Intelligent water level regulation and control method for water conservancy project

Abstract

The invention relates to the technical field of water conservancy dispatching management, in particular to an intelligent water level regulation and control method for a water conservancy project. The method comprises the following steps: firstly, analyzing and obtaining a water quantity increase parameter of a regional water resource project in a preset future time period, and further obtaining a water inflow of a to-be-regulated reservoir in the preset future time period; and then obtaining a first water level variation based on a preset prediction model, and obtaining a second water level variation based on the water inflow, thereby analyzing and obtaining the predicted water level of the reservoir to be regulated and controlled, and carrying out intelligent water level regulation and control. From the angle of indirect influence of the water quantity on the water level, the future water inflow change of the reservoir to be regulated and controlled is analyzed and evaluated to predict the water level; the water level is predicted based on the historical water level change rule angle; and the predicted water level of the reservoir to be regulated and controlled is evaluated by integrating the two methods, so that the limitation of singly analyzing the historical water level change rule is avoided, the water level prediction accuracy is improved, and the intelligent regulation and control effect on the reservoir water level is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of water conservancy dispatching and management, and in particular to an intelligent water level control method for a water conservancy project. Background Art

[0002] Water conservancy projects refer to a series of engineering activities that use water resources for planning, design, construction and management, aiming to regulate water flow, improve the use of water resources, ensure water safety and prevent and control floods. Reservoirs are an important part of water conservancy projects, mainly used to store and regulate water resources. Therefore, the water level regulation of reservoirs is directly related to the effective management of water resources, flood prevention and control, safety assurance and hydropower generation.

[0003] At present, the future water level changes are usually predicted based on the historical water level changes of the reservoir, so as to adjust the reservoir water level and scientifically dispatch the reservoir resources; and the seasonal autoregressive integrated moving average (SARIMA) model is usually used in the existing technology to predict the future water level of the reservoir. However, the water level of the reservoir is usually affected by factors such as meteorology, environment, and the upstream water volume in its local water resources project, which may lead to poor accuracy of the model prediction results, and then lead to poor regulation and control of the reservoir water level. Summary of the invention

[0004] In order to solve the technical problem that the prior art has poor control effect on the reservoir water level, the purpose of the present invention is to provide an intelligent water level control method for a water conservancy project. The technical solution adopted is as follows: In the preset historical period at the current moment, the historical indicator sequence of the regional water resources project under each water volume influencing indicator and the historical water level sequence of the reservoir to be regulated in the regional water resources project are obtained; in the preset future period at the current moment, the future indicator sequence of the regional water resources project under each water volume influencing indicator is obtained; According to the similarity of the change trend between the historical indicator sequence under each water volume influencing indicator and the historical water level sequence, combined with the future indicator sequence under each water volume influencing indicator, the water volume growth parameter of the regional water resources project in the preset future period is obtained; according to the water volume growth parameter and the contribution of each water inlet of the reservoir to be regulated in the regional water resources project to the water volume of the reservoir to be regulated, the water inflow of the reservoir to be regulated in the preset future period is obtained; Based on the preset prediction model and the historical water level sequence, a first water level change of the reservoir to be regulated within a preset future period is obtained; based on the water inflow, a second water level change of the reservoir to be regulated within a preset future period is obtained; based on the first water level change and the second water level change, combined with the similarity of changes between the future indicator sequence and the historical indicator sequence under the preset water volume influencing index, a predicted water level of the reservoir to be regulated within a preset future period is obtained; The water level of the reservoir to be regulated is intelligently regulated according to the predicted water level.

[0005] Furthermore, the water volume influencing index includes at least precipitation, temperature and total water inflow.

[0006] Furthermore, the method for obtaining the water volume growth parameter includes: According to the future indicator sequence under each water volume influencing indicator, obtaining the indicator change amount under the corresponding water volume influencing indicator; Under each water quantity impact indicator, according to the Pearson correlation coefficient between the historical indicator sequence and the historical water level sequence, the weight of the impact of the change of each water quantity impact indicator on the water quantity growth of the regional water resources project is obtained; The water volume growth impact weight under each water volume impact index is used to weight the corresponding indicator change, and the weighted sum result is used as the water volume growth parameter of the regional water resources project in a preset future period.

[0007] Furthermore, the method for obtaining the water volume growth impact weight includes: The positive correlation of the Pearson correlation coefficient under the precipitation index is normalized to obtain the first change correlation coefficient; the positive correlation of the Pearson correlation coefficient under the total water inflow index is normalized to obtain the second change correlation coefficient; the negative correlation of the Pearson correlation coefficient under the temperature index is normalized to obtain the third change correlation coefficient; Normalizing the first change correlation coefficient, the second change correlation coefficient, and the third change correlation coefficient respectively, wherein the sum of the normalized first change correlation coefficient, the second change correlation coefficient, and the third change correlation coefficient is 1; The normalized first change correlation coefficient is used as the weight of water volume growth under the precipitation index; the normalized second change correlation coefficient is used as the weight of water volume growth under the total water inflow index; the negative of the normalized third change correlation coefficient is taken as the weight of water volume growth under the temperature index.

