Water resource real-time monitoring and scheduling management system based on spatio-temporal data

Abstract

The invention provides a water resource real-time monitoring and scheduling management system based on spatio-temporal data, and relates to the technical field of water resource management, and the system comprises a data obtaining module which obtains the distribution data of surface water and underground water through the infrared and radar remote sensing technology; the water supply and demand obtaining module predicts supply and demand information based on historical water levels; the image model construction module is used for constructing three-dimensional image models of the shallow supply area and the deep storage area and mapping the three-dimensional image models to a space-time coordinate system; the monitoring optimization module is used for constructing an optimization objective function in combination with the real-time water level information and the supply and demand data to solve an optimal result; and the scheduling management module is used for allocating water resources according to an optimization result and realizing space and time balance optimization. The water resource utilization efficiency is improved, uneven distribution is relieved, and comprehensive optimization of economic, social and ecological benefits is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of water resource management, and in particular to a water resource real-time monitoring and dispatching management system based on spatiotemporal data. Background Art

[0002] Water resources are an important foundation for human survival and social and economic development. However, with the intensification of global climate change and human activities, the problems of water shortage and uneven distribution are becoming increasingly prominent. Traditional water resource management methods usually rely on historical data and empirical decisions, lack the ability to respond to real-time dynamic changes, and are difficult to meet the needs of modern water resource management.

[0003] In recent years, with the development of the Internet of Things, remote sensing technology, geographic information system (GIS) and big data analysis technology, water resources monitoring and management based on spatiotemporal data has become possible. By collecting multi-source data such as hydrology, meteorology, and water demand in real time and combining it with spatiotemporal analysis models, dynamic monitoring and scientific scheduling of water resources can be achieved. However, the current technical solutions still have shortcomings in data integration, real-time performance and scheduling optimization, and an efficient and intelligent real-time monitoring and scheduling management system for water resources is urgently needed. Summary of the invention

[0004] In order to overcome the shortcomings of the prior art, the purpose of the present invention is to provide a real-time monitoring and scheduling management system for water resources based on spatiotemporal data, which can significantly improve the efficiency of water resource utilization, alleviate the problem of uneven spatiotemporal distribution of water resources, and achieve comprehensive optimization of regional economic, social and ecological benefits.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A real-time monitoring and dispatching management system for water resources based on spatiotemporal data, comprising:

[0007] A data acquisition module is used to acquire the spatiotemporal distribution data of surface water and groundwater in the target area through infrared remote sensing technology and radar remote sensing technology; the spatiotemporal distribution data includes water resource distribution images of shallow recharge areas and deep storage areas;

[0008] A water supply and demand acquisition module is used to construct a water level prediction model based on a historical water level database, and predict the water supply and demand information in the target area through the water level prediction model;

[0009] An image model building module is used to establish a spatiotemporal coordinate system of the target area and to build a three-dimensional image model of the shallow recharge area and the deep storage area based on the water resource distribution image;

[0010] A mapping module, for mapping the three-dimensional image models of the shallow recharge area and the deep storage area onto the space-time coordinate system to obtain the water recharge potential of the shallow recharge area and the water storage capacity of the deep storage area;

[0011] A monitoring optimization module is used to obtain real-time monitoring water level information, and to construct an optimization objective function based on the real-time monitoring water level information and the water supply and demand information, and to solve the problem with the water replenishment potential and the water storage capacity as constraints to obtain an optimization result;

[0012] The scheduling management module is used to schedule the water resources in the target area according to the optimization result to achieve management based on optimal spatial balance optimization and temporal balance optimization.

[0013] Preferably, the water level prediction model is a network structure based on PSO-SVR-LSTM.

[0014] Preferably, the data acquisition module includes:

[0015] An image acquisition submodule is used to establish a UAV circular flight route in the target area, install an infrared remote sensing device and a radar remote sensing device on the UAV, set a number of UAVs to fly according to the UAV circular flight route, scan the target area through the infrared remote sensing device, and obtain infrared image data covered by the corresponding UAV circular flight route;

[0016] An image denoising submodule, used for denoising the infrared image data;

[0017] An image stitching submodule, used for overlapping and stitching the infrared image data according to the geographical location corresponding to the target area to obtain stitching image data;

[0018] A region division submodule is used to set a temperature interval threshold, mark the area in the stitched image data whose temperature is not within the temperature interval threshold as a surface water distribution area, and mark the area in the stitched image data whose temperature is within the temperature interval threshold as a groundwater distribution area;

[0019] The data determination submodule is used to determine the corresponding spatiotemporal distribution data according to the surface water area and the groundwater distribution area.

[0020] Preferably, the image denoising submodule comprises:

[0021] A noise monitoring unit, used for detecting noise points on the infrared image data using a filter window to obtain a noise mean;

[0022] A denoising unit, which denoises the infrared image data in the corresponding filtering window when the noise mean in the filtering window is greater than a preset threshold;

[0023] The window sliding unit is used to slide the filter window until the entire infrared image data is traversed.

