Method, system and electronic equipment for measuring efficiency of water-energy-food system in irrigation area and driving analysis of spatial and temporal heterogeneity

Abstract

The application discloses a kind of irrigation area water-energy-food system efficiency measure and spatio-temporal heterogeneity driving analysis method, system and electronic equipment, constructs including input index, desired output index and non-desired output index efficiency evaluation index system;Obtain the input index data, desired output index data and non-desired output index data of each evaluation period in each irrigation area;According to efficiency evaluation index system, establish efficiency measure model, and obtain the corresponding WEFSE of each evaluation period in each irrigation area;On the basis of obtaining WEFSE, construct driving analysis data set, regression analysis is carried out using spatio-temporal geographic weighted regression model, and the spatio-temporal heterogeneity result of driving factor influence is obtained.The present application reflects the characteristics of multi-element non-synchronous control and ecological environment constraint by constructing efficiency evaluation index system, based on efficiency measure model, carbon dioxide emission and grey water emission are included in efficiency measure as non-desired output, and spatio-temporal heterogeneity is described.

Description

Technical Field

[0001] This invention relates to the field of efficiency measurement and spatiotemporal heterogeneity-driven analysis technology, and in particular to a method, system, and electronic equipment for measuring the efficiency of an irrigation district water-energy-food system and for spatiotemporal heterogeneity-driven analysis. Background Technology

[0002] Data Envelopment Analysis (DEA) is a mathematical programming model-based method for evaluating the relative efficiency of production possibility sets composed of decision-making units. It has been widely applied in water resource utilization efficiency, agricultural ecological efficiency, and energy utilization rate evaluation. Traditional DEA models are mostly based on the radial assumption, typically assuming that input and output indicators change in the same proportion. This differs from the reality that irrigation districts' water, land, and energy factors exhibit different control ratios and asynchronous improvements in actual management, making it difficult to accurately identify input redundancy, output insufficiency, and their constraints. To overcome these problems, a super-efficiency model based on undesirable outputs using slack variables can characterize efficiency levels from the slack in input and output, allowing each indicator to be adjusted according to its own proportion. Furthermore, considering the super-efficiency of undesirable outputs allows for the differentiation and ranking of decision-making units at the efficiency frontier, and considers pollution load and carbon emissions as undesirable outputs to reflect the ecological and environmental constraints of the irrigation district's water-energy-food system, thus making it more suitable for evaluating the sustainable efficiency of the irrigation district's water-energy-food system.

[0003] However, the above studies still have the following problems: the explanation of efficiency driving mechanism mostly adopts static or global regression analysis, which makes it difficult to characterize the differences in the impact of driving factors such as socio-economic, hydrological and ecological factors on efficiency in different irrigation areas and at different times, that is, the spatiotemporal heterogeneity of driving effects, thus limiting the application of evaluation results in the refined management and regional regulation of irrigation areas. Summary of the Invention

[0004] The purpose of this invention is to address the problems in existing efficiency evaluations of irrigation district water-energy-food systems, which struggle to simultaneously reflect the asynchronous regulation characteristics of multiple factors and ecological constraints, and whose driving mechanisms are difficult to characterize spatiotemporal heterogeneity. To solve these problems, this invention introduces an efficiency measurement model that considers undesirable outputs in the efficiency measurement stage. By using slack variables to characterize the redundancy of each input factor, the insufficiency of desired outputs, and the excess of undesirable outputs, a comprehensive reflection of the asynchronous regulation characteristics of multiple factors and ecological constraints is achieved within a unified framework. Based on this, a spatiotemporal weighted regression model is further introduced to conduct spatiotemporal local regression analysis on the driving mechanism of synergistic efficiency. This yields results showing the changes in the direction and intensity of the driving factors' influence with spatial location and time, thereby revealing the spatiotemporal heterogeneity of the driving mechanism.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] In a first aspect, this invention proposes a method for measuring the efficiency of an irrigation district's water-energy-food system and for analyzing its spatiotemporal heterogeneity. This method includes:

[0007] Construct an efficiency evaluation index system that includes input indicators, expected output indicators, and unexpected output indicators;

[0008] Data on input indicators, expected output indicators, and unexpected output indicators for each irrigation district during each evaluation period were obtained to construct input data for efficiency measurement.

[0009] An efficiency measurement model is established based on the efficiency evaluation index system, and WEFSE corresponding to each irrigation district in each evaluation period is obtained based on the efficiency measurement input data.

[0010] Based on the obtained WEFSE, a driving analysis dataset is constructed, and a spatiotemporal geographic weighted regression model is used for regression analysis to obtain the spatiotemporal heterogeneity results of the driving factors.

[0011] More preferably, the input indicators include water resource input indicators, arable land resource input indicators, and energy input indicators;

[0012] The expected output indicators include grain output and grain output value, which are used to characterize the material output and economic benefits of agricultural production in the irrigation area.

[0013] The undesirable output indicators include carbon dioxide emissions generated during agricultural production and grey water emissions corresponding to agricultural non-point source pollution, which are used to characterize the ecological and environmental constraints of the irrigation district's water-energy-food system.

[0014] More preferably, the water resource input indicators include agricultural blue water input and agricultural green water input, the arable land resource input indicators are expressed by crop planting area, and the energy input indicators include the converted values ​​of electricity, diesel, fertilizer, pesticide and film energy consumption.