[0008] Furthermore, the method for obtaining the water contribution degree includes: In a preset historical period at the current moment, a first historical water inflow sequence of the reservoir to be regulated and a second historical water inflow sequence of each water inlet of the reservoir to be regulated in the regional water resources project corresponding to an upstream water source are obtained; The first historical water inflow sequence of the reservoir to be regulated is used as the dependent variable, and the second historical water inflow sequence of each water inlet is used as the independent variable to construct a linear regression model; each regression coefficient in the linear regression model is used as the first water volume contribution of the corresponding water inlet to the reservoir to be regulated; Normalize the storage capacity of the upstream water source corresponding to each water inlet of the reservoir to be regulated in the regional water resources project, and use the normalized result as the second water volume contribution of the corresponding water inlet to the reservoir to be regulated; The first water volume contribution and the second water volume contribution are integrated to obtain the water volume contribution of the corresponding water inlet to the reservoir to be regulated.

[0009] Furthermore, the method for obtaining the water inflow includes: The product of the water contribution of each water inlet to the reservoir to be regulated and the water volume growth parameter is taken as the contributed water volume of each water inlet to the reservoir to be regulated; the contributed water volumes of all water inlets to the reservoir to be regulated are combined to obtain the water inflow of the reservoir to be regulated in a preset future period.

[0010] Furthermore, the method for obtaining the first water level change comprises: The preset prediction model is a SARIMA model; the length of the preset historical period is greater than the length of the preset future period; the historical water level sequence is divided into a number of sequence sub-segments with the same length as the preset future period; a model training set is constructed based on all the sequence sub-segments to train the SARIMA model; The sequence sub-segment closest to the current moment is input into the trained SARIMA model, and the predicted water level sequence within a preset future time period is output; based on the water level difference corresponding to the first and last sequence elements in the predicted water level sequence, the first water level change of the reservoir to be regulated within the preset future time period is obtained.

[0011] Furthermore, the method for obtaining the second water level change comprises: The sum of the water inflow of the reservoir to be regulated and the water storage at the current moment is used as the predicted water storage of the reservoir to be regulated in a preset future time period; based on a preset water storage-water level model and the predicted water storage, the future water level of the reservoir to be regulated in the preset future time period is obtained; the water level at the current moment is subtracted from the future water level to obtain a second water level change.

[0012] Furthermore, the method for obtaining the predicted water level includes: The preset water volume influencing index is precipitation; in the historical precipitation sequence, a reference sequence of the future precipitation sequence is obtained; according to the sequence similarity between the reference sequence and the future precipitation sequence, a first reference weight of the first water level change is obtained; according to the first reference weight, a second reference weight of the second water level change is obtained, and the first reference weight is negatively correlated with the second reference weight and the sum value is 1; The first water level change is weighted using the first reference weight, and the second water level change is weighted using the second reference weight, and the weighted sum is used as the predicted water level change; the sum of the water level at the current moment and the predicted water level change is used as the predicted water level of the reservoir to be regulated within a preset future time period.

[0013] Furthermore, the method for intelligently regulating the water level of the reservoir to be regulated according to the predicted water level includes: If the predicted water level is greater than the preset flood control water level, the outlet sluice of the reservoir to be regulated is opened to release water until the water level is lower than the preset flood control water level; if the predicted water level is lower than the preset low water level, the outlet sluice of the reservoir to be regulated is closed until the water level is higher than the preset low water level.

[0014] The present invention has the following beneficial effects: The present invention first obtains the historical indicator sequence of the regional water resources project under each water quantity influencing indicator and the historical water level sequence of the reservoir to be regulated in the regional water resources project, and then obtains the future indicator sequence of the regional water resources project under each water quantity influencing indicator, so as to improve the data analysis basis for the subsequent water level prediction; according to the similarity of the change trend between the historical indicator sequence under each water quantity influencing indicator and the historical water level sequence, the influence of the change of the water quantity influencing indicator on the water quantity change of the regional water resources project is indirectly evaluated, and then the water quantity growth parameter of the regional water resources project in the preset future time period is obtained in combination with the future indicator sequence under each water quantity influencing indicator. The water quantity growth parameter reflects the future water quantity growth of the regional water resources project, which is used for the subsequent evaluation of the future water inflow of the reservoir to be regulated to further evaluate the water level change. Preparation; according to the water volume growth parameters and the water volume contribution of each water inlet of the reservoir to be regulated in the regional water resources project to the reservoir to be regulated, the water inflow of the reservoir to be regulated in the preset future period is obtained; firstly, based on the preset prediction model and the historical water level sequence, the water level change law is analyzed to predict the first water level change of the reservoir to be regulated; then the second water level change of the reservoir to be regulated is obtained according to the water inflow; according to the first water level change and the second water level change, combined with the similarity of the changes between the future indicator sequence and the historical indicator sequence under the preset water volume influencing index, the influence of season differences is reduced, the reference weights for the first water level change and the second water level change are determined, and the predicted water level of the reservoir to be regulated in the preset future period is obtained; finally, the water level of the reservoir to be regulated is intelligently regulated according to the predicted water level. The present invention analyzes and evaluates the future changes in water inflow to the reservoir to be regulated from the perspective of the indirect impact of water volume on water level to predict the water level; then predicts the water level based on the historical water level change law; and then comprehensively evaluates the predicted water level of the reservoir to be regulated, thereby avoiding the limitations of a single analysis of the historical water level change law, improving the accuracy of the predicted water level, and further improving the intelligent regulation effect of the reservoir water level. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 A flow chart of a method for intelligent water level control of a water conservancy project provided by one embodiment of the present invention; Figure 2 A flow chart of a method for obtaining water volume growth parameters provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of a water conservancy project intelligent water level control method proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0018] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0019] The specific scheme of the intelligent water level control method for a water conservancy project provided by the present invention is described in detail below with reference to the accompanying drawings.