[0024] Preferably, the noise monitoring unit comprises:

[0025] The detection model construction subunit is used to construct a noise point detection model according to the mean and median of each image point in the filtering window; the noise point detection model is: Among them, f(x) represents the similar noise value of pixel x, u(x) represents the gray value of pixel x, and u mean (x) represents the grayscale mean of all pixels in the filter window centered on pixel x, ▽u(x) represents the gradient mean of pixel x, ▽u mean (x) is the grayscale median of all pixels in the filter window centered on pixel x, ▽ x (x) represents the gradient value of pixel x in the horizontal direction, ▽ y (x) represents the gradient value of pixel x in the vertical direction;

[0026] A detection subunit, used to detect each image point in the filtering window using the noise point detection model to obtain a similar noise value for each image point;

[0027] A noise determination subunit, used for treating corresponding image points with a value greater than a similar noise value as noise points;

[0028] The mean value determination subunit obtains the noise mean value according to the number of noise points and the number of each image point in the filter window.

[0029] Preferably, the denoising unit comprises:

[0030] The variance calculation subunit is used to calculate the pseudo pixel variance according to the grayscale median of all pixels in the filter window; wherein the calculation formula of the pseudo pixel variance is: in, represents the pseudo pixel variance of the pixel point (a, b) in the area of ​​the filter window with a size of (2n+1)×(2n+1), mean(a,b) represents the grayscale median of the pixel point (a,b) in the filter window, and x(k,l) represents the grayscale value of the pixel point at the position (k,l);

[0031] The denoising model construction subunit is used to construct a window denoising model using the pseudo pixel variance; the calculation formula of the window denoising model is: Among them, f(a, b) represents the gray value of the pixel (a, b) after denoising, D is the adjustable coefficient, and x(a, b) represents the gray value of the pixel (a, b) in the filter window.

[0032] Preferably, the formula of the optimization objective function is:

[0033]

[0034] Among them, Z is the optimization target value, which represents the comprehensive optimization target of supply and demand balance deviation, time balance deviation, and economic and ecological costs; S i is the water supply of the ith region, including surface water and groundwater; D i is the water demand of the ith region, including agriculture, industry, life, and ecology, ΔW t is the water level fluctuation within time t, that is, the absolute value of the water level change; C eco is the regional economic cost; C env is the regional ecological cost; 1 ,ω 2 ,ω 3 ,ω 4 are weight coefficients used to balance the importance of supply-demand deviation, time balance, economic cost and ecological cost; N is the total number of spatial nodes in the target area; T is the total number of steps in the optimization time period;

[0035] The constraints include: water resource supply and demand balance constraint, water resource total amount constraint and water replenishment potential constraint; the expression of water resource supply and demand balance constraint is: S i ≥D i , The expression of the total water resources constraint is: Among them, W total is the total available water resources in the target area, including surface water and groundwater; the expression of the water replenishment potential constraint is: in, is the water supply to the shallow recharge area of ​​the ith region; The water supply to the deep storage area of ​​the ith region; is the recharge potential of the shallow recharge area in the ith region; is the water storage capacity of the deep storage area in the ith region.

[0036] Preferably, the optimization algorithm for solving the optimization objective function is a particle swarm algorithm.

[0037] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0038] The present invention provides a real-time monitoring and dispatching management system for water resources based on spatiotemporal data, comprising: a data acquisition module, used to acquire the spatiotemporal distribution data of surface water and groundwater in a target area through infrared remote sensing technology and radar remote sensing technology; the spatiotemporal distribution data includes water resource distribution images of shallow recharge areas and deep storage areas; a water supply and demand acquisition module, used to construct a water level prediction model based on a historical water level database, and predict water supply and demand information in a target area through the water level prediction model; an image model construction module, used to establish a spatiotemporal coordinate system of a target area, and construct a water resource distribution image of a shallow recharge area and a deep storage area based on the water resource distribution image. A three-dimensional image model; a mapping module for mapping the three-dimensional image models of the shallow recharge area and the deep storage area onto the spatiotemporal coordinate system to obtain the water replenishment potential of the shallow recharge area and the water storage capacity of the deep storage area; a monitoring and optimization module for obtaining real-time monitoring water level information, and constructing an optimization objective function based on the real-time monitoring water level information and the water supply and demand information, solving the function with the water replenishment potential and the water storage capacity as constraints to obtain an optimization result; a scheduling and management module for scheduling the water resources in the target area according to the optimization result to achieve management based on optimal spatial equilibrium optimization and temporal equilibrium optimization. The present invention provides an efficient, scientific, and intelligent real-time monitoring and scheduling management system for water resources by combining remote sensing technology, spatiotemporal data analysis, and optimization models, which can significantly improve the efficiency of water resource utilization, alleviate the problem of uneven spatiotemporal distribution of water resources, and achieve comprehensive optimization of regional economic, social, and ecological benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0040] Figure 1 A schematic diagram of the system structure provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] The purpose of the present invention is to provide a real-time monitoring and scheduling management system for water resources based on spatiotemporal data. By combining remote sensing technology, spatiotemporal data analysis and optimization models, an efficient, scientific and intelligent real-time monitoring and scheduling management system for water resources is provided, which can significantly improve the efficiency of water resource utilization, alleviate the problem of uneven spatiotemporal distribution of water resources, and achieve comprehensive optimization of regional economic, social and ecological benefits.