[0015] More preferably, the efficiency measurement model is established, using slack variables to characterize input redundancy, expected output deficiency, and undesired output redundancy. The objective function expression is as follows:

[0016] ;

[0017] In the formula, This indicates the WEFSE corresponding to each irrigation district in each evaluation period. Indicates the first The i-th input of a decision-making unit This represents the projected value of the i-th input. Indicates the first The r-th expected output of a decision-making unit Indicates the first The r-th undesired output of a decision-making unit This represents the projected value of the r-th expected output. This represents the projected value of the r-th undesired output. These represent the quantities of expected and unexpected output, respectively. Indicates the number of input indicators. This represents the projection value of the input vector. Indicates the first The input vector of each decision-making unit Indicates the first The input vector of each decision-making unit Indicates the first The linear combination weights of each decision unit. Indicates the first The undesired output vector of each decision unit Indicates the first The expected output vector of each decision unit Indicates the first The undesired output vector of each decision unit This represents the projection value of the undesired output vector. This represents the projection value of the expected output vector. Indicates the first The expected output vector of each decision-making unit.

[0018] More preferably, the process of constructing the driving analysis dataset is as follows:

[0019] Using the WEFSE of each irrigation district within each evaluation period as the dependent variable, and the irrigation district number and year as the sample index, for each sample, the dependent variable, driving factor index, irrigation district spatial coordinates, and year are recorded to construct the driving analysis dataset. The expression is as follows:

[0020] ;

[0021] In the formula, This represents the dependent variable for observation point i in the irrigation district during the evaluation period. This represents the spatial coordinates of observation point i in the irrigation area. Indicates the year of the irrigation district evaluation; The intercept term represents the observation points in the irrigation area; Indicators of driving factors Local regression coefficients at observation points in the irrigation area; This represents the random error term.

[0022] More preferably, spatiotemporal distance is introduced to characterize the comprehensive differences between different observation points in terms of spatial location and time. The formula for calculating spatiotemporal distance is as follows:

[0023] ;

[0024] In the formula, This indicates the degree of combined difference between observation point i and observation point j in both spatial and temporal dimensions. This represents the scale factor used to balance the effects of spatial and temporal distances under different units of measurement. and This represents the difference in spatial coordinates between the irrigation areas corresponding to observation point i and observation point j. This represents the difference between the corresponding observed samples in the time dimension.

[0025] More preferably, the regression analysis using a spatiotemporal geographic weighted regression model is performed as follows:

[0026] The spatiotemporal weighted regression model uses weighted least squares to estimate local regression coefficients, and the output calculation formula is as follows:

[0027] ;

[0028] In the formula, This indicates that the k-th component of the estimated local regression coefficient at observation point i in the irrigation area is... , Represents the matrix of independent variables. Represents the matrix of independent variables transpose; It is an n×n diagonal matrix, whose diagonal elements represent the spatiotemporal weights of other observation points relative to observation point i; The dependent variable is the observation point i in the irrigation area. The vector formed;

[0029] The sample weights are determined by the spatiotemporal distance between observation points, giving higher weights to samples that are closer in both space and time, in order to characterize the non-stationarity of the driving factors' influence in the spatial and temporal dimensions. The formula is as follows:

[0030] ;

[0031] In the formula, where Indicates a fixed kernel function. Indicates spatiotemporal bandwidth. This represents the sample weight of observation point j relative to regression point i;

[0032] ;

[0033] in, This represents the total number of observation samples in the irrigation area.

[0034] More preferably, based on the local regression coefficients corresponding to each observation point, the influence direction and intensity differences of different driving factors on WEFSE under different irrigation areas and different periods are analyzed, forming the result distribution of the driving factor effect with spatial location and time changes. By analyzing the spatial distribution characteristics and temporal evolution characteristics of the local regression coefficients, the differences in the dominant role of each driving factor on WEFSE and its changing trend under different irrigation areas and different development stages are identified, revealing the spatiotemporal heterogeneity characteristics of the driving factors affecting WEFSE, and obtaining the spatiotemporal heterogeneity results of the driving factors' influence.

[0035] In a second aspect, the present invention proposes a system for measuring the efficiency of an irrigation district's water-energy-food system and for analyzing its spatiotemporal heterogeneity, applicable to the method for measuring the efficiency of the irrigation district's water-energy-food system and for analyzing its spatiotemporal heterogeneity, comprising:

[0036] The first module is used to construct an efficiency evaluation index system that includes input indicators, expected output indicators, and unexpected output indicators.

[0037] The second module is used to obtain input indicator data, expected output indicator data, and unexpected output indicator data for each irrigation district during each evaluation period, and to construct efficiency measurement input data.

[0038] The third module establishes an efficiency measurement model based on the efficiency evaluation index system, and obtains the WEFSE corresponding to each irrigation district in each evaluation period by combining the efficiency measurement input data.

[0039] The fourth module is used to construct the driving analysis dataset, and to perform regression analysis using a spatiotemporal geographic weighted regression model to obtain the spatiotemporal heterogeneity results of the driving factors.

[0040] In a third aspect, the present invention provides an electronic device comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, characterized in that, when the processor executes the computer program, it implements the steps of the above-mentioned method for measuring the efficiency of the irrigation district water-energy-food system and driving spatiotemporal heterogeneity.