[0020] See also Figure 1 , which shows a flow chart of a method for intelligent water level control of a water conservancy project provided by an embodiment of the present invention, specifically comprising: Step S1, within the preset historical time period at the current moment, obtain the historical indicator sequence of the regional water resources project under each water quantity influencing indicator, and the historical water level sequence of the reservoir to be regulated in the regional water resources project; within the preset future time period at the current moment, obtain the future indicator sequence of the regional water resources project under each water quantity influencing indicator.

[0021] It should be noted that the water level control method for each reservoir is the same. Here, only any reservoir is taken as an example for analysis and description. Regional water resources engineering refers to all artificial water conservancy projects and natural rivers, lakes and other water bodies within the local area of ​​the reservoir to be regulated. The upstream and downstream of each reservoir to be regulated are usually connected to different water bodies, which affects the water level changes of the reservoir to be regulated.

[0022] Considering that the water level of the reservoir to be regulated is mainly affected by the water inflow into the reservoir, and the water inflow into the reservoir is mainly affected by the water volume in the regional water resources project, and the water volume in the regional water resources project is usually affected by multiple factors, the embodiment of the present invention first obtains the historical indicator sequence of the regional water resources project under each water volume influencing indicator, and the historical water level sequence of the reservoir to be regulated in the regional water resources project, so as to facilitate the subsequent analysis and evaluation of the impact of changes in water volume influencing indicators on changes in water volume; further obtain the future indicator sequence of the regional water resources project under each water volume influencing indicator, so as to predict the future water inflow situation, thereby analyzing and predicting the water level of the reservoir to be regulated, so as to intelligently control the water level.

[0023] In one embodiment of the present invention, first, the current moment is taken as the starting point, and a preset historical period is obtained in the reverse direction of the time sequence, and at the same time, a preset future period is obtained in the forward direction of the time sequence, wherein the duration of the preset historical period is greater than the preset future period, and the duration of the preset historical period is set to five years, and the duration of the preset future period is set to one week. The duration of the preset future period will affect subsequent regulation and control behaviors, and the implementer can also set it according to actual needs, which will not be elaborated here; Then, the historical indicator sequence of each water volume influencing indicator of the regional water resources project in the past five years is obtained, considering that precipitation will lead to an increase in the water volume of the entire regional water resources project, which may in turn lead to an increase in the water volume of the reservoir to be regulated; similarly, an increase in the total water inflow of the regional water resources project will also lead to an increase in the water volume of the reservoir to be regulated; however, temperature may accelerate water evaporation, thereby leading to a decrease in the water volume of the entire regional water resources project, and then a decrease in the water volume of the reservoir to be regulated; It should be noted that the analysis of the regional water resource project in the embodiment of the present invention does not include water bodies in frozen conditions, that is, it only considers the situation where the temperature rise causes water evaporation and thus the water volume decreases, and does not consider the situation where the temperature rise causes the ice and snow to melt, thereby increasing the water volume.

[0024] Based on this, in a preferred embodiment of the present invention, the water volume influencing index at least includes precipitation, temperature and total water inflow, and the historical precipitation sequence, historical temperature sequence and historical total water inflow sequence are obtained respectively. The implementer may also set other types and quantities of water volume influencing indicators; the sequence elements in the historical precipitation sequence and the historical temperature sequence are the daily average values ​​of the corresponding water volume influencing index within the past five years, such as using the average temperature of one day as the sequence element to construct the historical temperature sequence, and the sequence element in the historical total water inflow sequence is the sum of the daily cumulative water flow at all water inlets of the regional water resources project; Then, a historical water level sequence of the reservoirs to be regulated in the regional water resources project is constructed. The sequence elements in the historical water level sequence are the daily water level heights of the reservoirs to be regulated in the past five years. The implementer can use the water level height at zero o'clock every day as the sequence element to construct the historical water level sequence; At the same time, the future indicator sequence of each water volume influencing indicator of the regional water resources project in the next week is obtained. The future precipitation sequence and the future temperature sequence can be obtained through the forecast results of the meteorological station, while the future total water inflow sequence needs to evaluate the water flow rate per unit time based on the flow velocity at each water inlet of the reservoir to be regulated at the current moment and the water flow cross-sectional area at the inlet, and then evaluate the daily cumulative water flow rate in combination with time. The sum of the daily cumulative water flow rates at all water inlets is taken as the total daily water inflow of the reservoir to be regulated, and then used as a sequence element to construct the future total water inflow sequence.

[0025] It should be noted that the sequence elements in the historical total water intake sequence and the future total water intake sequence are all cumulative values. For example, the total water intake corresponding to the second day includes the total water intake on the first day, so as to facilitate the subsequent evaluation of the change in the total water intake.