[0043] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0044] Figure 1 A schematic diagram of the system structure provided by an embodiment of the present invention, such as Figure 1 As shown, the present invention provides a water resources real-time monitoring and scheduling management system based on spatiotemporal data, including:

[0045] A data acquisition module is used to acquire the spatiotemporal distribution data of surface water and groundwater in the target area through infrared remote sensing technology and radar remote sensing technology; the spatiotemporal distribution data includes water resource distribution images of shallow recharge areas and deep storage areas;

[0046] A water supply and demand acquisition module is used to construct a water level prediction model based on a historical water level database, and predict the water supply and demand information in the target area through the water level prediction model;

[0047] An image model building module is used to establish a spatiotemporal coordinate system of the target area and to build a three-dimensional image model of the shallow recharge area and the deep storage area based on the water resource distribution image;

[0048] A mapping module, for mapping the three-dimensional image models of the shallow recharge area and the deep storage area onto the space-time coordinate system to obtain the water recharge potential of the shallow recharge area and the water storage capacity of the deep storage area;

[0049] A monitoring optimization module is used to obtain real-time monitoring water level information, and to construct an optimization objective function based on the real-time monitoring water level information and the water supply and demand information, and to solve the problem with the water replenishment potential and the water storage capacity as constraints to obtain an optimization result;

[0050] The scheduling management module is used to schedule the water resources in the target area according to the optimization result to achieve management based on optimal spatial balance optimization and temporal balance optimization.

[0051] Preferably, the data acquisition module includes:

[0052] An image acquisition submodule is used to establish a UAV circular flight route in the target area, install an infrared remote sensing device and a radar remote sensing device on the UAV, set a number of UAVs to fly according to the UAV circular flight route, scan the target area through the infrared remote sensing device, and obtain infrared image data covered by the corresponding UAV circular flight route;

[0053] An image denoising submodule, used for denoising the infrared image data;

[0054] An image stitching submodule, used for overlapping and stitching the infrared image data according to the geographical location corresponding to the target area to obtain stitching image data;

[0055] The area division submodule is used to set a temperature interval threshold, mark the area in the stitched image data whose temperature is not within the temperature interval threshold as the surface water distribution area, and mark the area in the stitched image data whose temperature is within the temperature interval threshold as the groundwater distribution area.

[0056] The data determination submodule is used to determine the corresponding spatiotemporal distribution data according to the surface water area and the groundwater distribution area.

[0057] Specifically, the image acquisition submodule of this embodiment uses an infrared remote sensing device and a radar remote sensing device mounted on a drone to perform a comprehensive scan of the target area. First, a circular flight route of the drone is designed in the target area to ensure that the flight route can cover the entire geographical range of the target area. Several drones are set up to fly synchronously according to a predetermined flight route, and the surface is scanned using an infrared remote sensing device to obtain infrared image data of the target area. At the same time, radar laser signals are sent through a radar remote sensing device, and reflected signals are received to obtain groundwater distribution information. The flight altitude and scanning frequency of the drone are adjusted according to the terrain complexity and resolution requirements of the target area to ensure the comprehensiveness and accuracy of data collection.

[0058] The image denoising submodule preprocesses the acquired infrared image data to improve the quality and accuracy of the data. First, an image denoising algorithm based on median filtering and Gaussian filtering is used to remove random noise and environmental interference signals in the infrared image. Secondly, combined with the reflected signal data of the radar remote sensing device, the abnormal points in the infrared image are corrected to eliminate artifacts caused by equipment errors or external interference. Finally, by comparing the overlapping area data collected by multiple drones, the multi-source data fusion technology is used to further optimize the image quality to ensure that the denoised image data can truly reflect the surface and groundwater distribution in the target area.

[0059] The image stitching submodule overlaps and stitches the denoised infrared image data according to the geographical location of the target area to generate complete stitching image data. First, the geographical location of each infrared image is calibrated according to the geographical coordinate information of the UAV flight route. Then, the overlapping areas between adjacent images are identified using feature point matching algorithms (such as SIFT or SURF algorithms), and the overlapping areas are aligned using image registration technology. Finally, the weighted average fusion algorithm is used to smooth the pixel values ​​of the overlapping areas, eliminate the abruptness of the stitching boundaries, and generate seamless stitching image data to ensure that the surface and groundwater distribution information of the target area can be fully presented.

[0060] The regional division submodule classifies and annotates the surface water and groundwater distribution areas in the spliced ​​image data by setting temperature interval thresholds. First, according to the principle of infrared remote sensing technology, the temperature of the surface water area is usually lower than the surrounding surface temperature, while the temperature of the groundwater area fluctuates within a certain range. According to the climatic conditions and geological characteristics of the target area, an appropriate temperature interval threshold is set. Then, the spliced ​​image data is classified using an image segmentation algorithm (such as threshold segmentation or K-means clustering), and the areas with temperatures not within the threshold range are marked as surface water distribution areas, and the areas with temperatures within the threshold range are marked as groundwater distribution areas. Finally, the annotated image data is generated to provide basic data support for the subsequent division of shallow recharge areas and deep storage areas.

[0061] Preferably, the image denoising submodule comprises:

[0062] A noise monitoring unit, used for detecting noise points on the infrared image data using a filter window to obtain a noise mean;

[0063] A denoising unit, which denoises the infrared image data in the corresponding filtering window when the noise mean in the filtering window is greater than a preset threshold;

[0064] The window sliding unit is used to slide the filter window until the entire infrared image data is traversed.