[0041] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention constructs an efficiency evaluation index system for irrigation area water-energy-food system that includes water resource input, arable land resource input and energy input. Based on the constructed efficiency measurement model, it can reflect the asynchronous regulation characteristics of multiple factors and ecological environment constraints, and characterize spatiotemporal heterogeneity. It incorporates carbon dioxide emissions and grey water emissions as undesirable outputs into the efficiency measurement, while reflecting the economic output and ecological environment constraints of agricultural production, so that the analysis and evaluation results have a clearer sustainable significance. Attached Figure Description

[0042] Figure 1This is a flowchart of the spatiotemporal heterogeneity-driven analysis method for measuring the efficiency of the irrigation district water-energy-food system in this embodiment of the invention. Figure 2 This is a schematic diagram of the efficiency measurement model construction process in an embodiment of the present invention; Figure 3 This is a schematic diagram showing the distribution of WEFSE efficiency values ​​in various irrigation districts in a certain area according to an embodiment of the present invention; Figure 4 This is a block diagram of the module composition of the water-energy-food system efficiency measurement and spatiotemporal heterogeneity-driven analysis system in an embodiment of the present invention. Figure 5 This is a hardware structure diagram of the electronic device in an embodiment of the present invention. Detailed Implementation

[0043] To make the objectives, technical solutions, and beneficial effects of this invention clearer, the specific embodiments of this invention will be described in detail below with reference to the accompanying drawings. These embodiments are merely preferred examples of this invention, used to aid in understanding the inventive concept, and do not constitute a limitation on the scope of protection.

[0044] Water resources, energy, and food are key foundational resources for agricultural production and regional economic development in irrigation districts, and there is a complex coupling and mutual constraint among them: food production depends on stable irrigation water supply and land resource conditions; irrigation water intake, pumping, distribution, and return water treatment processes are highly dependent on energy support such as electricity or fuel; at the same time, fertilizers, pesticides, mechanized operations, and agricultural product processing in agricultural production also require energy input. With the increasing demand for food brought about by climate change and population growth, irrigation districts face multiple pressures such as tight water resources, rising energy consumption, and pollution emissions while ensuring food supply. Therefore, it is necessary to quantitatively evaluate the synergistic efficiency of the water-energy-food system in irrigation districts and support refined management.

[0045] Before providing a detailed description of the embodiments of this application, some of the nouns and terms involved in the embodiments of this application will be explained first. The nouns and terms that may be involved in the embodiments of this application are subject to the following interpretations.

[0046] WEF (Water-Energy-Food) refers to the irrigation district's water-energy-food system.

[0047] WEFSE (Water-Energy-Food Synergistic Efficiency) refers to the synergistic efficiency of the water-energy-food system in an irrigation district.

[0048] Super-Undesirable-SBM is an efficiency measure model that considers undesirable outputs and is used to further distinguish and rank decision-making units on the efficiency frontier.

[0049] GTWR (Geographically and Temporally Weighted Regression) is a spatiotemporally weighted regression model.

[0050] Existing research on efficiency evaluation methods for irrigation district water-energy-food systems largely relies on traditional radial DEA models to measure system efficiency. These models typically assume that inputs and outputs change proportionally, failing to reflect the asynchronous regulation of water resources, arable land resources, and energy inputs in actual management. Furthermore, ecological constraints such as carbon emissions and agricultural non-point source pollution accompanying agricultural production are often not fully incorporated into the efficiency evaluation system, making it difficult for efficiency measurements to reflect the system's sustainability. In addition, analyses of efficiency differences often employ static or global regression methods, which struggle to reveal the varying characteristics of driving factors' impact on efficiency across different irrigation districts and periods—that is, the spatiotemporal heterogeneity of driving mechanisms.

[0051] In view of this, embodiments of this application provide a method for measuring the efficiency of the water-energy-food system in irrigation districts and analyzing its spatiotemporal heterogeneity. This method constructs an efficiency evaluation index system including water resource input, arable land resource input, and energy input, and uses food yield and food output value as expected outputs, and carbon dioxide emissions and gray water emissions corresponding to agricultural non-point source pollution as undesirable outputs. Based on this, a Super-Undesirable-SBM model is established to measure the WEFSE (Weighted Energy-Effectiveness) of the irrigation district. Furthermore, GTWR (Gross-Gross-Resource Flow Analysis) is used to analyze the driving mechanism of WEFSE to reveal the spatiotemporal heterogeneity of the driving factors, thereby providing decision support for the coordinated management of water, energy, and food resources in irrigation districts and for energy conservation and emission reduction.

[0052] Figure 1 This is an optional flowchart of a spatiotemporal heterogeneity-driven analysis method for measuring the efficiency of an irrigation district water-energy-food system provided in this embodiment. Figure 1 The method flow may include, but is not limited to, steps S100 to S500.

[0053] S100, construct an efficiency evaluation index system that includes input indicators, expected output indicators and unexpected output indicators.

[0054] S200: Obtain input indicator data, expected output indicator data, and unexpected output indicator data for each irrigation district during each evaluation period to construct efficiency measurement input data; use existing technical means to preprocess the efficiency measurement input data by unifying units, handling missing values, and verifying consistency.

[0055] S300: Based on the efficiency evaluation index system, establish an input-oriented super-efficiency slack variable data envelopment analysis model that considers undesirable outputs to obtain an efficiency measurement model. In this model, input indicators and expected output indicators are set as strongly disposable variables, and undesirable output indicators are set as weakly disposable variables. Slack variables are used to characterize input redundancy, insufficient expected output, and undesirable output redundancy.