[0026] Step S2, based on the historical indicator sequence under each water volume influencing indicator and the similarity of the changing trend between the historical water level sequence, combined with the future indicator sequence under each water volume influencing indicator, obtain the water volume growth parameters of the regional water resources project in the preset future period; based on the water volume growth parameters and the water volume contribution of each water inlet of the reservoir to be regulated in the regional water resources project to the reservoir to be regulated, obtain the water inflow of the reservoir to be regulated in the preset future period.

[0027] Considering that different changes in different water quantity influencing indicators will affect the water quantity changes of regional water resources projects to different degrees, and then affect the water quantity of the reservoir to be regulated to indirectly affect the water level; therefore, based on the historical changes of each water quantity influencing indicator and the similarity of the change trends between the historical water level changes of the reservoir to be regulated, the impact of each water quantity influencing indicator on the water level of the reservoir to be regulated can be analyzed, thereby indirectly evaluating the impact of the changes in water quantity influencing indicators on the water quantity changes of regional water resources projects; then, combined with the future estimated changes of each water quantity influencing indicator, the future water quantity changes of regional water resources projects can be predicted; Based on the above logic, the embodiment of the present invention first obtains the water volume growth parameters of the regional water resources project in a preset future period according to the similarity of the changing trends between the historical indicator sequence under each water volume influencing indicator and the historical water level sequence, and then combines the future indicator sequence under each water volume influencing indicator; the water volume growth parameters reflect the future water volume growth of the regional water resources project, and prepare for the subsequent evaluation of the future water inflow of the reservoir to be regulated, thereby analyzing and predicting the water level of the reservoir to be regulated, so as to intelligently regulate the water level.

[0028] Preferably, in one embodiment of the present invention, the method for obtaining the water volume growth parameter includes: See also Figure 2 , which shows a flow chart of a method for obtaining water volume growth parameters provided by an embodiment of the present invention, specifically comprising: Step S201, according to the future indicator sequence under each water volume influencing indicator, obtain the indicator change amount under the corresponding water volume influencing indicator.

[0029] As an example, first obtain the indicator change amount under each water volume influencing indicator in the preset future time period. Specifically, directly subtract the first sequence element from the last sequence element in the future indicator sequence under each water volume influencing indicator to obtain the indicator change amount of the corresponding water volume influencing indicator in the preset future time period; for example, in the future precipitation sequence, subtract the temperature of the last day from the temperature of the first day to obtain the temperature change. The method for obtaining the change amounts of other indicators is the same and will not be repeated here.

[0030] It should be noted that the change in indicators is not only positive, but may also be zero or negative.

[0031] Step S202: Under each water quantity influencing indicator, according to the Pearson correlation coefficient between the historical indicator sequence and the historical water level sequence, the weight of the impact of the change of each water quantity influencing indicator on the water quantity growth of the regional water resources project is obtained.

[0032] Considering that the water level change of the reservoir to be regulated can indirectly reflect the change of the total water inflow of the regional water resources project, if the total water inflow of the regional water resources project decreases, the water level of the reservoir to be regulated may also decrease relatively; considering that the total water inflow of the regional water resources project is affected by a variety of water volume influencing indicators, an indirect analytical relationship between each water volume influencing indicator and the water level can be established. Through the correlation between the changes in the water volume influencing indicators and the water level, the correlation between the changes in the water volume influencing indicators and the total water volume of the regional water resources project can be indirectly evaluated, and the weight of the impact of the changes in each water volume influencing indicator on the water volume growth of the regional water resources project can be obtained, so as to facilitate the subsequent combination of the future changes in each water volume influencing indicator to evaluate the future changes in the water volume of the regional water resources project.

[0033] Preferably, in one embodiment of the present invention, considering that the Pearson correlation coefficient can evaluate the correlation of changes between sequences, and considering that precipitation and total water inflow are usually positively correlated with the change of water volume of regional water resources projects, while temperature is negatively correlated with the change of water volume of regional water resources projects; based on this, the method for obtaining the water volume growth impact weight includes: The positive correlation of the Pearson correlation coefficient under the precipitation index is normalized to obtain the first change correlation coefficient; the positive correlation of the Pearson correlation coefficient under the total water inflow index is normalized to obtain the second change correlation coefficient; the negative correlation of the Pearson correlation coefficient under the temperature index is normalized to obtain the third change correlation coefficient; Normalizing the first change correlation coefficient, the second change correlation coefficient, and the third change correlation coefficient respectively, wherein the sum of the normalized first change correlation coefficient, the second change correlation coefficient, and the third change correlation coefficient is 1; The normalized first change correlation coefficient is used as the weight of water volume growth under the precipitation index; the normalized second change correlation coefficient is used as the weight of water volume growth under the total water inflow index; the negative of the normalized third change correlation coefficient is taken as the weight of water volume growth under the temperature index.