[0065] Preferably, the noise monitoring unit comprises:

[0066] The detection model construction subunit is used to construct a noise point detection model according to the mean and median of each image point in the filtering window; the noise point detection model is: Among them, f(x) represents the similar noise value of pixel x, u(x) represents the gray value of pixel x, and u mean (x) represents the grayscale mean of all pixels in the filter window centered on pixel x, ▽u(x) represents the gradient mean of pixel x, ▽u mean(x) is the grayscale median of all pixels in the filter window centered on pixel x, ▽ x (x) represents the gradient value of pixel x in the horizontal direction, ▽ y (x) represents the gradient value of pixel x in the vertical direction;

[0067] A detection subunit, used to detect each image point in the filtering window using the noise point detection model to obtain a similar noise value for each image point;

[0068] A noise determination subunit, used for treating corresponding image points with a value greater than a similar noise value as noise points;

[0069] The mean value determination subunit obtains the noise mean value according to the number of noise points and the number of each image point in the filter window.

[0070] Preferably, the denoising unit comprises:

[0071] The variance calculation subunit is used to calculate the pseudo pixel variance according to the grayscale median of all pixels in the filter window; wherein the calculation formula of the pseudo pixel variance is: in, represents the pseudo pixel variance of the pixel point (a, b) in the area of ​​the filter window with a size of (2n+1)×(2n+1), mean(a,b) represents the grayscale median of the pixel point (a,b) in the filter window, and x(k,l) represents the grayscale value of the pixel point at the position (k,l);

[0072] The denoising model construction subunit is used to construct a window denoising model using the pseudo pixel variance; the calculation formula of the window denoising model is: Among them, f(a, b) represents the gray value of the pixel (a, b) after denoising, D is an adjustable coefficient, and x(a, b) represents the gray value of the pixel (a, b) in the filter window.

[0073] Specifically, the noise monitoring unit detects noise points in the infrared image data through a filtering window. First, a noise point detection model based on the mean and median is constructed, which can comprehensively consider the grayscale value, gradient information and overall grayscale distribution characteristics of the pixel point in the filtering window. By calculating the similar noise value of each pixel point in the filtering window, it is determined whether the pixel point is a noise point. The detection model uses the deviation of the grayscale mean and median of the pixel point, as well as the characteristics of the gradient change, to effectively identify random noise and artifacts in the image. Finally, the detected noise points are marked to provide a basis for subsequent denoising processing.

[0074] Optionally, the detection subunit uses a noise point detection model to detect each pixel point in the filter window one by one. By comparing the similar noise value of each pixel point with the grayscale distribution of the surrounding pixels, it is determined whether the pixel point is a noise point. During the detection process, the focus is on analyzing the degree of deviation between the grayscale value of the pixel point and the mean and median of the surrounding pixels, and combining the gradient information to ensure the accuracy of the detection result. The detection subunit can quickly locate the noise points in the image and mark them as abnormal points, laying the foundation for the subsequent noise mean calculation and denoising processing.

[0075] Furthermore, the noise determination subunit marks the pixels with similar noise values ​​greater than a certain threshold as noise points based on the results of the detection subunit. The number and distribution of noise points are calculated by counting all the pixels in the filter window. This subunit can dynamically adjust the noise threshold to adapt to the noise characteristics of different images, thereby improving the flexibility and adaptability of noise detection. Finally, the noise determination subunit outputs the distribution information of the noise points to provide data support for the subsequent noise mean calculation.

[0076] Furthermore, the denoising unit performs denoising on the detected noise points based on the noise monitoring unit. First, the variance calculation subunit calculates the pseudo-pixel variance based on the grayscale median in the filter window to measure the grayscale difference between the noise point and the surrounding pixels. Then, the window denoising model is constructed using the pseudo-pixel variance, and the grayscale value of the noise point is corrected by adjusting the adjustable coefficient D in the model. The denoising unit can effectively smooth the grayscale value of the noise point while retaining the edge and detail information of the image, avoiding image blur caused by over-smoothing.

[0077] Furthermore, the window sliding unit traverses the entire infrared image data by sliding the filter window to ensure that all pixels can be detected and processed. During the sliding process, the window sliding unit dynamically adjusts the window size to adapt to the noise characteristics of different areas. For example, for areas with more noise, the window sliding unit will select a smaller window to improve detection accuracy; while for areas with less noise, a larger window will be selected to improve processing efficiency. The design of the window sliding unit ensures the comprehensiveness and efficiency of the denoising process.

[0078] Through the collaborative work of the above modules, the image denoising submodule can significantly improve the quality of infrared image data. The noise point detection model combines grayscale value and gradient information to accurately identify random noise and artifacts; the denoising model uses pseudo-pixel variance to correct noise points, which can effectively remove noise while retaining the edge and detail information of the image. The overall denoising process is efficient and flexible, and is suitable for infrared image data with different noise characteristics. Ultimately, the denoised image data is clearer and more accurate, providing a reliable data basis for subsequent image stitching and region division.

[0079] Preferably, the water level prediction model is a network structure based on PSO-SVR-LSTM.