[0056] S400, based on the efficiency measurement model, inputs the efficiency measurement input data into the efficiency measurement model for solution, and obtains the WEFSE corresponding to each irrigation district in each evaluation period, that is, the coordinated efficiency of the irrigation district's water-energy-food system WEFSE.

[0057] Based on the WEFSE obtained, S500 constructs a driving analysis dataset and uses a spatiotemporal geographic weighted regression model to perform regression analysis on the driving factors of system collaborative efficiency, obtaining the spatiotemporal heterogeneity results of the driving factors' influence.

[0058] In some embodiments, step S100 may include, but is not limited to, steps S110 to S140:

[0059] S110: Construct input indicators, which include water resource input indicators, arable land resource input indicators, and energy input indicators; wherein, water resource input indicators include agricultural blue water input and agricultural green water input, and the arable land resource input indicators are expressed in terms of crop planting area; the energy input indicators include the converted values ​​of electricity, diesel, fertilizer, pesticide, and film energy consumption.

[0060] S120: Construct expected output indicators, including grain output and grain output value, to characterize the material output and economic benefits of agricultural production in the irrigation area;

[0061] S130: Construct undesirable output indicators, including carbon dioxide emissions generated during agricultural production and grey water emissions corresponding to agricultural non-point source pollution, to characterize the ecological and environmental constraints of the irrigation district's water-energy-food system.

[0062] S140: Based on input indicators, expected output indicators, and unexpected output indicators, a WEFSE evaluation index system for irrigation districts is formed, namely, an efficiency evaluation index system.

[0063] Reference Figure 2 In some embodiments, step S200 may include, but is not limited to, steps S210 to S240:

[0064] S210: Obtain agricultural green water input data and agricultural blue water input data; among which, agricultural green water input is determined based on effective precipitation and crop planting area, and agricultural blue water input is determined based on irrigation water statistics combined with irrigation quotas and crop area allocation, expressed by the following formula:

[0065]

[0066] in, This represents the agricultural water resource input in the i-th irrigation district. This represents the amount of agricultural green water resources input in the i-th irrigation district. This represents the amount of agricultural blue water resources input in the i-th irrigation district;

[0067]

[0068] in, Indicates the variety and quantity of food crops. Indicates the i-th irrigation district. The sown area for grain crops Indicates the i-th irrigation district. Effective precipitation during the growing season of grain crops;

[0069]

[0070]

[0071]

[0072] In the above formula, Rainfall amount, unit: mm. Represents the i-th irrigation district. The proportion of water used for irrigation of crops to the total water used for agricultural irrigation; This represents the irrigation water consumption for crops in the i-th irrigation district, in meters. 3 , Represents the i-th irrigation district. Irrigation quota for crops, unit: m 3 / hm 2 .

[0073] S220: Obtain data on arable land resource input, wherein arable land resource input is represented by the total crop sown area (arable land area).

[0074]

[0075] This indicates the amount of arable land resources invested.

[0076] S230: Acquire energy input data and perform conversion; among which, energy consumption data such as electricity, diesel, fertilizer, pesticides and agricultural film (mulch film) are uniformly converted into standard coal equivalent or uniform energy unit according to preset conversion factors.

[0077] The conversion factors corresponding to different energy types are shown in Table 1. To ensure the consistency of the accounting results for different energy types, the energy consumption data of different energy types are further converted into standard coal equivalent as energy input indicators. The conversion relationship between standard coal and energy units is: 1 tce = 29,307.6 MJ.

[0078] Table 1 Conversion Factors for Different Energy Types

[0079]

[0080] S240: Obtain expected and unexpected output data; expected output includes grain yield (grain output) and grain value; unexpected output includes carbon dioxide emissions and grey water emissions. Represented by the following formula:

[0081]

[0082] in, This represents the grain yield of the i-th irrigation district. Represents the i-th irrigation district. Yield of grain crops, in tons (t).

[0083]

[0084]

[0085]

[0086] in, This represents the grain output value of the i-th irrigation district, in yuan. This represents the j-th irrigation district. The yield of grain crops, This represents the price of the c-th type of grain crop in the i-th irrigation district, in yuan / ton.

[0087] This represents the carbon dioxide emissions of the i-th irrigation district, in kg. Indicates the quantity of different types of agricultural energy; This represents the agricultural energy consumption of the ith type in the ith irrigation district; This represents the sown area of ​​grain crops in the i-th irrigation district, in hectares (hm²). 2 ; This represents the total sown area of ​​all crops in the i-th irrigation district; This represents the carbon dioxide conversion factor for the e-th type of agricultural energy.

[0088] This represents the amount of grey water discharged from the i-th irrigation district, in m³. 3 ; This represents the total nitrogen input in grain production in the i-th irrigation district, in kg. This indicates the leaching rate of total nitrogen; This indicates the maximum permissible total nitrogen concentration in water, expressed in mg / L. This represents the background concentration of total nitrogen in natural water bodies. The carbon emission coefficients for agricultural energy are shown in Table 2.

