Water resource scheduling optimization method and system based on artificial intelligence

Abstract

The invention provides a water resource scheduling optimization method and system based on artificial intelligence. The method comprises the following steps: acquiring water resource multi-source heterogeneous data of a target area, generating a dynamic balance strategy according to a precipitation gradient change trend of precipitation distribution characteristic data and a consumption rate association relationship in user side water demand fluctuation mode data, and determining the water demand fluctuation mode of the user side on the basis of node pressure distribution characteristics of dynamic load data of a water delivery pipe network. Outputting a multi-target water delivery path optimization scheme in combination with a priority weight in a dynamic balance strategy, and converting the multi-target water delivery path optimization scheme into a scheduling instruction set for controlling a pump station valve in the geographic space topology; according to the technical scheme provided by the invention, the defects of non-uniform resource allocation, poor real-time response capability and the like of a traditional static scheduling strategy are overcome, the utilization rate of water resources is also improved, and the flexibility and the response speed of a water supply system are enhanced.

Description

Technical Field

[0001] This application relates to the technical field of water resource scheduling, and particularly to an artificial intelligence-based water resource scheduling optimization method and system. Background Art

[0002] With the acceleration of urbanization and industrialization, the demand for water resources has increased sharply, and the uneven distribution and seasonal variation of water resources have further exacerbated the contradiction between supply and demand. In agricultural irrigation, industrial production, and residential life, it is particularly important to rationally schedule water resources. Especially in arid or semi-arid regions, how to dynamically adjust the water supply strategy according to precipitation, the load of water pipelines, and the changes in user water demand has become the key to ensuring the efficient use of water resources.

[0003] Current water resource scheduling mainly relies on traditional hydraulic models and empirical rules, and plans by analyzing historical precipitation data, predicting future precipitation, and evaluating the carrying capacity of water pipelines. These methods usually adopt fixed scheduling rules, such as allocating water sources according to a fixed schedule or adjusting the working state of pump station valves based on a simple threshold trigger mechanism. Although these methods can ensure basic water supply requirements to a certain extent, they lack the ability to effectively respond to real-time changing factors (such as sudden precipitation events or sudden fluctuations in user water demand). Summary of the Invention

[0004] This application provides an artificial intelligence-based water resource scheduling optimization method and system to solve the problems of uneven resource allocation, poor ability to respond to real-time changes, and low operating efficiency caused by relying on traditional static scheduling strategies in the prior art.

[0005] In a first aspect, this application provides an artificial intelligence-based water resource scheduling optimization method, including:

[0006] Obtain multi-source heterogeneous data of water resources in the target area, where the multi-source heterogeneous data of water resources includes precipitation distribution characteristic data that changes with seasons, dynamic load data of water pipelines based on geographical space topology, and user-side water demand fluctuation pattern data;

[0007] Generate a dynamic balance strategy according to the precipitation gradient change trend of the precipitation distribution characteristic data and the consumption rate correlation relationship in the user-side water demand fluctuation pattern data;

[0008] Based on the node pressure distribution characteristics of the dynamic load data of the water pipelines, combine the priority weights in the dynamic balance strategy to output an optimized multi-objective water conveyance path plan;

[0009] Convert the optimized multi-objective water conveyance path plan into a scheduling instruction set for controlling pump station valves in the geographical space topology.

[0010] Optionally, based on the node pressure distribution characteristics in the dynamic load data of the water supply pipeline network, and in combination with the priority weights in the dynamic balance strategy, a multi-objective water conveyance path optimization scheme is generated, including:

[0011] Based on the node pressure distribution characteristics in the dynamic load data of the water supply pipeline network, the pressure volatility rate and pressure balance degree index of each water supply pipeline network node within a continuous monitoring period are extracted;

[0012] The priority weights in the dynamic balance strategy are split into a water source replenishment urgency coefficient and a user - end demand deviation coefficient;

[0013] The product of the pressure volatility rate and the water source replenishment urgency coefficient is used as the hydraulic coupling degree of the water supply pipeline network node, and based on the hydraulic coupling degree, the branch pipelines in the geospatial topology are initially sorted to generate an initial path sorting result;

[0014] Based on the water flow inertia factor in the dynamic load data of the water supply pipeline network, the inertia decay coefficient of each branch pipeline during the flow regulation process is calculated;

[0015] According to the user - end demand deviation coefficient, the initial path sorting result is dynamically adjusted to generate a multi - objective water conveyance path optimization scheme.

