calculates the average inflow to be 498 m³ / s, and the reservoir water level is 716.27 m. S32 obtains the most matching knowledge as knowledge 16. S33 determines the current period flood control scheduling scheme based on the obtained most matching flood control knowledge. S333 judges that the gate opening degree does not need to be maintained, and thus S335 is entered to determine whether the gate opening degree needs to be changed. Since the gate opening degree recommended by the knowledge is consistent with the opening degree of the previous period, the current opening degree is maintained.

[0072] Period 11, S1 calculates the average inflow to be 764 m³ / s, and the reservoir water level is 716.11 m. S32 obtains the most matching knowledge as knowledge 9. S333 judges that the gate opening degree does not need to be maintained. S335 is entered to maintain the gate opening degree. The corresponding abandoned water / outflow is 289 / 488 m³ / s, and the abandoned water / outflow corresponding to the flood control knowledge is 578 / 778 m³ / s. The gate opening degree recommended by the knowledge better fits the inflow, and thus the gate opening degree is changed to [0.125, 0.0, 0.125].

[0073] After the above steps, the scheduling scheme of reservoir B under the target flood is obtained after the calculation of each period. The operation of reservoir B is shown in FIGS. 1 and 2. Figure 12 and Figure 13

[0074] From the reservoir B operation diagram and the gate opening degree diagram Figure 12 , 13 , it can be seen that the outflow is adjusted in steps with the increase and decrease of the inflow, and the gate opening and closing follow the principle of step-by-step opening and step-by-step closing. The gate opening for flood discharge avoids sudden increase of the outflow, and reserves sufficient emergency response time for the downstream area. With the decrease of the inflow, the gate opening degree is gradually reduced, so that the reservoir water level is maintained at a high level. At the same time, the reservoir outflow always maintains full load to utilize water energy as much as possible, fully demonstrating the fine management and dynamic response capability of flood control scheduling. However, due to the limitation of flood control knowledge, although the inflow gradually decreases in the later stage of scheduling, the gate still maintains a part of the opening degree, resulting in a relatively fast water level drop rate. If long-term high water level operation is required, the gate can be closed in period 161. During the flood control scheduling period, there are two periods of short-time opening degree decrease-increase-decrease in period 50 and period 85. Artificial intervention can be used to balance the demand for flood control and water storage and gate operation.

[0075] ​The application selects the best matching case flood as a typical flood, extracts flood control related knowledge and stores the knowledge to form a knowledge base, then applies the extracted knowledge to actual flood control scheduling to intelligently generate a reservoir flood control scheduling scheme. The method mainly includes three stages: case selection, knowledge extraction and knowledge application. In the case selection stage, the reservoir inflow and water level data in the typical flood are traversed to detect whether the gate is moving; in the knowledge extraction stage, the related flood control knowledge is extracted and stored; in the knowledge application stage, the system calculates the average inflow and determines whether the gate opening needs to be adjusted and whether decision intervention is needed through knowledge matching, and automatically adjusts the intervention when the existing knowledge is insufficient to ensure that the water level of the reservoir remains in a safe range,

[0076] The above-mentioned embodiments only express several embodiments of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent. It should be noted that for ordinary skilled persons in the art, without departing from the concept of the present application, several modifications and improvements can be made, which are within the scope of protection of the present application.

Claims

1. A flood control and dispatching method for cascade reservoirs, characterized in that: include: Obtain characteristic parameters of a target flood and multiple case floods, determine the matching degree between the target flood and each case flood based on the characteristic parameters of the target flood and the characteristic parameters of each case flood, and take the case flood with the highest matching degree as the typical flood; Starting from the first period, the operation data of the cascade reservoirs under typical flood conditions are traversed period by period to determine whether the gate openings of the cascade reservoirs change in adjacent periods. If so, the average inflow of the cascade reservoirs in the next period is determined, and the average inflow, reservoir water level and gate opening of the cascade reservoirs in the next period are integrated into flood control knowledge and stored in the flood control knowledge base; if not, the operation data of the cascade reservoirs in the next period under typical flood conditions is continuously traversed; Obtain the average inflow of the cascade reservoirs under the target flood during a period of time, and obtain the most matching flood control knowledge from the flood control knowledge base based on the average inflow of the cascade reservoirs under the target flood during a period of time; perform flood control scheduling for the target flood based on the most matching flood control knowledge to obtain a flood control scheduling plan for the current period; Determine whether the total number of time periods of the currently obtained flood control scheduling plan is equal to the total number of time periods of a typical flood; If so, all time periods of the target flood are determined, and the flood control scheduling plans for each time period are summarized to obtain the flood control scheduling plans for the cascade reservoirs under the target flood in all time periods; if not, the average inflow flow of the cascade reservoirs under the target flood in the next time period is determined.

2. A flood control and dispatching method for cascade reservoirs according to claim 1, characterized in that: The above-mentioned process of integrating the average inflow, reservoir water level and gate opening of the cascade reservoirs in a future period into flood control knowledge and storing it in the flood control knowledge base specifically includes: Taking the time corresponding to the first flood control knowledge as the starting data point, the operation data of the cascade reservoirs under typical floods are traversed period by period to determine whether the gate openings in adjacent periods have changed. If so, the average inflow of the cascade reservoirs in the next period is determined based on the typical flood data, and a flood control knowledge is constructed based on the average inflow of the cascade reservoirs in the next period, the current reservoir water level and the gate opening, and stored in the flood control knowledge base. If not, the operation data of the cascade reservoirs under typical floods in the next period are continued to be traversed. The storage structure of the flood control knowledge base is determined based on the following formula: Flood prevention knowledge i : Average inflow of cascade reservoirs in the next period, reservoir water level, and gate opening; in, i A serial number representing flood control knowledge.