[0089] Table 2 Carbon emission coefficients of different agricultural energy sources

[0090]

[0091] Step S300 may include, but is not limited to: setting inputs and expected outputs as strongly disposable variables, setting undesirable outputs as weakly disposable variables, establishing an efficiency measurement model (Super-Undesirable-SBM) based on an efficiency evaluation index system under the condition of constant returns to scale, and using slack variables to characterize input redundancy, expected output insufficiency, and undesirable output redundancy. The objective function expression is as follows:

[0092]

[0093] In the formula, This indicates the WEFSE (efficiency of the WEF system) for each irrigation district during each evaluation period. Indicates the first The i-th input of a decision-making unit This represents the projected value of the i-th input. Indicates the first The r-th expected output of a decision-making unit Indicates the first The r-th undesired output of a decision-making unit This represents the projected value of the r-th expected output. This represents the projected value of the r-th undesired output. These represent the quantities of expected and unexpected output, respectively. Indicates the number of input indicators. This represents the projection value of the input vector. Indicates the first The input vector of each decision-making unit Indicates the first The input vector of each decision-making unit Indicates the first The linear combination weights of each decision unit. Indicates the first The undesired output vector of each decision unit Indicates the first The expected output vector of each decision unit Indicates the first The undesired output vector of each decision unit This represents the projection value of the undesired output vector. This represents the projection value of the expected output vector. Indicates the first The expected output vector of each decision-making unit.

[0094] Slack variables are obtained by comparing the original observations of each input factor, expected output, and undesired output with their projections on the efficiency frontier. They are used to characterize the degree of input redundancy, insufficient expected output, and excessive undesired output under given technological conditions, where the above differences constitute an implicit characterization of slack information.

[0095] In some embodiments, the sorting process includes: sequentially selecting each decision unit as the object to be evaluated, and removing the currently evaluated decision unit from the reference set when performing efficiency measurement; constructing a production possibility set based on the remaining decision units, projecting the input and output of the evaluated decision units to obtain the corresponding superefficiency value; and sorting each decision unit according to the magnitude of its corresponding superefficiency value, wherein, under input-oriented conditions, the larger the superefficiency value, the higher the efficiency level of the corresponding decision unit, and the higher its ranking position.

[0096] Following the efficiency measurement steps outlined above, an efficiency measurement model is used to calculate the synergistic efficiency of the water-energy-food system in irrigation districts, specifically the WEFSE value for each irrigation district within each evaluation period. This efficiency measurement model uses slack variables to characterize each input factor and undesirable output, enabling it to identify the degree of redundancy and constraints in water resource input, arable land resource input, energy input, and environmental constraints in different irrigation districts. Furthermore, by using a super-efficiency setting to differentiate and rank irrigation districts at the efficiency frontier, the synergistic efficiency evaluation results possess a higher degree of refinement and interpretability.

[0097] In some embodiments, step S400 may include, but is not limited to: inputting efficiency measurement input data into the efficiency measurement model based on the efficiency measurement model, and outputting the WEFSE corresponding to each irrigation district in each evaluation period, so as to identify the improvement direction of the irrigation district in terms of water resources, arable land resources, energy input and environmental constraints.

[0098] Based on the WEFSE obtained, S500 constructs a driving analysis dataset and uses a spatiotemporal geographic weighted regression model to perform regression analysis, obtaining the spatiotemporal heterogeneity results of the driving factors.

[0099] In some embodiments, step S500 may include, but is not limited to, steps S510 to S540:

[0100] S510: The process of constructing the driving analysis dataset is as follows:

[0101] Using the WEFSE of each irrigation district within each evaluation period as the dependent variable, and the irrigation district number and year as the sample index, for each sample, the dependent variable, driving factor index, irrigation district spatial coordinates, and year are recorded to construct the driving analysis dataset. The expression is as follows:

[0102]

[0103] In the formula, This represents the dependent variable corresponding to observation point i in the irrigation district during the evaluation period, i.e., the WEFSE corresponding to each irrigation district in each evaluation period; This represents the spatial coordinates of observation point i in the irrigation area. Indicates the year of the irrigation district evaluation; The intercept term represents the observation points in the irrigation area; Indicators of driving factors Local regression coefficients at observation points in the irrigation area; Represents the random error term; where the driving factor index Used to characterize external driving factors affecting collaborative efficiency values; in some embodiments, driving factor indicators include, but are not limited to, indicators reflecting socio-economic conditions, hydro-meteorological environment and ecological environment characteristics.

[0104] S520, the local regression coefficients are estimated using the weighted least squares method, and the calculation formula is as follows:

[0105]

[0106] In the formula, This indicates that the k-th component of the estimated local regression coefficient at observation point i in the irrigation area is... , Represents the matrix of independent variables. Represents the matrix of independent variables The transpose of the matrix contains the driving factor data for different irrigation district observation points in each row of the independent variable matrix, and includes the constant term corresponding to the intercept term. It is an n×n diagonal matrix, whose diagonal elements represent the spatiotemporal weights of other observation points relative to observation point i; The dependent variable is the observation point i in the irrigation area. The vector formed.

[0107] S530, in order to distinguish the degree of influence of different observation points on the regression results in the analysis of local regression coefficients, measures the proximity of observation points in the spatial and temporal dimensions. Therefore, spatiotemporal distance is introduced to characterize the comprehensive difference between different observation points in spatial location and time. The formula for calculating spatiotemporal distance is as follows:

[0108]

[0109] In the formula, This indicates the degree of combined difference between observation point i and observation point j in both spatial and temporal dimensions. This represents the scale factor used to balance the effects of spatial and temporal distances under different units of measurement. and This represents the difference in spatial coordinates between the irrigation areas corresponding to observation point i and observation point j. This represents the difference between the corresponding observed samples in the time dimension.

[0110] A spatiotemporal weighted regression model was used for regression analysis. The model employed weighted least squares to estimate local regression coefficients, using the formula... Output the calculation results.

