A reservoir operation module evaluation method and system based on information theory indicators

Abstract

The present invention relates to the technical field of hydrology and water resources, and proposes a reservoir scheduling module evaluation method and system based on information theory indicators, which includes the following steps: obtaining the inflow Q, outflow O and reservoir storage V based on time series in the reservoir; calculating the mutual information I(Q) between the inflow at the previous moment and the outflow at the current moment in the reservoir based on information theory; t‑1 ;O t ), the mutual information I(V t‑1 ;O t ), and the mutual information I(Q t‑1 ; V t‑1 ;O t ); Run the reservoir dispatching module to obtain the reservoir runoff simulation value based on the corresponding reservoir dispatching model; calculate the outflow runoff of the reservoir at the current moment t The mutual information between the simulated reservoir runoff values ​​of the corresponding reservoir operation model and the mutual information I(Q t‑1 ;O t )、I(V t‑1 ;O t ) and I(Q t‑1 ; V t‑1 ;O t ) are compared to obtain the evaluation result of the reservoir scheduling module.

Description

Technical Field

[0001] The present invention relates to the technical field of hydrology and water resources, and more specifically, to a reservoir dispatching module evaluation method and system based on information theory indicators. Background Art

[0002] Reservoirs play an important role in the allocation and management of water resources. With the increase in the number of reservoirs around the world, more and more global hydrological models have begun to consider the impact of reservoir construction on hydrological processes. Therefore, reservoir operation modules have become an important component of global hydrological models. Different types of reservoir operation modules used in different global hydrological models such as Water Balance Model plus (WBMplus), Water-Global Assessment and Prognosis 2.0 (WaterGAP 2) and CommunityWater Model (CWatM) have emerged. Faced with complex reservoir operation rules, it is necessary to evaluate the effectiveness of different reservoir operation modules.

[0003] At present, the evaluation of reservoir operation module mainly compares the correlation between observed values ​​and simulated values ​​to illustrate the closeness between simulated values ​​and observed values, so as to evaluate the performance of model simulation. However, common evaluation indicators such as the coefficient of certainty are only the square of the correlation coefficient, and the correlation coefficient can only quantify the linear relationship between observed values ​​and simulated values, and it is difficult to capture nonlinear relationships. Moreover, for the effectiveness of the model, that is, to what extent the model can represent the dependency between input variables and output variables, common evaluation indicators cannot characterize its inherent complex nonlinear relationship. Therefore, for the evaluation of reservoir operation module, the effectiveness of the module and the characteristics of the evaluation indicators that can characterize complex nonlinearity should be considered at the same time. Summary of the invention

[0004] In order to overcome the defect that the evaluation index of the reservoir scheduling module in the above-mentioned prior art can only quantify the linear relationship between the observed value and the simulated value, but is difficult to capture the nonlinear relationship, the present invention provides a reservoir scheduling module evaluation method based on information theory indicators, and a reservoir scheduling module evaluation system based on information theory indicators.

[0005] In order to solve the above technical problems, the technical solution of the present invention is as follows:

[0006] A reservoir operation module evaluation method based on information theory indicators includes the following steps:

[0007] S1, obtain the inflow Q, outflow O and reservoir storage V based on the time series in the reservoir;

[0008] S2, based on information theory, calculate the mutual information I(Q t-1 ;O t ), the mutual information I(V t-1 ;O t ), and the mutual information I(Q t-1 ; V t-1 ;O t );

[0009] S3, running a reservoir dispatching module, wherein the reservoir dispatching module is preset with several reservoir dispatching models, and obtaining a reservoir runoff simulation value based on the corresponding reservoir dispatching model;

[0010] S4. Calculate the outflow O of the reservoir at the current moment based on information theory t The mutual information between the simulated reservoir runoff values ​​of the corresponding reservoir operation model and the mutual information I(Q t-1 ;O t )、I(V t-1 ;O t ) and I(Q t-1 ; V t-1 ;O t ) are compared to obtain the evaluation result of the reservoir scheduling module.