[0034] As an example, under the precipitation index and the total water inflow index, the Pearson correlation coefficient is used as a function x in the equation is used for positive correlation normalization, so that the closer the Pearson correlation coefficient is to 1, the larger the normalization result is, indicating that the increase in the precipitation index or the total water inflow index will lead to an increase in the water level, which further indicates that the precipitation index or the total water inflow index has a greater impact on the water level change of the reservoir to be regulated, and indirectly indicates that the greater the impact on the water volume change of the regional water resources project, the greater the first change correlation coefficient and the second change correlation coefficient; Under the temperature indicator, the Pearson correlation coefficient is used as a function x in the figure is used for negative correlation normalization, so that the closer the Pearson correlation coefficient is to -1, the larger the normalization result is, indicating that the increase in the temperature index will cause the water level of the reservoir to be regulated to drop, further indicating that the temperature index has a greater impact on the water level change, which indirectly indicates that the greater the impact on the water volume change of the regional water resources project, the greater the third change correlation coefficient; Then the three change correlation coefficients are normalized, for example, the first change correlation coefficient is normalized, , where a is the normalized first change correlation coefficient, and is also the water volume growth impact weight under the precipitation index; D1 is the first change correlation coefficient; D2 is the second change correlation coefficient; D3 is the third change correlation coefficient; The first change correlation coefficient is normalized by dividing it by the sum of the three change correlation coefficients, so that the sum of the normalized first change correlation coefficient, the second change correlation coefficient and the third change correlation coefficient is 1; then the second change correlation coefficient and the third change correlation coefficient are normalized in the same way to obtain the water volume growth impact weight b under the total water inflow index and the water volume growth impact weight c under the temperature index.

[0035] In other embodiments, the implementer may also use other methods such as mutual correlation coefficient or DTW similarity instead of Pearson correlation coefficient to evaluate the change correlation coefficient under the corresponding water volume influencing index; it is also possible to directly measure the difference between sequence elements with the same sequence number in two historical indicator sequences, and use the mean of all differences as the corresponding change correlation coefficient; it and Pearson correlation coefficient are both existing technologies and will not be described in detail here.

[0036] Step S203, using the water volume growth impact weight under each water volume impact index, weighting the corresponding indicator change, and taking the weighted sum result as the water volume growth parameter of the regional water resources project in the preset future period.

[0037] As an example, the water volume growth impact weight under each water volume impact indicator is multiplied by the corresponding indicator change, and the sum of the products is used as the water volume growth parameter of the regional water resources project in the preset future period.

[0038] Considering that the water body structure of each water inlet of the reservoir to be regulated in the regional water resources project is different, such as capacity or water flow trend, which may lead to different contributions to the water inflow of the reservoir to be regulated, after obtaining the water volume growth parameters of the regional water resources project in the preset future period, the water volume contribution of each water inlet of the reservoir to be regulated in the regional water resources project to the reservoir to be regulated can be combined to obtain the water volume of the reservoir to be regulated in the preset future period. The water volume contribution reflects the contribution of the water body corresponding to each water inlet of the reservoir to be regulated in the regional water resources project to the water volume of the reservoir to be regulated, that is, how much water of the water body can flow into the reservoir to be regulated, so as to evaluate the water volume of the reservoir to be regulated in the future by combining the contribution water of all water inlets.

[0039] Preferably, in one embodiment of the present invention, considering that each water inlet of the reservoir to be regulated in the regional water resources project corresponds to the inflow of the upstream water source, it will usually contribute to the reservoir to be regulated to varying degrees as the inflow of the reservoir to be regulated; therefore, the change in the inflow of each water inlet of the reservoir to be regulated corresponding to the upstream water source can be analyzed, and the impact on the change in the inflow of the reservoir to be regulated can be analyzed, and then the contribution of each water inlet can be evaluated; considering that the flow of water in the water resources system is usually more complicated, and regression analysis is usually based on the relationship between the change in the inflow of the upstream water source corresponding to each water inlet and the inflow of the reservoir to be regulated, providing a systematic method to understand the relative contribution of each water source to the reservoir inflow, and the regression coefficient in the regression analysis model can accurately reflect the contribution intensity of each water inlet; considering that the smaller the storage capacity of each water inlet corresponding to the upstream water source, the lower its total water volume may be, and the lower the degree of contribution to the water volume of the reservoir to be regulated may be; based on this, the method for obtaining the water volume contribution degree includes: In a preset historical period at the current moment, a first historical water inflow sequence of the reservoir to be regulated and a second historical water inflow sequence of each water inlet of the reservoir to be regulated in the regional water resources project corresponding to an upstream water source are obtained; The first historical water inflow sequence of the reservoir to be regulated is taken as the dependent variable, and the second historical water inflow sequence of each water inlet is taken as the independent variable to construct a linear regression model; each regression coefficient in the linear regression model is taken as the contribution of the corresponding water inlet to the first water volume of the reservoir to be regulated; Normalize the storage capacity of the upstream water source corresponding to each water inlet of the reservoir to be regulated in the regional water resources project, and use the normalized result as the second water volume contribution of the corresponding water inlet to the reservoir to be regulated; The first water contribution and the second water contribution are integrated to obtain the water contribution of the corresponding water inlet to the reservoir to be regulated.

[0040] As an example, firstly, the daily water inflow of the reservoir to be regulated in the past five years is obtained, and the water inflow is used as a sequence element to construct a first historical water inflow sequence; then, the daily water inflow of the upstream water source corresponding to each water inlet of the reservoir to be regulated in the past five years is used as a sequence element to construct a second historical water inflow sequence for each water inlet; wherein the historical water inflows in the first historical water inflow sequence and the second historical water inflow sequence are not accumulated; Then, a linear regression model is constructed, wherein the construction of the linear regression model is an existing technology well known to those skilled in the art and will not be described in detail here; then, the regression coefficient of the independent variable corresponding to each water inlet in the constructed linear regression model is used as the first water volume contribution of the corresponding water inlet to the regulated reservoir; Then, the storage capacity of the upstream water source corresponding to each water inlet of the reservoir to be regulated in the regional water resources project is divided by the sum of the storage capacities of the upstream water sources corresponding to all water inlets to perform normalization, and the normalized result is used as the second water volume contribution of the corresponding water inlet; Finally, the first water contribution and the second water contribution are multiplied and combined to obtain the water contribution of the corresponding water inlet to the reservoir to be regulated; in other examples, the implementer can also use other basic mathematical operations such as addition or weighted summation to combine the two, which will not be repeated here.