[0111] The sample weights are determined by the spatiotemporal distance between observation points, giving higher weights to samples that are closer in both space and time, in order to characterize the non-stationarity of the driving factors' influence in the spatial and temporal dimensions. The formula is as follows:

[0112]

[0113] In the formula, where Indicates a fixed kernel function. Indicates spatiotemporal bandwidth. This represents the sample weight of observation point j relative to regression point i;

[0114]

[0115] in, This represents the total number of observation samples in the irrigation area.

[0116] In spatiotemporal weighted regression analysis, the selection of spatiotemporal bandwidth directly affects the locality and parameter stability of the regression results. Too small a bandwidth can easily lead to overfitting of the model to local samples, while too large a bandwidth may weaken the spatiotemporal differences in the driving factors. Therefore, a balance needs to be struck between model fitting accuracy and parameter stability. Based on this, this invention uses a fixed kernel function for spatiotemporal weighting during the spatiotemporal weighted regression analysis and introduces an automatic spatiotemporal bandwidth determination mechanism based on the AICc minimization criterion. This mechanism can be automatically implemented by existing spatiotemporal weighted regression model analysis tools during the regression calculation process, enabling the model to adaptively determine the optimal spatiotemporal bandwidth during the regression calculation process, thereby obtaining stable and reliable local regression coefficients and achieving a stable characterization of the spatiotemporal heterogeneity of the driving mechanism.

[0117] In some embodiments, observation samples from different years of each irrigation district are used as input data for the efficiency measurement model. WEFSE is used as the dependent variable and driving factor indicators are used as independent variables. Combining the spatial coordinates and time information of the irrigation district, the optimal spatiotemporal bandwidth is automatically determined using the spatiotemporal geographic weighted regression model analysis module and AICc criterion of ArcGIS 10.8 software, and the local regression coefficients of the driving factors corresponding to each irrigation district under different time conditions are calculated accordingly.

[0118] S540, based on the local regression coefficients corresponding to each observation point, analyzes the differences in the direction and intensity of the influence of different driving factors on WEFSE under different irrigation areas and different time periods, forming the distribution of the driving factors' effects as a function of spatial location and time. By analyzing the spatial distribution and temporal evolution characteristics of the local regression coefficients, it identifies the differences and changing trends of the dominant role of each driving factor on WEFSE under different irrigation areas and different development stages, revealing the spatiotemporal heterogeneity of the driving factors' influence on WEFSE, and obtaining the spatiotemporal heterogeneity results of the driving factors' influence. Among them, the spatiotemporal heterogeneity results refer to the analytical results formed by the inconsistent direction and intensity of the influence of the same driving factor on WEFSE under different irrigation areas and different time conditions, specifically manifested as the differentiated distribution of local regression coefficients in the spatial and temporal dimensions.

[0119] In some embodiments, the nighttime light index is used as one of the driving factors. A spatiotemporally weighted regression model is used to perform regression analysis on samples from different irrigation districts and at different times, obtaining the corresponding local regression coefficients at each observation point in each irrigation district. When the local regression coefficient corresponding to the driving factor in a specific irrigation district is positive within a specific period, it indicates that the factor has a promoting effect on WEFSE under those spatial and temporal conditions; conversely, when the regression coefficient is negative, it indicates that it has an inhibitory effect on WEFSE. Simultaneously, the absolute value of the regression coefficient reflects the difference in the intensity of the driving factor's influence; the larger the absolute value, the more significant its influence on WEFSE. By comparing and analyzing the spatial distribution characteristics of the local regression coefficients of each irrigation district and their evolutionary trends over time, the spatiotemporal heterogeneity of the direction and intensity of the same driving factor's influence on WEFSE can be identified between different irrigation districts and within the same irrigation district under different time conditions.

[0120] Therefore, this application can generate quantitative analysis results characterizing the spatiotemporal differences of the effects of driving factors. These results can reflect the changing characteristics of the positive or negative impact of the same driving factor on WEFSE in different irrigation areas or at different times, as well as the spatial distribution pattern and temporal evolution trend of the impact intensity, thereby revealing the spatiotemporal heterogeneity of the effects of driving factors.

[0121] like Figure 3 As shown, Figure 3 To obtain the WEFSE (corresponding to the water-energy-food system efficiency measurement method for irrigation districts proposed in this invention) after calculating the collaborative efficiency of multiple irrigation districts in a certain area within the same evaluation period. Figure 3 The spatial distribution results of the efficiency values ​​are shown in the figure. This is used to intuitively reflect the spatial differences in the collaborative efficiency levels of different irrigation districts during the same evaluation period, and to provide basic efficiency data support for the subsequent analysis of the spatiotemporal heterogeneity of driving factors.

[0122] In some embodiments, based on the input, expected output and unexpected output data of each irrigation district during the same evaluation period, the efficiency of each irrigation district is measured using the formula described in step S300 according to steps S300-S400, and the corresponding WEFSE is obtained.

[0123] In some embodiments, taking a specific irrigation district within a certain evaluation period as an example, the water resource input, arable land resource input, energy input, and corresponding expected and unexpected output data of the irrigation district during the evaluation period are substituted into the efficiency measurement model described in step S300. The WEFSE of the irrigation district during the corresponding evaluation period is obtained by solving the model. The above calculation process is repeated for each irrigation district to obtain the WEFSE of each irrigation district. The WEFSE is then associated with the spatial location information of the corresponding irrigation district, and the WEFSE of each irrigation district is spatially represented using ArcGIS 10.8 software, thereby forming a model as shown below. Figure 3The WEFSE spatial distribution results for each irrigation district are shown in the figure.