[0080] Specifically, the PSO-SVR-LSTM water level prediction model combines the advantages of particle swarm optimization (PSO), support vector regression (SVR) and long short-term memory network (LSTM) to improve the accuracy and adaptability of water level prediction. First, the data in the historical water level database is used as input, and after preprocessing (such as denoising, normalization and missing value filling), it is input into the SVR and LSTM models respectively. The SVR model is good at processing small sample data and nonlinear relationships, and can quickly capture the short-term trend of water level changes; the LSTM model can effectively learn the long-term dependency of water level changes through its memory unit and gating mechanism. The PSO algorithm is used in this framework to optimize the key parameters of the SVR and LSTM models (such as the penalty coefficient and kernel parameter of SVR, the number of neurons and learning rate of LSTM) to ensure that the prediction performance of the model is optimal.

[0081] In the PSO algorithm, the parameters of the SVR and LSTM models are used as the search space of the particle swarm, and each particle represents a set of parameter combinations. PSO initializes the position and velocity of the particle swarm and uses an objective function (such as mean square error MSE) to evaluate the fitness value of each set of parameters. In each iteration, the particles update their positions and velocities according to the individual optimal position and the global optimal position, gradually approaching the optimal solution. Specifically, PSO optimizes the penalty coefficient (C) and kernel parameter (g) of the SVR model, as well as the number of neurons (m) and learning rate (lr) of the LSTM model. Through multiple iterations, the PSO algorithm can automatically adjust these parameters to minimize the prediction error of the SVR and LSTM models on the training data, thereby improving the overall performance of the model.

[0082] After the PSO optimization is completed, the optimized parameters are used to train the SVR and LSTM models respectively, and the two are iteratively trained and fused. Specifically, the SVR model is used to capture the short-term trend of water level changes, and the LSTM model is used to learn the long-term change law. The prediction results of the two are weighted fused to obtain the final water level prediction value. The distribution of weights can be dynamically adjusted according to the performance of the model on the validation set to ensure the accuracy and robustness of the fusion results. Finally, the water level prediction model based on PSO-SVR-LSTM can efficiently and accurately predict the water level changes in the target area, providing reliable data support for water resources regulation and optimal configuration.

[0083] Specifically, the core of the image model construction module is to establish a three-dimensional image model of the shallow recharge area and the deep storage area based on radar remote sensing images. First, the distribution image data of surface water and groundwater in the target area are obtained through radar remote sensing technology, and these data are preprocessed (such as denoising, normalization and splicing). Then, according to the geological characteristics of the shallow recharge area and the deep storage area, these areas are divided into recharge layer, transmission layer and receiving layer using three-dimensional modeling technology. The recharge layer represents the source of water resources in the shallow recharge area, the receiving layer represents the water resource storage area in the deep storage area, and the transmission layer represents the water resource transmission channel between the two. The three-dimensional image model can clearly display the structure of the shallow recharge area and the deep storage area and their relationship through precise spatial resolution and hierarchical division.

[0084] In the three-dimensional image model, several channel openings are set in the recharge layer and the receiving layer to represent the input and output positions of water resources; H transport channels (H>0) are set in the transport layer to simulate the path of transporting water resources from the shallow recharge area to the deep storage area. The number and distribution of transport channels are set according to the geological structure and water resource flow characteristics reflected in the radar remote sensing image. By parametrically modeling the geometry, length and width of the transport channel, the transport capacity of water resources in different areas can be dynamically reflected. In addition, the reflection characteristics of the radar laser signal are combined in the model to capture the dynamic changes of water resources in the transport channel.

[0085] In order to realize dynamic monitoring of water resource flow, the multiple reflection mutation characteristics of radar laser signals are used to obtain the total amount of water resources transported from the shallow recharge area to the deep storage area in each evaluation period. By analyzing the spectrum changes of the radar laser reflection signal, the water resource flow rate and unit water content in the transmission channel are calculated, and these data are dynamically displayed in the three-dimensional transmission image model. The water resource flow situation in each evaluation period is recorded in units of time nodes to form a series of dynamic transmission images, which intuitively show the water resource transmission process from the shallow recharge area to the deep storage area.

[0086] When establishing the space-time coordinate system, this embodiment combines the geographic spatial information of the target area with the time dimension to form a four-dimensional space-time framework. The three-dimensional image models of the shallow recharge area and the deep storage area are mapped into the space-time coordinate system, so that the water resource delivery situation at each time node can be accurately located in space. Through this mapping, the water replenishment potential of the shallow recharge area and the water storage capacity of the deep storage area in different time periods can be dynamically displayed. The establishment of the space-time coordinate system not only provides an intuitive visualization effect, but also provides data support for subsequent water resource scheduling and optimization.

[0087] In the space-time coordinate system, the water recharge potential of the shallow recharge area is calculated based on the dynamic data of the three-dimensional image model of the transport. Specifically, the total amount of water resources transported from the recharge layer to the receiving layer through each transport channel in the transport layer during different evaluation periods is analyzed to obtain the recharge water flow rate V and unit water content θ. The water recharge potential of each transport channel is calculated by combining the time length t of the evaluation period and the divergence div(V) of the groundwater flow. Finally, the water recharge potential of the shallow recharge area in all evaluation periods is accumulated and averaged to obtain the water recharge potential Q of the corresponding groundwater distribution area. This calculation process can accurately quantify the water resource recharge capacity of the shallow recharge area.