[0016] Optionally, according to the user - end demand deviation coefficient, the initial path sorting result is dynamically adjusted to generate a multi - objective water conveyance path optimization scheme, including:

[0017] A path sorting correction factor is generated based on the ratio of the user - end demand deviation coefficient to the inertia decay coefficient, and the priority weights of the branch pipelines in the initial path sorting result are scaled proportionally using the path sorting correction factor to obtain the scaled - down priority weights;

[0018] According to the scaled - down priority weights, the flow distribution ratios of each branch pipeline are updated to obtain the updated flow distribution ratios. Based on the updated flow distribution ratios, a set of flow distribution ratios is generated, and the change rate of the node pressure variance of the water supply pipeline network is calculated in combination with the pressure balance degree index;

[0019] According to the water flow inertia factor in the dynamic load data of the water supply pipeline network, based on the proportional change amplitude of each branch pipeline in the dynamically adjusted set of flow distribution ratios, the path redundancy decay coefficient of the corresponding branch pipeline is calculated;

[0020] When the change rate of the pressure variance does not reach the preset optimization threshold and the path redundancy attenuation coefficient is higher than the preset redundancy critical value, the product of the reciprocal of the water source replenishment urgency coefficient and the change rate of the pressure variance is used as the correction factor adjustment amount;

[0021] Based on the correction factor adjustment amount, update the path sorting correction factor, and re-execute scaling the priority weights, updating the flow distribution ratio, calculating the change rate of the pressure variance, and the path redundancy attenuation coefficient until the change rate of the pressure variance is less than the preset optimization threshold and the path redundancy attenuation coefficient is lower than the preset redundancy critical value, and generate a multi-objective water conveyance path optimization scheme.

[0022] Optionally, according to the correlation relationship between the precipitation gradient change trend of the precipitation distribution characteristic data and the consumption rate in the user-side water demand fluctuation pattern data, generate a dynamic balance strategy, including:

[0023] Based on the precipitation gradient change trend, divide the precipitation replenishment time interval of the target area within the preset time period, and calculate the cumulative precipitation gap of each precipitation replenishment time interval to generate a water source replenishment urgency index at the seasonal scale;

[0024] According to the user-side water demand fluctuation pattern data, extract the correlation relationship of the consumption rate in different daily time periods, and calculate the water demand deviation rate of each daily time period;

[0025] Couple the water source replenishment urgency index and the water demand deviation rate in space and time to generate dynamic constraint conditions across time scales;

[0026] Based on the coupled constraint conditions, establish an iterative optimization model, and use the iterative optimization model to constrain the total water source allocation within the precipitation replenishment time interval and the weighted sum of the water demand deviation rates of each daily time period to be minimized, and generate a dynamic balance strategy that meets the cross-time scale resource matching.

[0027] Optionally, couple the water source replenishment urgency index and the water demand deviation rate in space and time to generate dynamic constraint conditions across time scales, including:

[0028] According to the number of days between preset time periods in the precipitation replenishment time interval, calculate the time decay factor, and multiply the time decay factor by the water source replenishment urgency index to generate a basic weight of the replenishment priority distributed by daily granularity;

[0029] Based on the absolute value of the difference between the peak period deviation rate and the valley period deviation rate in the water demand deviation rate, and combining the water smoothing coefficient of the same type of daily time period within the historical irrigation cycle, calculate the peak-valley compensation intensity of each daily time period;

[0030] Align the basic weight of the supply priority with the peak-valley compensation intensity according to daily time periods to generate a spatio-temporal correlation matrix;

[0031] Calculate the pressure fluctuation suppression coefficient of each branch pipeline according to the node pressure balance degree index in the dynamic load data of the water supply pipeline network, and embed the pressure fluctuation suppression coefficient as a weight into the column vector of the spatio-temporal correlation matrix to obtain a weighted spatio-temporal correlation matrix;

[0032] Perform principal component analysis on the weighted spatio-temporal correlation matrix, extract the eigenvectors that satisfy the coupling relationship between the water source supply priority and the peak-valley time periods of user-side water consumption, and generate dynamic constraint conditions across time scales.