[0011] In the technical scheme, firstly, the time series data of the reservoir's inflow, outflow and reservoir storage are obtained, and the inflow and outflow, the reservoir storage and outflow, and the mutual information values ​​of the inflow, reservoir storage and outflow are calculated respectively. Then, the reservoir scheduling module under the global hydrological model is run to obtain the simulated reservoir outflow value, and then the mutual information between the observed outflow value and the simulated outflow value is calculated, and then the mutual information values ​​of the comprehensive elements are compared, so as to fully consider the effectiveness of the reservoir scheduling module and the characteristics that the evaluation indicators can characterize the complex nonlinearity between variables, and avoid the one-sidedness of ordinary evaluation indicators that only consider linear relationships.

[0012] Furthermore, the present invention also proposes a reservoir dispatching module evaluation system based on information theory indicators, which is applied to the above-mentioned reservoir dispatching module evaluation method based on information theory indicators. It includes a reservoir data acquisition module, a mutual information calculation module, a reservoir dispatching module and an evaluation module, specifically:

[0013] The reservoir data acquisition module is used to obtain the inflow Q, outflow O and reservoir storage V based on the time series in the reservoir;

[0014] The mutual information calculation module is used to calculate the mutual information I(Q t-1 ;O t ), the mutual information I(V t-1 ;O t ), and the mutual information I(Q t-1 ; V t-1 ;O t ) and is used to calculate the outflow O of the reservoir at the current moment t The mutual information between the simulated reservoir runoff values ​​and the corresponding reservoir operation model;

[0015] The reservoir dispatching module is preset with several reservoir dispatching models, the data collected by the reservoir data acquisition module is input into the reservoir dispatching module for operation, and the reservoir runoff simulation value based on the corresponding reservoir dispatching model is output;

[0016] The evaluation module is used to calculate the outflow flow of the reservoir output by the mutual information calculation module. t The mutual information between the reservoir runoff simulation value of the corresponding reservoir operation model and the mutual information I(Q t-1 ;O t )、I(V t-1 ;O t ) and I(Q t-1 ; V t-1 ;O t ) are compared to generate the evaluation results of the reservoir scheduling module.

[0017] Furthermore, the present invention also proposes a reservoir scheduling module evaluation system based on information theory indicators, which includes a processor and a memory, and a computer program is stored in the memory. When the processor executes the computer program in the memory, the steps of the above-mentioned reservoir scheduling module evaluation method are implemented.

[0018] Compared with the prior art, the beneficial effect of the technical solution of the present invention is as follows: the present invention evaluates the effectiveness of the reservoir scheduling module based on information theory indicators by comparing the mutual information values ​​between the comprehensive inflow and outflow runoff and the reservoir water storage with the mutual information values ​​of the simulation values ​​under different reservoir scheduling modules, quantifies the complex dependency relationship between the input variables and the output variables and the maximum possible information value captured, and compares them with the information value captured by the simulation value, ensuring the correlation between the simulation value and the observation value without losing all data information, and avoiding the situation where only linear relationships are considered and nonlinear relationships are ignored. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is a flow chart of the reservoir scheduling module evaluation method based on information theory indicators in Example 1.

[0020] Figure 2 This is a schematic diagram of the location of the FOL reservoir in Example 3 and an elevation distribution map of the area controlled by the reservoir.

[0021] Figure 3 Schematic diagram of the monthly average inflow, outflow and reservoir storage capacity of the FOL reservoir in Example 3.

[0022] Figure 4 It is a three-dimensional scatter plot of the inflow runoff, outflow runoff and reservoir storage capacity of the FOL reservoir in Example 3.

[0023] Figure 5 Schematic diagram of mutual information value data comparison in Example 3.

[0024] Figure 6 It is a scatter plot of the observed and simulated outflow runoff in January, April, July and October of Example 3.

[0025] Figure 7 This is an architectural diagram of the reservoir scheduling module evaluation system based on information theory indicators in Example 4. DETAILED DESCRIPTION

[0026] The drawings are for illustrative purposes only and should not be construed as limiting the present patent;

[0027] It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0028] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0029] Example 1

[0030] This embodiment proposes a reservoir operation module evaluation method based on information theory indicators, such as Figure 1 As shown, it is a flow chart of the reservoir scheduling module evaluation method based on information theory indicators of this embodiment.