[0088] This embodiment evaluates the water storage capacity of the deep storage area by mapping the three-dimensional image model of the deep storage area into the space-time coordinate system, combining the water resource input of the transmission channel and the geological characteristics of the deep storage area. The evaluation of water storage capacity mainly considers the size of the water storage space, permeability and dynamic changes of water resources in the deep storage area. By analyzing the changes in the total amount of water resources in the receiving layer at different time nodes, the water storage capacity of the deep storage area can be dynamically evaluated, providing a scientific basis for subsequent water resource scheduling. Ultimately, the water replenishment potential of the shallow recharge area and the water storage capacity of the deep storage area together constitute the basic data for the optimal allocation of water resources in the target area.

[0089] Preferably, the formula of the optimization objective function is:

[0090]

[0091] Among them, Z is the optimization target value, which represents the comprehensive optimization target of supply and demand balance deviation, time balance deviation, and economic and ecological costs; S i is the water supply of the ith region, including surface water and groundwater; D i is the water demand of the ith region, including agriculture, industry, life, and ecology, ΔW t is the water level fluctuation within time t, that is, the absolute value of the water level change; C eco is the regional economic cost; C env is the regional ecological cost; 1 ,ω 2 ,ω 3 ,ω 4 are weight coefficients used to balance the importance of supply-demand deviation, time balance, economic cost and ecological cost; N is the total number of spatial nodes in the target area; T is the total number of steps in the optimization time period;

[0092] The constraints include: water resource supply and demand balance constraint, water resource total amount constraint and water replenishment potential constraint; the expression of water resource supply and demand balance constraint is: S i ≥D i , The expression of the total water resources constraint is: Among them, W total is the total available water resources in the target area, including surface water and groundwater; the expression of the water replenishment potential constraint is: in, is the water supply to the shallow recharge area of ​​the ith region; The water supply to the deep storage area of ​​the ith region; is the recharge potential of the shallow recharge area in the ith region; is the water storage capacity of the deep storage area in the ith region.

[0093] Preferably, the optimization algorithm for solving the optimization objective function is a particle swarm algorithm.

[0094] Specifically, in the monitoring and optimization module of this embodiment, it is first necessary to construct an optimization objective function and related constraints. The optimization objective function comprehensively considers the effects of supply and demand balance deviation, time balance deviation, economic cost and ecological cost, aiming to achieve spatial balance optimization and time balance optimization of water resources. Based on the real-time monitored water level information and the predicted water supply and demand information, the objective function can dynamically adjust the water supply to meet the water demand requirements, while controlling the water level fluctuation range and reducing regional economic and ecological costs. Constraints include water resource supply and demand balance constraints, water resource total amount constraints and water replenishment potential constraints, which are used to ensure the feasibility of the optimization results. The supply and demand balance constraint ensures that the water supply in each area meets the water demand; the water resource total amount constraint limits the total water supply in the target area to not exceed the total available amount of surface water and groundwater; the water replenishment potential constraint ensures that the water supply in the shallow recharge area and the deep storage area does not exceed their water replenishment potential and water storage capacity.

[0095] For solving the optimization objective function, the particle swarm optimization algorithm (PSO) is used to improve the efficiency and accuracy of the solution. First, the parameters of the particle swarm are initialized, including the number of particles, the maximum number of iterations, the inertia weight, and the learning factor. Each particle represents a possible water resource allocation scheme, including the water supply in each area and the water level fluctuation at the time node. The initial position and velocity of the particle are randomly generated and constrained in the search space to ensure that the initial solution meets the constraints of supply and demand balance, total water resources, and water replenishment potential. The fitness value of the objective function is used to evaluate the pros and cons of each particle. The lower the fitness value, the closer the water resource allocation scheme of the particle is to the optimal one.

[0096] During the iteration process of the algorithm, particles dynamically adjust their speed and position according to their own historical optimal position and the global optimal position. Specifically, when each particle moves in the search space, it will be affected by two important factors: one is its own experience, that is, the particle's historical optimal position; the other is the experience of the group, that is, the global optimal position. By continuously updating the position and speed of the particles, the particle swarm gradually approaches the optimal solution. In each iteration, the current fitness value of each particle is calculated, and it is determined whether it meets the requirements of the optimization objective function. If the current solution does not meet the requirements, the search direction of the particle is adjusted according to the gradient information of the objective function, and at the same time, it is checked whether all constraints are met. Through multiple iterations, the particle swarm can quickly converge to the optimal solution area.

[0097] When the particle swarm algorithm reaches the maximum number of iterations or the fitness value meets the set accuracy requirements, the optimal water resource allocation plan is output as the optimization result. The optimization results include the water supply of each area and the water level fluctuation at the time node, which can achieve the spatial balance and temporal balance of water resources in the target area. At the same time, the monitoring and optimization module will dynamically feedback the optimization results and the real-time monitored water level information. If there is a deviation between the actual monitoring data and the optimization result, the weight coefficient and constraint conditions of the optimization objective function are readjusted, and the particle swarm algorithm is re-run for iterative optimization. Through this dynamic control process, the monitoring and optimization module of this embodiment can continuously optimize the water resource allocation plan to ensure the rational use and sustainable management of water resources in the target area.