[0033] Optionally, convert the multi-objective water conveyance path optimization scheme into a scheduling instruction set for controlling the pump station valves in the geospatial topology, including:

[0034] Generate an initial pump station valve opening and closing timing parameter set based on the branch pipeline flow distribution ratio and path redundancy attenuation coefficient in the multi-objective water conveyance path optimization scheme;

[0035] Calculate the inertia compensation coefficient during the flow regulation process of each branch pipeline according to the water flow inertia factor in the dynamic load data of the water supply pipeline network and in combination with the water source supply priority weight in the dynamic balance strategy;

[0036] Calculate the pressure fluctuation suppression threshold of each monitoring node based on each monitoring node of the water supply pipeline network corresponding to the pressure balance degree index;

[0037] Align the initial pump station valve opening and closing timing parameter set and the inertia compensation coefficient in terms of time and space according to the pipeline connection relationship in the geospatial topology to generate a dynamic regulation instruction template, where the flow regulation parameter of each pump station valve in the dynamic regulation instruction template is the product of the corresponding parameter in the opening and closing timing parameter set and the inertia compensation coefficient;

[0038] Perform segmented correction on the dynamic regulation instruction template according to the pressure fluctuation suppression threshold until the pressure fluctuation suppression thresholds of all monitoring nodes are within a preset safe range, to obtain a scheduling instruction set for controlling the pump station valves in the geospatial topology, where the scheduling instruction set includes the opening and closing timing parameters, flow regulation parameters of the pump station valves, and the corresponding pressure fluctuation suppression threshold warning levels.

[0039] Optionally, perform segmented correction on the dynamic regulation instruction template according to the pressure fluctuation suppression threshold until the pressure fluctuation suppression thresholds of all monitoring nodes are within a preset safe range to generate a scheduling instruction set for controlling the pump station valves in the geospatial topology, including:

[0040] Calculate the predicted pressure fluctuation values of each monitoring node based on the flow rate adjustment parameters in the dynamic adjustment instruction template;

[0041] Generate a pipeline adjustment sensitivity coefficient according to the ratio between the path redundancy attenuation coefficient and the water source replenishment priority weight;

[0042] Establish a pressure fluctuation - flow rate adjustment feedback matrix, where the row vector of the pressure fluctuation - flow rate adjustment feedback matrix represents the predicted pressure fluctuation values of each monitoring node, and the column vector represents the pipeline adjustment sensitivity coefficients of the corresponding branch pipelines;

[0043] Perform eigenvalue decomposition on the pressure fluctuation - flow rate adjustment feedback matrix, and extract the eigenvectors with eigenvalues greater than the preset eigenvalue threshold as the key adjustment channels;

[0044] Determine the target branch pipeline and the adjustment direction based on the eigenvector element values of each key adjustment channel, and update the flow rate adjustment parameters according to the maximum allowable adjustment amplitude of the inertia compensation coefficient;

[0045] Based on the updated flow rate adjustment parameters, recalculate the pressure fluctuation suppression threshold of the associated monitoring nodes;

[0046] If the recalculated pressure fluctuation suppression threshold does not reach the preset safety interval, optimize the recalculated pressure fluctuation suppression threshold by reducing the adjustment amplitude in geometric progression according to the pipeline adjustment sensitivity coefficient until the pressure fluctuation suppression thresholds of the monitoring nodes corresponding to all key adjustment channels are within the preset safety interval, so as to generate a scheduling instruction set for controlling the pump station valves in the geospatial topology.