[0031] The reservoir operation module evaluation method based on information theory indicators proposed in this embodiment includes the following steps:

[0032] S1. Obtain the inflow Q, outflow O and reservoir storage V based on time series in the reservoir.

[0033] In this step, the obtained inflow Q, outflow O and reservoir storage V are stored in time series to obtain the inflow Q t ={q0,q1,...,q N-1}, outflow Ot ={o0,o1,...,o N-1}, reservoir water storage capacity V t ={v0,v1,...,v N-1}, where N is the length of the time series.

[0034] S2, based on information theory, calculate the mutual information I(Q t-1 ;O t ), the mutual information I(V t-1 ;O t ), and the mutual information I(Q t-1 ; V t-1 ;O t ).

[0035] In this step, the mutual information I(Q t-1 ;O t )、I(V t-1 ;O t ) and I(Q t-1 ; V t-1 ;O t ).

[0036] Specifically, the mutual information I(Q t-1 ;O t ), and its expression formula is as follows:

[0037]

[0038] Based on information theory, the mutual information I(V) between the reservoir water storage at the previous moment and the outflow at the current moment is calculated. t-1 ;O t ), and its expression formula is as follows:

[0039]

[0040] Based on information theory, the mutual information I(Q) between the inflow at the previous moment, the reservoir storage and the outflow at the current moment is calculated. t-1 ; V t-1 ;O t ), and its expression formula is as follows:

[0041]

[0042] In the above formula, p(·) is the probability density function, p(·,·) is the joint probability density function, and p(·,·,·) is the multidimensional joint density function.

[0043] S3. Running a reservoir dispatching module, in which several reservoir dispatching models are preset, to obtain reservoir runoff simulation values ​​based on corresponding reservoir dispatching models.

[0044] The reservoir dispatching module in this embodiment adopts the reservoir dispatching module under the global hydrological model, in which several reservoir dispatching models are preset, such as the WBMplus model, the WaterGAP2 model and the CWatM model. The inflow Q, outflow O and reservoir storage V collected in step S1 are input into the reservoir dispatching module to obtain the simulated outflow value O′ of the reservoir.

[0045] S4. Calculate the outflow O of the reservoir at the current moment based on information theory t The mutual information between the simulated reservoir runoff values ​​of the corresponding reservoir operation model and the mutual information I(Q t-1 ;O t )、I(V t-1 ;O t ) and I(Q t-1 ; V t-1 ;O t ) are compared to obtain the evaluation result of the reservoir scheduling module.

[0046] In this step, the outflow O of the reservoir at the current moment is first calculated. t The mutual information value I′ between the simulated reservoir outflow value O′ and the mutual information value I(Q t-1 ;O t )、I(V t-1 ;O t ) and I(Q t-1 ; V t-1 ;O t ) are compared to obtain the evaluation result of the reservoir scheduling module.

[0047] This embodiment evaluates the effectiveness of the reservoir scheduling module based on information theory indicators by comparing the mutual information values ​​of the inflow and outflow, the reservoir water storage and the outflow, and the inflow, the reservoir water storage and the outflow with the mutual information values ​​of the simulation values ​​under different reservoir scheduling modules, quantifies the complex dependency between the input variables and the output variables and the maximum possible information value, and compares them with the information value captured by the simulation value, to ensure the correlation between the simulation value and the observation value without losing all data information, and avoids considering only linear relationships and ignoring nonlinear relationships.

[0048] Example 2

[0049] This embodiment proposes a specific implementation process based on the reservoir scheduling module evaluation method based on information theory indicators proposed in Example 1. Specifically, the reservoir scheduling models preset in the reservoir scheduling module in this embodiment include the WBMplus model, the WaterGAP2 model and the CWatM model.

[0050] In step S3, the time series-based inflow Q, outflow O and reservoir storage V in the reservoir are input into the reservoir scheduling module, and the reservoir scheduling module runs and calls its preset reservoir scheduling model to obtain the reservoir runoff simulation value of the corresponding reservoir scheduling model.