[0098] Optionally, the scheduling management module of this embodiment first generates a water resource scheduling plan for the target area based on the optimization results output by the monitoring optimization module. The optimization results include the water supply of each area (the specific water supply allocated to the shallow recharge area and the deep storage area) and the water level fluctuation information at the time node. The scheduling management module converts the optimization results into specific scheduling instructions, including the allocation path of water resources, water delivery, scheduling time and priority. The scheduling plan will combine the actual needs of the region (such as agricultural irrigation, industrial water, domestic water and ecological water needs) to ensure the scientificity and rationality of water resource allocation, while giving priority to key areas and high-priority needs.

[0099] During the scheduling implementation process, the scheduling management module uses real-time monitoring data to dynamically control the delivery and allocation of water resources. By integrating the water supply capacity of shallow recharge areas and deep storage areas, the module coordinates water delivery scheduling between different regions to achieve spatial balance optimization. For example, when the water demand in a certain area surges or the water level drops beyond the threshold, the system will dynamically adjust the water supply allocated from the shallow recharge area or deep storage area to ensure regional supply and demand balance. At the same time, the module will also optimize the timing of water resource delivery based on the water level fluctuation information at the time node to avoid drastic fluctuations in water levels due to excessive extraction or delivery.

[0100] The dispatching management module optimizes the water transmission network in the target area to ensure that water resources are transported through the optimal path. Based on the optimization results and the geographical information in the area, the module selects water transmission pipelines, channels or natural rivers as the main path for water resource allocation, and evaluates and dynamically adjusts the water transmission capacity. For example, the module will give priority to paths with high water transmission efficiency and low energy consumption, while avoiding resource waste and ecological damage caused by long water transmission distances or excessive flow rates. In addition, during the water transmission process, the module will monitor the operating status of the water transmission network in real time, such as flow rate, pipeline pressure and water transmission volume, to ensure the safety and efficiency of the water transmission process.

[0101] During the implementation process, the scheduling management module will continuously monitor the water level changes, supply and demand balance, and water transmission network operation in the target area, and compare the monitoring data with the optimization results. If there is a deviation between the monitoring data and the actual scheduling results (such as increased water demand, reduced water transmission efficiency, or water level fluctuations exceeding the threshold), the module will dynamically adjust the scheduling plan. For example, increase the water transmission volume in high water demand areas, or re-plan the water transmission route to improve scheduling efficiency. Through this dynamic feedback mechanism, the scheduling management module can continuously optimize the water resource allocation and transmission plan, achieve the optimal regulation of water resources in space and time, and ensure the sustainability and efficiency of regional water resource management.

[0102] The beneficial effects of the present invention are as follows:

[0103] (1) The present invention realizes dynamic monitoring of surface water and groundwater through infrared remote sensing technology and radar remote sensing technology, combined with real-time monitoring of water level information, and can timely grasp the spatiotemporal distribution of water resources in the target area; by optimizing the objective function and the scheduling management module, the dynamic scheduling of water resources is realized to meet the water demand at different time and space nodes.

[0104] (2) Based on the historical water level database and water level prediction model, the present invention accurately predicts the water supply and demand information of the target area, avoiding scheduling errors caused by prediction bias in traditional water resource management; by constructing a three-dimensional image model of the shallow recharge area and the deep storage area and mapping it to the space-time coordinate system, the water replenishment potential and water storage capacity are accurately calculated, thereby improving the scientific nature of the water resource scheduling plan.

[0105] (3) The present invention analyzes the water replenishment potential and water storage capacity of shallow recharge areas and deep storage areas, rationally allocates water resources, optimizes the allocation of water resources among different regions, and alleviates the problem of uneven temporal and spatial distribution of water resources.

[0106] (5) The present invention controls the range of water level fluctuations through real-time monitoring of water level fluctuations and solving the optimization objective function, thereby ensuring the stability of the water supply system and avoiding the impact of drastic changes in water levels on the ecology and water supply system.

[0107] (5) The present invention takes into account both regional economic costs and ecological costs in the optimization objective function, achieves a balance between economic and ecological benefits through comprehensive evaluation, and improves the comprehensive benefits of water resources scheduling and management.

[0108] (6) The present invention combines remote sensing technology, spatiotemporal data analysis and optimization algorithms (such as PSO optimization algorithm) to realize the intelligence and automation of water resource scheduling and management system, reduce the complexity of manual intervention and improve management efficiency.

[0109] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0110] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A real-time monitoring and dispatching management system for water resources based on spatiotemporal data, characterized in that: include: A data acquisition module is used to acquire the spatiotemporal distribution data of surface water and groundwater in the target area through infrared remote sensing technology and radar remote sensing technology; the spatiotemporal distribution data includes water resource distribution images of shallow recharge areas and deep storage areas; A water supply and demand acquisition module is used to construct a water level prediction model based on a historical water level database, and predict the water supply and demand information in the target area through the water level prediction model; An image model building module is used to establish a spatiotemporal coordinate system of the target area and to build a three-dimensional image model of the shallow recharge area and the deep storage area based on the water resource distribution image; A mapping module, for mapping the three-dimensional image models of the shallow recharge area and the deep storage area onto the space-time coordinate system to obtain the water recharge potential of the shallow recharge area and the water storage capacity of the deep storage area; A monitoring optimization module is used to obtain real-time monitoring water level information, and to construct an optimization objective function based on the real-time monitoring water level information and the water supply and demand information, and to solve the problem with the water replenishment potential and the water storage capacity as constraints to obtain an optimization result; The scheduling management module is used to schedule the water resources in the target area according to the optimization result to achieve management based on optimal spatial balance optimization and temporal balance optimization.