[0047] In a second aspect, the present application provides an artificial - intelligence - based water resource scheduling optimization system, including:

[0048] An acquisition module, configured to acquire multi - source heterogeneous data of water resources in a target area, where the multi - source heterogeneous data of water resources includes precipitation distribution characteristic data varying with seasons, dynamic load data of a water supply pipeline network based on a geospatial topology, and user - side water consumption demand fluctuation pattern data;

[0049] A generation module, configured to generate a dynamic balance strategy according to the precipitation gradient change trend in the precipitation distribution characteristic data and the consumption rate correlation relationship in the user - side water consumption demand fluctuation pattern data;

[0050] An output module, configured to output a multi - objective water conveyance path optimization scheme based on the node pressure distribution characteristics of the water supply pipeline network dynamic load data and in combination with the priority weight in the dynamic balance strategy;

[0051] A conversion module, configured to convert the multi-objective water conveyance path optimization scheme into a scheduling instruction set for controlling the pump station valves in the geospatial topology.

[0052] In a third aspect, an embodiment of the present application provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement an artificial intelligence-based water resource scheduling optimization method as described in the first aspect above.

[0053] In a fourth aspect, an embodiment of the present application provides a computer storage medium storing a computer program, which, when executed by a computer, implements an artificial intelligence-based water resource scheduling optimization method as described in the first aspect.

[0054] In the embodiment of the present application, by comprehensively analyzing multi-source heterogeneous data in the target area, including key information such as precipitation distribution characteristics, dynamic load of the water supply network, and user water demand fluctuation patterns, intelligent optimization scheduling of water resources is achieved. This method uses a dynamic balance strategy combined with node pressure distribution characteristics to generate a multi-objective water conveyance path optimization scheme, and converts this scheme into a specific pump station valve scheduling instruction set, which not only improves the utilization rate of water resources, but also enhances the flexibility and response speed of the water supply system. This enables the system to still ensure efficient and stable operation when facing complex and changeable environmental conditions, effectively solving the deficiencies of traditional static scheduling strategies in aspects such as uneven resource allocation and poor real-time response ability.

[0055] Furthermore, a more refined data processing and calculation mechanism is introduced in the process of generating the multi-objective water conveyance path optimization scheme. By extracting the pressure volatility and pressure balance degree index of each water supply network node, and combining the priority weights after splitting the water source replenishment urgency coefficient and the user end demand deviation coefficient, the hydraulic coupling degree of each branch pipeline can be more accurately evaluated and adjusted, so as to effectively sort the initial paths. At the same time, based on the water flow inertia factor, the inertia decay coefficient is calculated, and the initial path sorting result is dynamically adjusted accordingly, and finally a more optimized water conveyance path scheme is generated. This method significantly improves the accuracy and adaptability of the entire scheduling system, ensuring the safe and stable supply of water resources even in extreme cases, and greatly improving the reliability and efficiency of the system.

[0056] These aspects or other aspects of the present application will be more clearly understood in the following description of the embodiments. Description of the Drawings

[0057] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0058] Figure 1 The flowchart of an artificial intelligence-based water resource scheduling optimization method provided by the present application is shown;

[0059] Figure 2 The structural schematic diagram of an artificial intelligence-based water resource scheduling optimization system provided by the present application is shown;

[0060] Figure 3 The structural schematic diagram of a computing device provided by the present application is shown. Detailed implementation manners

[0061] To enable those skilled in the art to better understand the solutions of the present application, the following will clearly and completely describe the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application.

[0062] In some processes described in the specification and claims of the present application and the above drawings, a plurality of operations appear in a specific order. However, it should be clearly understood that these operations may not be executed in the order in which they appear in this document or may be executed in parallel. The operation numbers such as 101 and 102 are only used to distinguish different operations, and the numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions such as "first" and "second" in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequence, nor do they limit that "first" and "second" are of different types.

[0063] An artificial intelligence-based water resource scheduling optimization method proposed by the present application mainly focuses on how to efficiently and intelligently manage water resources, especially in the face of complex and changeable environmental conditions (such as seasonal precipitation changes, fluctuations in user water demand, etc.). This method dynamically adjusts the water supply strategy by integrating multiple data sources and using advanced data analysis and artificial intelligence technologies, so as to achieve the optimal allocation of resources. It first collects multi-source heterogeneous data in the target area, including precipitation distribution characteristic data that changes with seasons, dynamic load data of the water supply pipeline network, and data on the fluctuation patterns of user-side water demand. Then, it generates a dynamic balance strategy based on these data, and combines the node pressure distribution characteristics of the water supply pipeline network to output an optimized multi-objective water conveyance path plan. Finally, this plan is converted into a specific pump station valve scheduling instruction set to control the actual operation.