[0051] Among them, the reservoir runoff simulation value O based on the WBMplus model WBMplus,t The expression formula is:

[0052]

[0053] Where, κ and λ are model parameters of the WBMplus model. In this embodiment, κ is 0.16 and λ is 0.6; Q in,avg is the historical mean of the reservoir's inflow.

[0054] Simulated value of reservoir runoff based on WaterGAP2 model WaterGAP2,t The expression formula is:

[0055]

[0056] Where V max and V min are the maximum and minimum values ​​of the reservoir water storage, respectively, r is the outflow coefficient, in this embodiment, k r The value is 0.01; Δt represents the time step.

[0057] Simulated value of reservoir runoff based on CWatM model CWatM,t The expression formula is:

[0058]

[0059] Where V total is the total storage capacity of the reservoir, F is the reservoir filling rate; L c To protect the water storage limit, the value is taken as 0.1; L n is the normal water storage limit, which is 0.5; L f is the flood storage limit, which is 0.9; min is the minimum outflow, set to 20% of the average flow; norm is the normal outflow, taking the average flow value; Ond For non-destructive outflow, set to 400% of the average flow rate.

[0060] Furthermore, in step S4, the outflow O of the reservoir at the current moment is calculated. t The steps of calculating the mutual information between the simulated reservoir runoff values ​​and the corresponding reservoir operation model are as follows:

[0061] Calculate the outflow O of the reservoir at the current moment t Compared with the reservoir runoff simulation value O based on WBMplus model WBMplus,t The mutual information I(O WBMplus,t ;O t ), and its expression formula is as follows:

[0062]

[0063] Calculate the outflow O of the reservoir at the current moment t Compared with the reservoir runoff simulation value O based on WaterGAP2 model WaterGAP2,t The mutual information I(O WaterGAP2,t ;O t ), and its expression formula is as follows:

[0064]

[0065] Calculate the outflow O of the reservoir at the current moment t Compared with the reservoir runoff simulation value O based on CWatM model CWatM,t The mutual information I(O CWatM,t ;O t ), and its expression formula is as follows:

[0066]

[0067] Where, outflow O t ={o0,o1,...,o N-1}, N is the length of the time series; p(·) is the probability density function, and p(·,·) is the joint probability density function.

[0068] In this embodiment, the mutual information value I(Q t-1 ;O t )、I(V t-1 ;O t ) and I(Q t-1 ; V t-1 ;O t ) and the mutual information value I(O WBMplus,t ;O t )、I(O WaterGAP2,t ;O t ) and I(O CWatM,t ;O t) satisfy the following constraints:

[0069] I(O WBMplus,t ;O t )≤I(Q t-1 ;O t ),

[0070] I(O WaterGAP2,t ;O t )≤I(V t-1 ;O t ),

[0071] I(O CWatM,t ;O t )≤I(Q t-1 ; V t-1 ;O t ).

[0072] In this embodiment, the mutual information value I(Q t-1 ;O t )、I(V t-1 ;O t ) and I(Q t-1 ; V t-1 ;O t ) are respectively related to the mutual information value I(O WBMplus,t ;O t )、I(O WaterGAP2,t ;O t ) and I(O CWatM,t ;O t ) are compared to reflect the effectiveness of information theory indicators and their characteristics in characterizing complex nonlinear relationships between variables. It is found that the reservoir operation module under the CWatM model has better simulation performance.

[0073] Example 3

[0074] This embodiment conducts experiments based on the reservoir scheduling module evaluation method based on information theory indicators proposed in Embodiment 2.

[0075] This embodiment uses the time series data of reservoir inflow, outflow and reservoir storage on a daily scale provided by the California Data Exchange Center (CDEC), specifically taking the Folsom (FOL) reservoir as an example.

[0076] like Figure 2The figure shows the location of the FOL Reservoir and the elevation distribution of the area controlled by the reservoir. The FOL Reservoir was built by the United States Army Corps of Engineers (USACE) in 1955 with a total capacity of 1.20 cubic kilometers. Located on the American River in northern California, the FOL Reservoir plays an important role in hydroelectric power generation, irrigation, and municipal water supply, especially in flood control. It collects runoff from the American River basin and is a key flood protection for the Sacramento area.