[0041] After obtaining each water inlet of the reservoir to be regulated in the regional water resources project, the water inflow of the reservoir to be regulated in a preset future period can be further obtained.

[0042] In a preferred embodiment of the present invention, the product of the water contribution of each water inlet to the reservoir to be regulated and the water volume growth parameter is used as the contributed water of each water inlet to the reservoir to be regulated; the sum of the contributed water of all water inlets to the reservoir to be regulated is used as the water inflow of the reservoir to be regulated in a preset future period.

[0043] Step S3, based on the preset prediction model and the historical water level sequence, obtain the first water level change of the reservoir to be regulated in the preset future time period; obtain the second water level change of the reservoir to be regulated in the preset future time period according to the water inflow; based on the first water level change and the second water level change, combined with the similarity of the changes between the future indicator sequence and the historical indicator sequence under the preset water volume influencing index, obtain the predicted water level of the reservoir to be regulated in the preset future time period.

[0044] Considering that the historical water level information of the reservoir to be regulated can provide certain information on the water level change pattern, the water level change in the preset future period can be predicted based on the relevant prediction model.

[0045] Preferably, in one embodiment of the present invention, the SARIMA model can provide effective prediction results and reflect the regularity of water level changes when processing water level data with seasonal, trend and autocorrelation characteristics; therefore, the preset prediction model is set to the SARIMA model; since the length of the preset historical period is greater than the length of the preset future period, the historical water level sequence is divided into a number of sequence sub-segments with the same length as the preset future period; further, a model training set is constructed based on all the sequence sub-segments to train the SARIMA model; The sequence sub-segment closest to the current moment is input into the trained SARIMA model, and the predicted water level sequence within the preset future period is output; based on the water level difference corresponding to the first and last sequence elements in the predicted water level sequence, the first water level change of the reservoir to be regulated within the preset future period is obtained.

[0046] As an example, first take the current moment as the starting point and divide the historical water level sequence into several sequence sub-segments with the same length as the preset future time period, that is, the length of the sequence sub-segment is 7. After traversing and dividing the historical water level sequence from the tail end into sequence sub-segments with a length of 7, each sequence sub-segment is used as input and the next adjacent sequence sub-segment is used as output to construct a model training set to train the SARIMA model; then the sequence sub-segment corresponding to the tail end is used as the sub-segment to be predicted and input into the trained model to output the predicted water level sequence; the first sequence element is subtracted from the tail sequence element to obtain the first water level change.

[0047] It should be noted that the training and application of the SARIMA model is an existing technology well known to those skilled in the art and will not be described in detail here. In other embodiments, the implementer may also construct other forms of model training sets, such as taking the water level sequence sub-segments corresponding to each month in the historical water level sequence as input and the water level sequence sub-segments within the following week as output to construct a training set, which will not be described in detail here.

[0048] Considering that the water inflow into the reservoir to be regulated in the preset future period will affect the water level change, the second water level change of the reservoir to be regulated in the preset future period can be obtained according to the water inflow.

[0049] Considering that there is a certain relationship between the water level change and the water inflow of the reservoir to be regulated, the implementer can build a change model between the reservoir water storage and the water level by modeling the reservoir to be regulated and analyzing its structure, so as to evaluate the water level when the total water storage of the reservoir to be regulated is known; therefore, in a preferred embodiment of the present invention, the method for obtaining the second water level change includes: The sum of the water inflow of the reservoir to be regulated and the water storage at the current moment is used as the predicted water storage of the reservoir to be regulated in a preset future time period; based on the preset water storage-water level model and the predicted water storage, the future water level of the reservoir to be regulated in the preset future time period is obtained; the water level at the current moment is subtracted from the future water level to obtain a second water level change.

[0050] As an example, by substituting the obtained predicted water storage into the preset water storage-water level model, the future water level of the reservoir to be regulated can be directly obtained. The future water level minus the current water level is the second water level change.

[0051] It should be noted that the evaluation and acquisition of the preset water storage capacity-water level model can be based on hydrological data or reservoir design drawings, which is an existing technology well known to those skilled in the art and will not be elaborated here.

[0052] Considering that if the changes between the future indicator sequence and the historical indicator sequence under each water volume influencing indicator are more similar within the same time period, the first water level change amount predicted based on the model will be less affected by the seasonal difference, and the prediction result will be more accurate; on the contrary, it is more inclined to refer to the second water level change amount based on the predicted water inflow analysis and evaluation to improve the prediction accuracy; considering that there may be a certain waste of computing resources when analyzing the similarity of changes between the future indicator sequence and the historical indicator sequence under all water volume influencing indicators, some representative indicators can be selected for analysis; therefore, the embodiment of the present invention obtains the predicted water level of the reservoir to be regulated within a preset future time period based on the first water level change amount and the second water level change amount, combined with the similarity of changes between the future indicator sequence and the historical indicator sequence under the preset water volume influencing indicator.