[0064] The following will clearly and completely describe the technical solutions in the embodiments of the present application in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present application.

[0065] Figure 1 A flowchart of an artificial intelligence-based water resource scheduling optimization method is provided for an embodiment of the present application, as Figure 1 shown, the method includes:

[0066] Step 101, obtain multi-source heterogeneous data of water resources in the target area, where the multi-source heterogeneous data of water resources includes precipitation distribution characteristic data that changes with seasons, dynamic load data of water supply pipe networks based on geospatial topology, and user-side water consumption demand fluctuation pattern data;

[0067] In this step, the multi-source heterogeneous data of water resources refers to various types of information related to water resources collected from multiple different sources and formats. These data include, but are not limited to, precipitation distribution characteristic data that changes with seasons, dynamic load data of water supply pipe networks based on geospatial topology, and user-side water consumption demand fluctuation pattern data. Among them, the precipitation distribution characteristic data describes the precipitation amount and its spatial distribution in a specific area at different time periods; the dynamic load data of the water supply pipe network records the operating parameters such as pressure and flow rate at each node in the entire water supply network; the user-side water consumption demand fluctuation pattern data reflects the change trend of the water consumption of users in different time periods.

[0068] In this embodiment, first, an integration platform needs to be established to automatically collect the above three types of data from multiple channels such as weather stations, sensor networks, and user databases. Through API interfaces or direct connections, these scattered data sources are integrated into a unified data warehouse and necessary preprocessing is performed, such as data cleaning and format conversion, to ensure the consistency and accuracy of the data. The data set also needs to be updated regularly to ensure that the data used is the latest, so as to accurately reflect the current environmental conditions and user needs.

[0069] For example, in a certain urban area prone to droughts, multiple meteorological observation points are set up to monitor precipitation in real time, and a series of sensors are installed on the main water supply pipes in the city to track the changes in water flow velocity and pressure. At the same time, the daily water consumption data of residential areas is collected through an intelligent water meter system. All these data are centrally transmitted to the central server and, after being standardized, provide solid data support for subsequent water resource scheduling optimization.

[0070] Step 102: Generate a dynamic balance strategy based on the precipitation gradient change trend in the precipitation distribution characteristic data and the consumption rate correlation relationship in the user terminal water demand fluctuation pattern data.

[0071] In this step, the dynamic balance strategy refers to a strategy that can flexibly adjust water resource allocation by analyzing the relationship between the precipitation gradient change trend (i.e., the change of precipitation in different time periods) and the user water consumption rate based on the precipitation distribution characteristic data and the user terminal water demand fluctuation pattern data; the precipitation gradient change trend refers to the change rate or change speed of precipitation in different time periods within a specific area. It reflects the change of precipitation over time (such as daily, weekly, monthly) and space (such as different geographical regions), and is one of the important indicators for analyzing precipitation patterns.

[0072] In this embodiment, first, statistical methods are used to calculate the average precipitation in different time periods, and a trend graph of precipitation change over time is drawn to determine the precipitation gradient change trend. Then, the user water consumption data is analyzed to identify the water use patterns during peak and off-peak periods. Next, this information is input into a machine learning model to establish a correlation model between precipitation and user water consumption. Based on the results output by the model, a set of dynamic balance strategies including a priority weight adjustment mechanism are formulated to ensure that while meeting the user water demand, the natural precipitation resources are utilized to the maximum extent. In addition, a monitoring and feedback mechanism needs to be set up to continuously track the implementation effect of the strategy and adjust and optimize it in a timely manner.

[0073] For example, in the above case of the urban area with frequent droughts, it is found that summer rainfall is concentrated from evening to night, while the peak of residents' water use appears in the morning and evening. By analyzing the meteorological data of the past few years, it is determined that the daily precipitation during summer shows a gradually increasing trend (i.e., the precipitation gradient change trend). Based on this, a dynamic balance strategy is formulated: reduce the water supply from the reservoir to the urban water supply network during the expected rainfall period, and improve the operation efficiency of the rainwater collection system; for example, when the forecast shows that there will be heavy rainfall every night in the next three days, the system will automatically reduce the water supply pressure during the day to encourage residents to reserve domestic water in advance; after the rainfall occurs, the system will adjust the water supply plan for the following days according to the actual rainfall data collected to ensure that normal water supply services can be maintained even in the case of insufficient rainfall; this not only effectively utilizes natural resources but also relieves the pressure on the water supply system and improves the overall operation efficiency.