[0077] like Figure 3 The following table shows the monthly averages of the inflow, outflow and storage capacity of the FOL reservoir. The maximum monthly average inflow occurred in March, reaching 196m 3 / s, the minimum value appeared in October, only 29m 3 / s. The relative values ​​of inflow and outflow show seasonal differences. In January-May, the inflow is greater than the outflow, while the opposite is true in other months. The peak value of reservoir water storage has a certain lag.

[0078] like Figure 4 Shown is a three-dimensional scatter plot of the normalized inflow, outflow and reservoir storage capacity of the FOL reservoir. Figure 4 The midpoints are represented by different grayscale values ​​to help identify different intervals of reservoir water storage.

[0079] Depend on Figure 4 It can be seen that the reservoir water storage is not large and the water level is relatively low in January, February, March and December. Therefore, when the inflow runoff increases, part of the inflow can be stored in the reservoir, resulting in less outflow runoff than inflow runoff. In April, May and June, the reservoir water storage gradually increases to reach the total capacity of the reservoir, resulting in the reservoir release mainly relying on the reservoir inflow because there is no empty volume to store the inflow. Although the inflow and outflow runoff in August, September and October decreased significantly, the water storage did not decrease, which means that most of the water in the reservoir is stored and a small part is released to provide ecological flow. The above situation shows that there is a complex nonlinear relationship between inflow runoff, outflow runoff and reservoir water storage.

[0080] like Figure 5 As shown, we will I(Q t-1 ;O t ), I(V t-1 ;O t ) and I(Q t-1 ,V t-1 ;O t ) respectively with I(O WBMplus,t ;O t ), I(O WaterGAP2,t ;O t ) and I(O CWatM,t ;O t) is a comparative diagram. Wherein, a), b), and c) are the mutual information values ​​I( WBMplus,t ;O t ) and the mutual information value I(Q t-1 ;O t ),、I(V t-1 ;O t ) and I(Q t-1 ,V t-1 ;O t ), d), e), and f) are the mutual information values ​​I(O WaterGAP2,t ;O t ) and the mutual information value I(Q t-1 ;O t ),、I(V t-1 ;O t ) and I(Q t-1 ,V t-1 ;O t ), g), h), i) are the mutual information values ​​I(O CWatM,t ;O t ) and the mutual information value I(Q t-1 ;O t ),、I(V t-1 ;O t ) and I(Q t-1 ,V t-1 ;O t ) is a comparative diagram.

[0081] Figure 5 The points in a), e) and i) are below the 1:1 line, which is consistent with information theory, that is, the information of the observed value that can be captured by the simulated value has an upper limit, and this upper limit is the information of the observed value. Due to the limitation of model structure or parameters, the simulated value will lose some information compared with the observed value. However, in Figure 5 In g) and h), these points are almost above the 1:1 line, which indicates that the information captured by the CWatM model exceeds the mutual information value I(Q t-1 ;O t ) and I(V t-1 ;O t )information. Figure 5 The point ratio in i) Figure 5 The points in c) and f) are closer to the 1:1 line, which means that the outflow simulated by the reservoir operation module under the CWatM model reduces the O t The uncertainty of the reservoir operation module under the CWatM model is reduced more than that under the WBMplus model and the WaterGAP2 model. This shows that the simulation performance of the reservoir operation module under the CWatM model is better. Figure 5In b) and d), nearly half of the points are above the 1:1 line and the other half are below the 1:1 line, while some points representing months are just the opposite, which indicates that the reservoir operation modules under the WBMplus model and WaterGAP 2 model show different applicability in different seasons.

[0082] Further, such as Figure 6 As shown in FIG. 1 , it is a scatter plot of the observed and simulated outflow runoff in January, April, July, and October in this embodiment. The scatter plot of the observed and simulated outflow runoff is calculated using linear regression and the coefficient of certainty r 2 Draw it. Figure 6 It can be seen that the r of the CWatM model 2 The values ​​are larger than those of the WaterGAP 2 model and the WBMplus model, and the line fitted by the least squares method of the CWatM model is closer to the 1:1 line than the other two models, which is consistent with Figure 5 The same results are obtained for the information-theoretic indicators shown in .