[0053] Preferably, in one embodiment of the present invention, considering that the change in precipitation has the greatest impact on the prediction of the water level, the preset water volume impact index is first set to precipitation; then, in the historical precipitation sequence, a reference sequence of the future precipitation sequence is obtained; according to the sequence difference between the reference sequence and the future precipitation sequence, a first reference weight of the first water level change is obtained; according to the first reference weight, a second reference weight of the second water level change is obtained, and the first reference weight is negatively correlated with the second reference weight and the sum value is 1; The first water level change is weighted using the first reference weight, and the second water level change is weighted using the second reference weight, and the weighted sum is used as the predicted water level change; the sum of the water level at the current moment and the predicted water level change is used as the predicted water level of the reservoir to be regulated within a preset future time period.

[0054] As an example, if the future precipitation sequence is the precipitation sequence from July 1 to July 7 of this year, the corresponding precipitation sequence from July 1 to July 7 of last year in the historical precipitation sequence is used as the reference sequence; then the absolute values ​​of the differences between the elements with the same sequence number between the reference sequence and the future precipitation sequence are averaged, and the mean is used as the x in the exponential function exp(-x) with the natural constant e as the base for negative correlation normalization. The smaller the mean, the more similar the sequences are, and the larger the normalization result. The normalization result is used as the first reference weight of the first water level change; 1 is subtracted from the first reference weight to obtain the second reference weight of the second water level change.

[0055] In other examples, implementers may also use other negative correlation normalization methods, such as adding a very small positive parameter that does not affect the result and then performing a reciprocal operation; implementers may also measure sequence similarity using similarity metrics such as cosine similarity and DTW similarity, which will not be described in detail here.

[0056] Step S4: intelligently regulate the water level of the reservoir to be regulated according to the predicted water level.

[0057] After obtaining the predicted water level of the reservoir to be regulated in a preset future period, the water level can be assessed for risk and adjusted intelligently.

[0058] Preferably, in one embodiment of the present invention, the method for intelligently regulating the water level of a reservoir to be regulated according to the predicted water level includes: if the predicted water level is greater than the preset flood control water level, opening the outlet sluice of the reservoir to be regulated to release water until the water level is lower than the preset flood control water level; if the predicted water level is lower than the preset low water level, closing the outlet sluice of the reservoir to be regulated until the water level is higher than the preset low water level; wherein, the setting of the preset flood control water level and the preset low water level are both related to the terrain design of the reservoir to be regulated, and the implementer can set them according to relevant hydrological data, etc.

[0059] Through the predicted water level of the reservoir to be regulated, the relevant reservoir managers can reasonably regulate the water resource utilization of regional water resource projects, ensure water conservancy safety, and even improve the power generation efficiency of downstream hydropower stations.

[0060] In summary, the present invention first analyzes and obtains the water volume growth parameters of the regional water resources project in a preset future period, and then combines the water volume contribution of each water inlet of the reservoir to be regulated in the regional water resources project to the reservoir to be regulated, and obtains the water inflow of the reservoir to be regulated in the preset future period; then obtains the first water level change based on the preset prediction model and the historical water level sequence, obtains the second water level change based on the inflow, and then analyzes and obtains the predicted water level of the reservoir to be regulated in the preset future period, so as to perform intelligent regulation of the water level of the reservoir to be regulated. The present invention analyzes and evaluates the future water inflow changes of the reservoir to be regulated from the perspective of the indirect impact of water volume on water level to predict the water level; then predicts the water level based on the historical water level change law; and then comprehensively evaluates the predicted water level of the reservoir to be regulated, thereby avoiding the limitations of a single analysis of the historical water level change law, improving the accuracy of the predicted water level, and thus improving the intelligent regulation effect of the reservoir water level.

[0061] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0062] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A method for intelligent water level control in a hydraulic project, characterized in that: The method comprises: In the preset historical period at the current moment, the historical indicator sequence of the regional water resources project under each water volume influencing indicator and the historical water level sequence of the reservoir to be regulated in the regional water resources project are obtained; in the preset future period at the current moment, the future indicator sequence of the regional water resources project under each water volume influencing indicator is obtained; According to the similarity of the change trend between the historical indicator sequence under each water volume influencing indicator and the historical water level sequence, combined with the future indicator sequence under each water volume influencing indicator, the water volume growth parameter of the regional water resources project in the preset future period is obtained; according to the water volume growth parameter and the contribution of each water inlet of the reservoir to be regulated in the regional water resources project to the water volume of the reservoir to be regulated, the water inflow of the reservoir to be regulated in the preset future period is obtained; Based on the preset prediction model and the historical water level sequence, a first water level change of the reservoir to be regulated within a preset future period is obtained; based on the water inflow, a second water level change of the reservoir to be regulated within a preset future period is obtained; based on the first water level change and the second water level change, combined with the similarity of changes between the future indicator sequence and the historical indicator sequence under the preset water volume influencing index, a predicted water level of the reservoir to be regulated within a preset future period is obtained; Intelligently regulating the water level of the reservoir to be regulated according to the predicted water level; The water volume influencing index at least includes precipitation, temperature and total water inflow; The method for obtaining the water volume growth parameter includes: According to the future indicator sequence under each water volume influencing indicator, obtaining the indicator change amount under the corresponding water volume influencing indicator; Under each water quantity impact indicator, according to the Pearson correlation coefficient between the historical indicator sequence and the historical water level sequence, the weight of the impact of the change of each water quantity impact indicator on the water quantity growth of the regional water resources project is obtained; The water volume growth impact weight under each water volume impact index is used to weight the corresponding indicator change, and the weighted sum result is used as the water volume growth parameter of the regional water resources project in a preset future period.