[0124] Please see Figure 4 This application also provides a system for measuring the efficiency of an irrigation district's water-energy-food system and for analyzing its spatiotemporal heterogeneity, used to implement the aforementioned method for measuring the efficiency of an irrigation district's water-energy-food system and for analyzing its spatiotemporal heterogeneity. The system includes:

[0125] The first module 101 is used to construct an efficiency evaluation index system that includes input indicators, expected output indicators and unexpected output indicators.

[0126] The second module 102 is used to acquire input indicator data, expected output indicator data and unexpected output indicator data of each irrigation district in each evaluation period, and to construct efficiency measurement input data.

[0127] The third module 103 establishes an efficiency measurement model based on the efficiency evaluation index system, and obtains the corresponding WEFSE of each irrigation district in each evaluation period based on the efficiency measurement input data.

[0128] Module 4, 104, is used to construct the driving analysis dataset. It employs a spatiotemporal geographic weighted regression model to perform regression analysis and obtain the spatiotemporal heterogeneity results of the driving factors.

[0129] It is understood that the content of the above-mentioned methods for measuring the efficiency of the irrigation area water-energy-food system and driving analysis of spatiotemporal heterogeneity is applicable to this system embodiment. The specific functions implemented in this system embodiment are the same as those in the above-mentioned method embodiments, and the beneficial effects achieved are also the same as those achieved in the above-mentioned method embodiments. Therefore, it will not be repeated here.

[0130] Please see Figure 5 , Figure 5 The hardware structure of an electronic device according to another embodiment is illustrated, the electronic device comprising:

[0131] The processor 201 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application.

[0132] The memory 202 can be implemented in the form of read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 202 is used to store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 202 and executed by the processor 201 to realize the above-mentioned method for measuring the efficiency of the irrigation district water-energy-food system and driving spatiotemporal heterogeneity.

[0133] Input / output interface 203 is used to implement information input and output;

[0134] The communication interface 204 is used to enable communication between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WiFi, Bluetooth, etc.).

[0135] Bus 205 is used to transmit information between various components of the device. The processor 201, memory 202, input / output interface 203 and communication interface 204 are connected to each other within the device via bus 205.

[0136] This application embodiment also provides a computer-readable storage medium storing a computer program, which, when executed by processor 201, implements the above-described method for measuring the efficiency of the irrigation district water-energy-food system and driving spatiotemporal heterogeneity.

[0137] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0138] Memory 202, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory 202 may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory 202 may also include memory remotely disposed relative to the processor, the remote memory being connected to the processor via a network.

[0139] Although the steps in the above embodiments are described in the above order, those skilled in the art will understand that in order to achieve the effect of this embodiment, different steps do not need to be executed in such an order. They can be executed simultaneously (in parallel) or in a reverse order. These simple variations are all within the protection scope of this invention.

[0140] Those skilled in the art will understand that although some embodiments described herein include certain features but not others included in other embodiments, combinations of features from different embodiments are intended to be within the scope of the invention and form different embodiments. For example, in the claims of this invention, any of the claimed embodiments can be used in any combination.

[0141] It should be noted that the above embodiments are illustrative of the invention and not restrictive, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, the word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed PC.

[0142] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it should be noted that the parts not covered in this invention are the same as or can be implemented using existing technology. It will be readily understood by those skilled in the art that the scope of protection of this invention is obviously not limited to these specific embodiments. Without departing from the principles of this invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions resulting from these changes or substitutions will all fall within the scope of protection of this invention.

Claims

1. A method for measuring the efficiency of an irrigation district's water-energy-food system and for analyzing its spatiotemporal heterogeneity, characterized in that... Construct an efficiency evaluation index system that includes input indicators, expected output indicators, and unexpected output indicators; Data on input indicators, expected output indicators, and unexpected output indicators for each irrigation district during each evaluation period were obtained to construct input data for efficiency measurement. An efficiency measurement model is established based on the efficiency evaluation index system, and WEFSE corresponding to each irrigation district in each evaluation period is obtained based on the efficiency measurement input data. Based on the obtained WEFSE, a driving analysis dataset is constructed, and a spatiotemporal geographic weighted regression model is used for regression analysis to obtain the spatiotemporal heterogeneity results of the driving factors.

2. The method for measuring the efficiency of the irrigation district water-energy-food system and driving spatiotemporal heterogeneity according to claim 1, characterized in that, The input indicators include water resource input indicators, arable land resource input indicators and energy input indicators; The expected output indicators include grain output and grain output value, which are used to characterize the material output and economic benefits of agricultural production in the irrigation area. The undesirable output indicators include carbon dioxide emissions generated during agricultural production and grey water emissions corresponding to agricultural non-point source pollution, which are used to characterize the ecological and environmental constraints of the irrigation district's water-energy-food system.

3. The method for measuring the efficiency of the irrigation district water-energy-food system and driving spatiotemporal heterogeneity according to claim 2, characterized in that, The water resource input indicators include agricultural blue water input and agricultural green water input; the arable land resource input indicators are expressed as crop planting area; the energy input indicators include the converted values ​​of electricity, diesel, fertilizer, pesticide and film energy consumption.