[0074] Step 103: Based on the node pressure distribution characteristics of the water transmission pipeline network dynamic load data, combined with the priority weights in the dynamic balance strategy, output a multi-objective optimized water transmission path plan.

[0075] In this step, the dynamic load data of the water supply pipeline network refers to the data that describes the pressure and flow rate changes over time at each node (such as pump stations, valves, pipeline connection points, etc.) in the water supply network. These data reflect the operating status of the entire water supply system and its response ability to different water usage demands; the node pressure distribution characteristics refer to the pressure values and their fluctuations at each node. By analyzing these characteristics, the distribution of water flow in the pipeline network and potential bottlenecks can be understood; the multi-objective water conveyance path optimization scheme is a set of path adjustment suggestions generated based on the above data analysis results and combined with the priority weights in the dynamic balance strategy (such as the emergency coefficient of water source replenishment and the deviation coefficient of user-end demand), aiming to optimize the water resource allocation efficiency, reduce energy consumption, and ensure the stability of the system.

[0076] In this embodiment, first, the dynamic load data of the water supply pipeline network is obtained in real time from the sensor network, including the pressure and flow rate information of each node. Then, algorithms are used to extract the pressure volatility and pressure balance degree index of each node to evaluate the overall health status and efficiency of the pipeline network. Next, this information is combined with the previously formulated dynamic balance strategy to calculate the hydraulic coupling degree of each branch pipeline, and based on this, an initial path sorting is performed. In addition, the influence of the water flow inertia factor on flow regulation needs to be considered, and the inertia decay coefficient of each branch pipeline is calculated. Finally, according to the deviation coefficient of user-end demand, the preliminary path sorting result is dynamically adjusted to generate a multi-objective water conveyance path optimization scheme that can not only meet the user's water usage demand but also ensure the efficient operation of the system.

[0077] For example, in the case of the urban area with frequent droughts mentioned above, a dynamic balance strategy has been formulated for the precipitation gradient change trend and the user water usage demand fluctuation pattern. Now, further using the data collected by the sensors installed in the urban water supply pipeline network, it is found that there is a significant pressure drop at some key nodes during the peak water usage period. Based on this, the system automatically adjusts the flow distribution on some high-energy-consuming paths to give priority to ensuring the water supply to residential areas and important facilities; for example, when it is predicted that there will be heavy rainfall at night, the system will reduce the water supply to areas with insufficient rainwater collection and increase the water supply support to other areas to avoid service interruption caused by too low local pressure, which not only improves the utilization efficiency of water resources but also enhances the reliability and adaptability of the water supply system. As a natural extension of the previous step, the entire water resource management process is further improved.

[0078] Step 104, convert the multi-objective water conveyance path optimization scheme into a set of scheduling instructions for controlling the pump station valves in the geospatial topology;

[0079] In this step, the scheduling instruction set refers to a series of specific operation commands designed to adjust the working states of the pump station valves in the geographical space topology according to the multi-objective water conveyance path optimization scheme. These instructions include, but are not limited to, information such as the opening or closing time of the pump station valves, flow regulation parameters, and the corresponding pressure fluctuation suppression threshold warning levels. The role of the scheduling instruction set is to ensure that the entire water supply network can operate efficiently according to the predetermined optimized path, thereby achieving the best allocation of resources and the stable operation of the system.

[0080] In this embodiment, first, based on the branch pipeline flow distribution ratio and path redundancy attenuation coefficient in the multi-objective water conveyance path optimization scheme, an initial pump station valve opening and closing time sequence parameter set is generated. Then, by using the water flow inertia factor in the dynamic load data of the water supply network and combining the water source replenishment priority weight in the dynamic balance strategy, the inertia compensation coefficient during the flow regulation process of each branch pipeline is calculated. Next, according to each monitoring node of the water supply network corresponding to the pressure balance degree index, the pressure fluctuation suppression threshold of each monitoring node is calculated, and these parameters are spatially and temporally aligned with the initial pump station valve opening and closing time sequence parameter set to generate a dynamic regulation instruction template. Finally, the dynamic regulation instruction template is segmented and corrected according to the pressure fluctuation suppression threshold until the pressure fluctuation suppression thresholds of all monitoring nodes are within the preset safe range, forming the final scheduling instruction set for controlling the actual operation of the pump station valves in the geographical space topology.