[0083] However, the highest r of the WaterGAP 2 model 2 The values ​​appear in October, while for the WBMplus model, the best fit of the linear regression occurs in January, which is not an exact match. Figure 5 The seasonal variation of the mutual information index in the is due to the linear regression and r 2 Only the linear relationship between variables is considered, without involving nonlinear correlation, resulting in incomplete results.

[0084] The above experimental results show that the reservoir scheduling module evaluation method based on information theory indicators proposed in this embodiment has the characteristics of being able to effectively consider the effectiveness and evaluation indicators of the module and being able to characterize complex nonlinearities.

[0085] Example 4

[0086] This embodiment proposes a reservoir dispatching module evaluation system based on information theory indicators, which is applied to the reservoir dispatching module evaluation method based on information theory indicators proposed in Embodiment 1 or Embodiment 2. Figure 7 , which is an architecture diagram of the reservoir scheduling module evaluation system based on information theory indicators of this embodiment.

[0087] The reservoir dispatching module evaluation system based on information theory indicators in this embodiment includes a reservoir data acquisition module, a mutual information calculation module, a reservoir dispatching module and an evaluation module.

[0088] Among them, the reservoir data acquisition module is used to obtain the inflow Q, outflow O and reservoir storage V based on time series in the reservoir.

[0089] The mutual information calculation module is used to calculate the mutual information I(Q t-1 ;O t ), the mutual information I(V t-1 ;O t ), and the mutual information I(Q t-1 ; V t-1 ;O t ) and is used to calculate the outflow O of the reservoir at the current moment t The mutual information between the simulated reservoir runoff values ​​and the corresponding reservoir operation model.

[0090] The reservoir dispatching module is preset with several reservoir dispatching models to be evaluated. The data collected by the reservoir data acquisition module is input into the reservoir dispatching module for operation, and the reservoir runoff simulation value based on the corresponding reservoir dispatching model is output.

[0091] The evaluation module is used to calculate the outflow flow of the reservoir output by the mutual information calculation module. t The mutual information between the reservoir runoff simulation value of the corresponding reservoir operation model and the mutual information I(Q t-1 ;O t )、I(V t-1 ;O t ) and I(Q t-1 ; V t-1 ;O t ) are compared to generate the evaluation results of the reservoir scheduling module.

[0092] In the specific implementation process, the reservoir data acquisition module collects the inflow Q, outflow O and reservoir storage V in the reservoir and stores them in time series. The reservoir data acquisition module sends the collected data to the mutual information calculation module.

[0093] The mutual information calculation module calculates the mutual information I(Q t-1 ;O t ), the mutual information I(V t-1 ;O t ), and the mutual information I(Q t-1 ; V t-1 ;O t ).

[0094] At the same time, the reservoir data acquisition module sends the collected data to the reservoir dispatching module, and the reservoir dispatching module outputs the reservoir runoff simulation value O' based on the corresponding reservoir dispatching model after operation. The reservoir data acquisition module sends the output reservoir runoff simulation value O' to the mutual information calculation module.

[0095] The mutual information calculation module calculates the outflow O of the reservoir at the current moment based on information theory. t The mutual information I′ between the reservoir runoff simulation value O′ of the corresponding reservoir operation model, and then the mutual information value I(Q t-1 ;O t )、I(V t-1 ;O t ) and I(Q t-1 ; V t-1 ;O t ), and the mutual information I′ are sent to the evaluation module to evaluate the reservoir scheduling module and generate the evaluation result of the reservoir scheduling module.

[0096] In another specific embodiment, the reservoir dispatching module includes a WBMplus model unit, a WaterGAP2 model unit and a CWatM model unit; the data collected by the reservoir data acquisition module is input into the reservoir dispatching module for operation, and the reservoir runoff simulation value based on the WBMplus model is output. WBMplus,t , reservoir runoff simulation value O based on WaterGAP2 model WaterGAP2,t and the reservoir runoff simulation value O based on the CWatM model CWatM,t .