4. The method for measuring the efficiency of the irrigation district water-energy-food system and driving spatiotemporal heterogeneity according to claim 1, characterized in that, The efficiency measurement model is established, using slack variables to characterize input redundancy, expected output underperformance, and undesired output redundancy. The objective function is as follows: ； In the formula, This indicates the WEFSE corresponding to each irrigation district in each evaluation period. Indicates the first The i-th input of a decision-making unit This represents the projected value of the i-th input. Indicates the first The r-th expected output of a decision-making unit Indicates the first The r-th undesired output of a decision-making unit This represents the projected value of the r-th expected output. This represents the projected value of the r-th undesired output. These represent the quantities of expected and unexpected output, respectively. Indicates the number of input indicators. This represents the projection value of the input vector. Indicates the first The input vector of each decision-making unit Indicates the first The input vector of each decision-making unit Indicates the first The linear combination weights of each decision unit. Indicates the first The undesired output vector of each decision unit Indicates the first The expected output vector of each decision unit Indicates the first The undesired output vector of each decision unit This represents the projection value of the undesired output vector. This represents the projection value of the expected output vector. Indicates the first The expected output vector of each decision-making unit.

5. The method for measuring the efficiency of the irrigation district water-energy-food system and driving spatiotemporal heterogeneity according to claim 1, characterized in that, The process of constructing the driving analysis dataset is as follows: Using the WEFSE of each irrigation district within each evaluation period as the dependent variable, and the irrigation district number and year as the sample index, for each sample, the dependent variable, driving factor index, irrigation district spatial coordinates, and year are recorded to construct the driving analysis dataset. The expression is as follows: ； In the formula, This represents the dependent variable for observation point i in the irrigation district during the evaluation period. This represents the spatial coordinates of observation point i in the irrigation area. Indicates the year of the irrigation district evaluation; The intercept term represents the observation points in the irrigation area; Indicators of driving factors Local regression coefficients at observation points in the irrigation area; This represents the random error term.

6. The method for measuring the efficiency of the irrigation district water-energy-food system and driving spatiotemporal heterogeneity according to claim 5, characterized in that, Spatiotemporal distance is introduced to characterize the comprehensive differences between different observation points in terms of spatial location and time. The formula for calculating spatiotemporal distance is as follows: ； In the formula, This indicates the degree of combined difference between observation point i and observation point j in both spatial and temporal dimensions. This represents the scale factor used to balance the effects of spatial and temporal distances under different units of measurement. and This represents the difference in spatial coordinates between the irrigation areas corresponding to observation point i and observation point j. This represents the difference between the corresponding observed samples in the time dimension.

7. The method for measuring the efficiency of the irrigation district water-energy-food system and driving spatiotemporal heterogeneity according to claim 5, characterized in that, The regression analysis using a spatiotemporal geographic weighted regression model is as follows: The spatiotemporal weighted regression model uses weighted least squares to estimate local regression coefficients, and the output calculation formula is as follows: ； In the formula, This indicates that the k-th component of the estimated local regression coefficient at observation point i in the irrigation area is... , Represents the matrix of independent variables. Represents the matrix of independent variables transpose; It is an n×n diagonal matrix, whose diagonal elements represent the spatiotemporal weights of other observation points relative to observation point i; The dependent variable is the observation point i in the irrigation area. The vector formed; The sample weights are determined by the spatiotemporal distance between observation points, giving higher weights to samples that are closer in both space and time, in order to characterize the non-stationarity of the driving factors' influence in the spatial and temporal dimensions. The formula is as follows: ； In the formula, where Indicates a fixed kernel function. Indicates spatiotemporal bandwidth. This represents the sample weight of observation point j relative to regression point i; ； in, This represents the total number of observation samples in the irrigation area.

8. The method for measuring the efficiency of the irrigation district water-energy-food system and driving spatiotemporal heterogeneity according to claim 7, characterized in that, Based on the local regression coefficients corresponding to each observation point, the influence direction and intensity differences of different driving factors on WEFSE under different irrigation areas and different periods are analyzed, forming the result distribution of the driving factor effect with spatial location and time changes. By analyzing the spatial distribution characteristics and temporal evolution characteristics of the local regression coefficients, the differences and changing trends of the dominant role of each driving factor on WEFSE under different irrigation areas and different development stages are identified, revealing the spatiotemporal heterogeneity characteristics of the driving factors affecting WEFSE, and obtaining the spatiotemporal heterogeneity results of the driving factors' influence.

9. A system for measuring the efficiency of an irrigation district's water-energy-food system and for analyzing its spatiotemporal heterogeneity, applied to the methods for measuring the efficiency of an irrigation district's water-energy-food system and analyzing its spatiotemporal heterogeneity as described in claims 1-8, characterized in that... include: The first module is used to construct an efficiency evaluation index system that includes input indicators, expected output indicators, and unexpected output indicators. The second module is used to obtain input indicator data, expected output indicator data, and unexpected output indicator data for each irrigation district during each evaluation period, and to construct efficiency measurement input data. The third module establishes an efficiency measurement model based on the efficiency evaluation index system, and obtains the corresponding WEFSE of each irrigation district in each evaluation period based on the efficiency measurement input data. The fourth module is used to construct the driving analysis dataset, and to perform regression analysis using a spatiotemporal geographic weighted regression model to obtain the spatiotemporal heterogeneity results of the driving factors.

10. An electronic device comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1-8.

Images