3. The flood control and dispatching method for cascade reservoirs according to claim 1, characterized in that: After all time periods of the target flood are determined, the flood control scheduling plan for each time period is summarized to obtain the flood control scheduling plan for the cascade reservoirs under the target flood in all time periods, specifically including: Based on the average inflow of the cascade reservoirs during a period of time under the target flood, a piece of current flood control knowledge that best matches the average inflow is obtained from the knowledge base. When the cascade reservoirs are opened, current flood control knowledge is searched from the opening period of the cascade reservoirs. The current flood control knowledge includes the gate opening corresponding to the water level and inflow. Determine the water-discharge period for cascade reservoirs, which is the period when the cascade reservoirs must discharge floodwaters due to water levels exceeding the limit while generating power without opening the sluice gates. Perform flood control scheduling for the target flood based on the most relevant current flood control knowledge, and determine the flood control scheduling plan for the current period. Determine whether the total number of the current period is equal to the total number of the target flood period; if so, all the periods of the target flood are determined, and the flood control scheduling plan for each period is summarized to obtain the flood control scheduling plan for the cascade reservoirs under the target flood in all periods, and the flood control scheduling plan includes outflow flow, water level process, power generation flow, flood discharge flow and gate opening; if not, determine the average inflow flow of the cascade reservoirs under the target flood in the next period.

4. A flood control and dispatching method for cascade reservoirs as claimed in claim 3, characterized in that: The flood control dispatching for the target flood based on the most matching flood control knowledge is performed to determine the flood control dispatching plan for the current period, specifically including: If the cascade reservoirs do not release flood water, and when the actual water level difference of the reservoirs is less than the set threshold or exceeds the water level matched in the knowledge base, the matched gate openings are not all 0, and the current period is greater than or equal to 4 periods away from the start period or within 3 periods away from the water abandonment period, then flood release will begin, the gate opening will be set to the gate opening recommended by the flood control knowledge, and the future operation status of the cascade reservoirs will be predicted; If the cascade reservoirs do not release floodwaters and the gate openings recommended by flood control knowledge are all 0, then no flood release will be made, the gate openings will be set to 0, and the future operating status of the cascade reservoirs will be predicted; If the cascade reservoir is releasing floodwater, find the time of the most recent gate operation and determine whether to maintain the current gate opening based on the gate operation interval; Determine whether to stop flood discharge. If the water level of the cascade reservoir is lower than the water level corresponding to the flood control knowledge and the corresponding gate openings are all 0, then stop flood discharge, set the gate openings to 0, and predict the future operating status of the cascade reservoir. If the cascade reservoir does not discharge flood and the gate openings recommended by flood control knowledge are all 0, then do not discharge flood, set the gate openings to 0, and predict the future operating status of the cascade reservoir. If flood discharge continues, determine whether to change the gate opening. If the difference between the water level and the water level of flood control knowledge is less than the set threshold, and compared with maintaining the original gate opening, the difference between the outflow and inflow corresponding to the flood control knowledge is less than the set threshold, then update the gate opening, and determine the current operation status of the cascade reservoir based on the gate opening and output.

5. A flood control and dispatching method for cascade reservoirs as claimed in claim 4, characterized in that: Determining the current operating conditions of the cascade reservoirs based on gate opening and output specifically includes: Determine flood discharge based on gate opening and water level; According to the output and flood discharge, the tailwater level is determined and the power generation flow is obtained; According to the flood discharge, power generation flow and inflow, the reservoir capacity is determined by the water balance equation to update the reservoir water level; Determine the current operating status of the cascade reservoirs based on the updated reservoir water levels and gate openings.

6. A flood control and dispatching method for cascade reservoirs as claimed in claim 4, characterized in that: If the cascade reservoir is releasing floodwater, the most recent gate operation time is found, and based on the gate operation time interval, whether the current gate opening is maintained is determined, including: Traverse the time series and find the first period with non-zero opening, record the first opening time and gate opening; if no period with non-zero opening is found, do not keep the same opening, and determine whether to stop flood discharge; If the current period is within the initial gate-operating interval after the first gate opening, the same gate opening is maintained, and the future operating status of the cascade reservoir is predicted; Initialize the most recent gate opening time to the first gate opening time plus the initial gate opening interval; traverse forward from the current period to find the time point when the gate opening last changed, and update the most recent gate opening time; Determine the interval between the current period and the most recent gate opening time. If it is less than the gate opening interval, maintain the same opening and predict the future operating status of the cascade reservoir; if it exceeds the gate opening interval, do not maintain the same opening and determine whether to stop flood discharge.

7. A flood control and dispatching method for cascade reservoirs according to claim 6, characterized in that: The predicted future operating status of cascade reservoirs includes: Determine the reservoir water level for the next period based on the power generation flow and gate opening of the cascade reservoirs in the current period; If the reservoir water level in the next period does not exceed the difference between the maximum limit water level and the reserved margin, the current operation status of the cascade reservoir is determined based on the gate opening and output of the current period; If the reservoir water level in the next period exceeds the difference between the maximum limit water level and the reserved margin, the power generation flow is determined based on the method of not abandoning water, and the flood discharge flow is determined based on the principle of inflow and outflow balance to obtain the intermediate gate opening; the calculation time is returned to the previous gate operation time, and the intermediate gate opening is used as the gate opening for the next period, and the future operating status of the cascade reservoir is determined based on the gate opening and output of the next period.

Images