[0097] The reservoir dispatching module evaluation system based on information theory indicators proposed in this embodiment fully considers the effectiveness of the reservoir dispatching module based on information theory indicators and the characteristics that the evaluation indicators can characterize the complex nonlinearity between variables, avoiding the one-sidedness of ordinary evaluation indicators that only consider linear relationships.

[0098] Example 5

[0099] This embodiment proposes a reservoir scheduling module evaluation system based on information theory indicators, which is applied to the reservoir scheduling module evaluation method based on information theory indicators in the above-mentioned embodiment 1 or embodiment 2.

[0100] The reservoir scheduling module evaluation system based on information theory indicators proposed in this embodiment includes a processor and a memory, the memory stores a computer program, and the processor implements the steps of the above-mentioned reservoir scheduling module evaluation method when executing the computer program in the memory.

[0101] The same or similar reference numerals correspond to the same or similar components;

[0102] The terms used in the drawings to describe positional relationships are only used for illustrative purposes and should not be construed as limiting this patent;

[0103] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the embodiments here. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A reservoir operation module evaluation method based on information theory indicators, characterized in that: The following steps are involved: S1, obtain the inflow Q, outflow O and reservoir storage V based on the time series in the reservoir; S2, based on information theory, calculate the mutual information I(Q t-1 ;O t ), the mutual information I(V t-1 ;O t ), and the mutual information I(Q t-1 ; V t-1 ;O t ); S3, running the reservoir dispatching module, wherein the reservoir dispatching module is preset with several reservoir dispatching models, and obtaining the reservoir runoff simulation value based on the corresponding reservoir dispatching model; the reservoir dispatching models preset in the reservoir dispatching module include the WBMplus model, the WaterGAP2 model and the CWatM model; the inflow runoff Q, the outflow runoff O and the reservoir storage capacity V based on the time series in the reservoir are input into the reservoir dispatching module, and the reservoir dispatching module runs and calls its preset reservoir dispatching model to obtain the reservoir runoff simulation value O based on the WBMplus model WBMplus,t , reservoir runoff simulation value based on WaterGAP2 model WaterGAP2,t and the reservoir runoff simulation value O based on the CWatM model CWatM,t ; S4. Calculate the outflow O of the reservoir at the current moment based on information theory t The mutual information between the simulated reservoir runoff values ​​of the corresponding reservoir operation model and the mutual information I(Q t-1 ;O t )、I(V t-1 ;O t ) and I(Q t-1 ; V t-1 ;O t ) to obtain the evaluation result of the reservoir dispatching module; among which, the outflow O of the reservoir at the current moment is calculated t The steps of calculating the mutual information between the simulated reservoir runoff values ​​and the corresponding reservoir operation model are as follows: Calculate the outflow O of the reservoir at the current moment t Compared with the reservoir runoff simulation value O based on WBMplus model WBMplus,t The mutual information I(O WBMplus,t ;O t ), and its expression formula is as follows: Calculate the outflow O of the reservoir at the current moment t Compared with the reservoir runoff simulation value O based on WaterGAP2 model WaterGAP2,t The mutual information I(O WaterGAP2,t ;O t ), and its expression formula is as follows: Calculate the outflow O of the reservoir at the current moment t Compared with the reservoir runoff simulation value O based on CWatM model CWatM,t The mutual information I(O CWatM,t ;O t ), and its expression formula is as follows: Where, outflow O t ={o0,o1,...,o N-1 }, N is the length of the time series; p(·) is the probability density function, and p(·,·) is the joint probability density function.

2. The reservoir operation module evaluation method based on information theory indicators according to claim 1 is characterized in that: In the step S2, the mutual information I(Q t-1 ;O t )、I(V t-1 ;O t ) and I(Q t-1 ; V t-1 ;O t ).

3. The reservoir operation module evaluation method based on information theory indicators according to claim 2 is characterized in that: In the step S2, the specific steps are as follows: S2.