[0081] For example, in the aforementioned case of the urban area with frequent droughts, after completing the multi-objective water conveyance path optimization scheme formulated based on the analysis of the precipitation gradient change trend, user water demand fluctuation pattern, and dynamic load data of the water supply network, now, the system automatically generates a detailed scheduling instruction set according to this scheme. For example, in the case of predicting heavy rainfall every night in the next three days, the system not only reduces the water supply volume to some urban areas during the day but also correspondingly adjusts the opening and closing times of the relevant pump station valves and the flow regulation parameters. Specifically, for those areas with relatively perfect rainwater collection systems, the system will reduce the working frequency of the pump stations in advance and reduce the water supply volume to these areas; while for the areas relying on external water source supply, the working intensity of the pump stations will be increased to ensure sufficient water supply. In this way, both the natural precipitation is fully utilized and the reasonable allocation of water resources in the whole city is ensured, further improving the response speed and efficiency of the water supply system, and as an extension of the previous step, making the entire water resource management process more perfect and intelligent.

[0082] To address the lack of effective response capabilities to real-time changing factors (such as sudden precipitation events or sudden fluctuations in user water demand) in traditional water resource scheduling schemes, and to enhance the flexibility, efficiency, and stability of the entire water supply system, in some embodiments, as described in step 103, based on the node pressure distribution characteristics in the dynamic load data of the water transmission pipeline network, and in combination with the priority weights in the dynamic balance strategy, a multi-objective optimized water transmission path scheme is generated, including:

[0083] Step 201: Based on the node pressure distribution characteristics in the dynamic load data of the water transmission pipeline network, extract the pressure volatility and pressure balance degree index of each water transmission pipeline network node within a continuous monitoring period.

[0084] In this step, the dynamic load data of the water transmission pipeline network refers to the data describing the pressure and flow rate changes of each node (such as pump stations, valves, pipeline connection points, etc.) in the water supply network over time. These data reflect the operating state of the entire water supply system and its response capabilities to different water use demands; the node pressure distribution characteristics refer to the pressure values and their fluctuations of each node. By analyzing these characteristics, the water flow distribution and potential bottlenecks in the pipeline network can be understood; the pressure volatility is an index measuring the pressure change amplitude of a certain node within a specific time period, reflecting the pressure stability of that node; and the pressure balance degree index is a comprehensive index used to evaluate the pressure balance status among nodes in the entire pipeline network system. The higher the value, the more uniform the pressure distribution.

[0085] In this embodiment, first, the dynamic load data of the water transmission pipeline network is obtained in real time from the sensor network, including the pressure and flow rate information of each node. Then, statistical methods are used to calculate the pressure volatility of each node within a continuous monitoring period (such as one day or one week). Specifically, the difference between the maximum pressure and the minimum pressure of each node within this period is calculated and divided by the average pressure value to obtain a relative volatility. Then, the pressure balance degree index is further calculated, which usually involves standardizing the pressure values of all nodes and then calculating their standard deviation or coefficient of variation to obtain a value reflecting the overall pressure balance status. In addition, machine learning algorithms can also be used to analyze historical data to identify the key factors affecting pressure fluctuations and balance, providing a basis for subsequent optimization strategies.

[0086] Step 202: Split the priority weights in the dynamic balance strategy into a water source replenishment urgency coefficient and a user-side demand deviation coefficient.

[0087] In this step, the priority weight refers to the relative importance value assigned to different water supply targets or regions when formulating the dynamic balance strategy, which is used to guide the allocation and scheduling of resources; the water source replenishment urgency coefficient is an index for quantitatively evaluating the urgency of water source supply for each water supply region or node, which reflects the severity of the water resource shortage situation caused by insufficient precipitation or other reasons in a certain region; the user - end demand deviation coefficient is an index for measuring the difference between the actual water consumption of each user group and the expected or standard water consumption, which is used to reflect the change of user water demand in different time periods and its impact on the overall supply - demand balance.