2. The water resources real-time monitoring and dispatching management system based on spatiotemporal data according to claim 1 is characterized in that: The water level prediction model is a network structure based on PSO-SVR-LSTM.

3. The water resources real-time monitoring and dispatching management system based on spatiotemporal data according to claim 1 is characterized in that: The data acquisition module comprises: An image acquisition submodule is used to establish a UAV circular flight route in the target area, install an infrared remote sensing device and a radar remote sensing device on the UAV, set a number of UAVs to fly according to the UAV circular flight route, scan the target area through the infrared remote sensing device, and obtain infrared image data covered by the corresponding UAV circular flight route; An image denoising submodule, used for denoising the infrared image data; An image stitching submodule, used for overlapping and stitching the infrared image data according to the geographical location corresponding to the target area to obtain stitching image data; A region division submodule is used to set a temperature interval threshold, mark the area in the stitched image data whose temperature is not within the temperature interval threshold as a surface water distribution area, and mark the area in the stitched image data whose temperature is within the temperature interval threshold as a groundwater distribution area; The data determination submodule is used to determine the corresponding spatiotemporal distribution data according to the surface water area and the groundwater distribution area.

4. The water resources real-time monitoring and dispatching management system based on spatiotemporal data according to claim 3 is characterized in that: The image denoising submodule comprises: A noise monitoring unit, used for detecting noise points on the infrared image data using a filter window to obtain a noise mean; A denoising unit, which denoises the infrared image data in the corresponding filtering window when the noise mean in the filtering window is greater than a preset threshold; The window sliding unit is used to slide the filter window until the entire infrared image data is traversed.

5. The water resources real-time monitoring and dispatching management system based on spatiotemporal data according to claim 4 is characterized in that: The noise monitoring unit comprises: The detection model construction subunit is used to construct a noise point detection model according to the mean and median of each image point in the filtering window; the noise point detection model is: Among them, f(x) represents the similar noise value of pixel x, u(x) represents the gray value of pixel x, and u mean (x) represents the grayscale mean of all pixels in the filter window centered on pixel x. represents the mean gradient of pixel x, is the grayscale median of all pixels in the filter window centered on pixel x, Represents the gradient value of pixel x in the horizontal direction. Represents the gradient value of pixel x in the vertical direction; A detection subunit, used to detect each image point in the filtering window using the noise point detection model to obtain a similar noise value for each image point; A noise determination subunit, used for treating corresponding image points with a value greater than a similar noise value as noise points; The mean value determination subunit obtains the noise mean value according to the number of noise points and the number of each image point in the filter window.

6. The water resources real-time monitoring and dispatching management system based on spatiotemporal data according to claim 5 is characterized in that: The denoising unit comprises: The variance calculation subunit is used to calculate the pseudo pixel variance according to the grayscale median of all pixels in the filter window; wherein the calculation formula of the pseudo pixel variance is: in, represents the pseudo pixel variance of the pixel point (a, b) in the area of ​​the filter window with a size of (2n+1)×(2n+1), mean(a,b) represents the grayscale median of the pixel point (a,b) in the filter window, and x(k,l) represents the grayscale value of the pixel point at the position (k,l); The denoising model construction subunit is used to construct a window denoising model using the pseudo pixel variance; the calculation formula of the window denoising model is: Among them, f(a, b) represents the gray value of the pixel (a, b) after denoising, D is the adjustable coefficient, and x(a, b) represents the gray value of the pixel (a, b) in the filter window.

7. The water resources real-time monitoring and dispatching management system based on spatiotemporal data according to claim 1 is characterized in that: The formula of the optimization objective function is: Among them, Z is the optimization target value, which represents the comprehensive optimization target of supply and demand balance deviation, time balance deviation, and economic and ecological costs; S i is the water supply of the ith region, including surface water and groundwater; D i is the water demand of the ith region, including agriculture, industry, life, and ecology, ΔW t is the water level fluctuation within time t, that is, the absolute value of the water level change; C eco is the regional economic cost; C env is the regional ecological cost; ω1, ω2, ω3, ω4 are weight coefficients used to balance the importance of supply and demand deviation, time balance, economic cost and ecological cost; N is the total number of spatial nodes in the target area; T is the total number of steps in the optimization time period; The constraints include: water resource supply and demand balance constraint, water resource total amount constraint and water replenishment potential constraint; the expression of water resource supply and demand balance constraint is: The expression of the total water resources constraint is: Among them, W total is the total available water resources in the target area, including surface water and groundwater; the expression of the water replenishment potential constraint is: in, is the water supply to the shallow recharge area of ​​the ith region; The water supply to the deep storage area of ​​the ith region; is the recharge potential of the shallow recharge area in the ith region; is the water storage capacity of the deep storage area in the ith region.

8. The water resources real-time monitoring and dispatching management system based on spatiotemporal data according to claim 1 is characterized in that: The optimization algorithm for solving the optimization objective function is a particle swarm algorithm.

Images