2. The intelligent water level control method for a hydraulic project according to claim 1, characterized in that: The method for obtaining the water volume growth impact weight includes: The positive correlation of the Pearson correlation coefficient under the precipitation index is normalized to obtain the first change correlation coefficient; the positive correlation of the Pearson correlation coefficient under the total water inflow index is normalized to obtain the second change correlation coefficient; the negative correlation of the Pearson correlation coefficient under the temperature index is normalized to obtain the third change correlation coefficient; Normalizing the first change correlation coefficient, the second change correlation coefficient, and the third change correlation coefficient respectively, wherein the sum of the normalized first change correlation coefficient, the second change correlation coefficient, and the third change correlation coefficient is 1; The normalized first change correlation coefficient is used as the weight of water volume growth under the precipitation index; the normalized second change correlation coefficient is used as the weight of water volume growth under the total water inflow index; the negative of the normalized third change correlation coefficient is taken as the weight of water volume growth under the temperature index.

3. The intelligent water level control method for a hydraulic project according to claim 1 is characterized in that: The method for obtaining the water contribution degree includes: In a preset historical period at the current moment, a first historical water inflow sequence of the reservoir to be regulated and a second historical water inflow sequence of each water inlet of the reservoir to be regulated in the regional water resources project corresponding to an upstream water source are obtained; The first historical water inflow sequence of the reservoir to be regulated is used as the dependent variable, and the second historical water inflow sequence of each water inlet is used as the independent variable to construct a linear regression model; each regression coefficient in the linear regression model is used as the first water volume contribution of the corresponding water inlet to the reservoir to be regulated; Normalize the storage capacity of the upstream water source corresponding to each water inlet of the reservoir to be regulated in the regional water resources project, and use the normalized result as the second water volume contribution of the corresponding water inlet to the reservoir to be regulated; The first water volume contribution and the second water volume contribution are integrated to obtain the water volume contribution of the corresponding water inlet to the reservoir to be regulated.

4. The intelligent water level control method for a hydraulic project according to claim 1, characterized in that: The method for obtaining the water inflow comprises: The product of the water contribution of each water inlet to the reservoir to be regulated and the water volume growth parameter is taken as the contributed water volume of each water inlet to the reservoir to be regulated; the contributed water volumes of all water inlets to the reservoir to be regulated are combined to obtain the water inflow of the reservoir to be regulated in a preset future period.

5. The intelligent water level control method for a hydraulic project according to claim 1 is characterized in that: The method for obtaining the first water level change comprises: The preset prediction model is a SARIMA model; the length of the preset historical period is greater than the length of the preset future period; the historical water level sequence is divided into a number of sequence sub-segments with the same length as the preset future period; a model training set is constructed based on all the sequence sub-segments to train the SARIMA model; The sequence sub-segment closest to the current moment is input into the trained SARIMA model, and the predicted water level sequence within a preset future time period is output; based on the water level difference corresponding to the first and last sequence elements in the predicted water level sequence, the first water level change of the reservoir to be regulated within the preset future time period is obtained.

6. The intelligent water level control method for a hydraulic project according to claim 1, characterized in that: The method for obtaining the second water level change comprises: The sum of the water inflow of the reservoir to be regulated and the water storage at the current moment is used as the predicted water storage of the reservoir to be regulated in a preset future time period; based on a preset water storage-water level model and the predicted water storage, the future water level of the reservoir to be regulated in the preset future time period is obtained; the water level at the current moment is subtracted from the future water level to obtain a second water level change.

7. The intelligent water level control method for a hydraulic project according to claim 1, characterized in that: The method for obtaining the predicted water level comprises: The preset water volume influencing index is precipitation; in the historical precipitation sequence, a reference sequence of the future precipitation sequence is obtained; according to the sequence similarity between the reference sequence and the future precipitation sequence, a first reference weight of the first water level change is obtained; according to the first reference weight, a second reference weight of the second water level change is obtained, and the first reference weight is negatively correlated with the second reference weight and the sum value is 1; The first water level change is weighted using the first reference weight, and the second water level change is weighted using the second reference weight, and the weighted sum is used as the predicted water level change; the sum of the water level at the current moment and the predicted water level change is used as the predicted water level of the reservoir to be regulated within a preset future time period.

8. The intelligent water level control method for a hydraulic project according to claim 1 is characterized in that: The method for intelligently regulating the water level of the reservoir to be regulated according to the predicted water level comprises: If the predicted water level is greater than the preset flood control water level, the outlet sluice of the reservoir to be regulated is opened to release water until the water level is lower than the preset flood control water level; if the predicted water level is lower than the preset low water level, the outlet sluice of the reservoir to be regulated is closed until the water level is higher than the preset low water level.

Images