1. Calculate the mutual information I(Q) between the inflow at the last moment and the outflow at the current moment in the reservoir based on information theory t-1 ;O t ), and its expression formula is as follows: S2.2, Based on information theory, calculate the mutual information I(V) between the reservoir water storage at the previous moment and the outflow at the current moment t-1 ;O t ), and its expression formula is as follows: S2.3, Based on information theory, calculate the mutual information I(Q) between the inflow at the previous moment, the reservoir storage and the outflow at the current moment t-1 ; V t-1 ;O t ), and its expression formula is as follows: In the formula, the inflow Q t-1 ={q0,q1,...,q N-1 }, outflow O t ={o0,o1,...,o N-1 }, reservoir water storage capacity V t-1 ={v0,v1,...,v N-1 }, where N is the length of the time series; p(·) is the probability density function, p(·,·) is the joint probability density function, and p(·,·,·) is the multidimensional joint density function.

4. The reservoir operation module evaluation method based on information theory indicators according to claim 1 is characterized in that: In the step S3, the reservoir runoff simulation value O based on the WBMplus model WBMplus,t The expression formula is: Where κ and λ are the model parameters of the WBMplus model, Q in,avg is the historical mean of the reservoir’s inflow runoff; Simulated value of reservoir runoff based on WaterGAP2 model WaterGAP2,t The expression formula is: Where V max and V min are the maximum and minimum values ​​of the reservoir water storage, respectively, r is the outflow coefficient, Δt represents the time step; Simulated value of reservoir runoff based on CWatM model CWatM,t The expression formula is: Where V total is the total storage capacity of the reservoir, F is the reservoir filling rate; L c To protect the water storage limit, L n is the normal water storage limit, L f The flood storage limit; min is the minimum outflow, O norm is the normal outflow, O nd It is a non-destructive outflow.

5. The reservoir operation module evaluation method based on information theory indicators according to any one of claims 1 to 4, characterized in that: The mutual information value I(Q t-1 ;O t )、I(V t-1 ;O t ) and I(Q t-1 ; V t-1 ;O t ) and the mutual information value I(O WBMplus,t ;O t )、I(O WaterGAP2,t ;O t ) and I(O CWatM,t ;O t ) satisfy the following constraints: I(O WBMplus,t ;O t )≤I(Q t-1 ;O t ) I(O WaterGAP2,t ;O t )≤I(V t-1 ;O t ) I(O CWatM,t ;O t )≤I(Q t-1 ;V t-1 ;O t )。 6. A reservoir dispatching module evaluation system based on information theory indicators, applying the reservoir dispatching module evaluation method based on information theory indicators according to any one of claims 1 to 5, characterized in that: include: Reservoir data acquisition module, used to obtain the inflow Q, outflow O and reservoir storage V based on time series in the reservoir; The mutual information calculation module is used to calculate the mutual information I(Q t-1 ;O t ), the mutual information I(V t-1 ;O t ), and the mutual information I(Q t-1 ; V t-1 ;O t ) and is used to calculate the outflow O of the reservoir at the current moment t The mutual information between the simulated reservoir runoff values ​​and the corresponding reservoir operation model; A reservoir dispatching module, wherein the reservoir dispatching module is preset with several reservoir dispatching models, the data collected by the reservoir data acquisition module is input into the reservoir dispatching module for operation, and the reservoir runoff simulation value based on the corresponding reservoir dispatching model is output; the module includes a WBMplus model unit, a WaterGAP2 model unit and a CWatM model unit; the data collected by the reservoir data acquisition module is input into the reservoir dispatching module for operation, and the reservoir runoff simulation value based on the WBMplus model is output. WBMplus,t , reservoir runoff simulation value O based on WaterGAP2 model WaterGAP2,t and the reservoir runoff simulation value O based on the CWatM model CWatM,t ; An evaluation module is used to calculate the outflow O of the reservoir output by the mutual information calculation module. t The mutual information between the reservoir runoff simulation value of the corresponding reservoir operation model and the mutual information I(Q t-1 ;O t )、I(V t-1 ;O t ) and I(Q t-1 ; V t-1 ;O t ) are compared to generate the evaluation results of the reservoir scheduling module.

7. A reservoir dispatching module evaluation system based on information theory indicators, characterized in that: The method comprises a processor and a memory, wherein a computer program is stored in the memory, and when the processor executes the computer program in the memory, the steps of the reservoir scheduling module evaluation method according to any one of claims 1 to 5 are implemented.

Images