[0088] In this embodiment, first, based on the dynamic balance strategy obtained from the previous analysis, identify the key water supply regions or nodes that need special attention and determine their corresponding priority weights. Next, through data analysis tools and algorithms (such as machine - learning models), calculate the water source replenishment urgency coefficient for each key node. Information such as historical precipitation data, current reservoir water levels, and future weather forecasts can be used to evaluate the urgency of water source supply. At the same time, according to the user water consumption data provided by the intelligent water meter system, calculate the user - end demand deviation coefficient in different time periods, which includes the comparison of water consumption in peak and trough periods, as well as the change in water - using patterns caused by special events (such as holidays). Then, split the original priority weight into two parts, namely the water source replenishment urgency coefficient and the user - end demand deviation coefficient, according to certain rules, so that each part can independently reflect the impact of its corresponding factor on the overall water supply strategy.

[0089] Step 203: Take the product of the pressure volatility and the water source replenishment urgency coefficient as the hydraulic coupling degree of the water pipeline network nodes, and perform an initial path sorting on the branch pipelines in the geospatial topology according to the hydraulic coupling degree to generate an initial path sorting result;

[0090] In this step, the hydraulic coupling degree of the water pipeline network nodes refers to a comprehensive index that combines the pressure fluctuation situation of the nodes and the urgency of water source supply, which is used to evaluate the importance and stability of each node in the water supply system; the branch pipelines refer to the pipeline segments in the water supply network that connect different nodes (such as pumping stations, valves, etc.), which are responsible for distributing water resources from the source to each user - end; the initial path sorting result is the result of prioritizing all branch pipelines based on the above - mentioned hydraulic coupling degree, which is used to guide subsequent resource scheduling and path optimization.

[0091] In this embodiment, first, the pressure volatility of each node is multiplied by its corresponding water source replenishment urgency coefficient to obtain a value representing the stability and urgency of the node. Then, based on these hydraulic coupling degree values, the branch pipes in the entire water supply network are preliminarily sorted. This process generally involves constructing a list containing all branch pipes and their corresponding hydraulic coupling degrees, and sorting them from high to low in terms of hydraulic coupling degree to determine which pipes should be given priority for adjustment or optimization. For example, at some key nodes, if it is found that their pressure fluctuations are large and the water source replenishment urgency is high, the branch pipes related to these nodes will obtain a higher priority in the sorting. The initially generated path sorting result provides a basic framework for subsequent further refinement and optimization, ensuring that when formulating specific scheduling strategies, those parts that most need adjustment can be given priority attention, thereby improving the operating efficiency and stability of the overall system.

[0092] Step 204: Calculate the inertia decay coefficient of each branch pipe during the flow regulation process based on the water flow inertia factor in the dynamic load data of the water transmission network.

[0093] In this step, the water flow inertia factor is a parameter that describes the ability of water flow in a pipe to resist changes due to the changes in its mass and velocity, reflecting the response characteristics of water flow when facing sudden changes in flow or pressure. The specific value can be determined by analyzing historical flow and pressure data; the inertia decay coefficient is an index that measures the difference between the actual flow adjustment amplitude and the ideal adjustment amplitude caused by water flow inertia during the flow regulation process. The higher the inertia decay coefficient, the greater the impact of water flow inertia on flow regulation, and the longer the time required for the system to reach a new stable state.

[0094] In this embodiment, first, relevant information on the water flow inertia factor is extracted from the dynamic load data of the water transmission network. This generally includes monitoring the historical flow data of each branch pipe and identifying the characteristic parameters of water flow inertia through time series analysis methods. Next, these characteristic parameters are used to construct a mathematical model to calculate the inertia decay coefficient of each branch pipe during the flow regulation process. The specific operation is to simulate the actual response situations of each pipe under different flow regulation scenarios, compare the difference between the actual flow change and the expected flow change, so as to obtain the specific value of the inertia decay coefficient. For example, in some relatively long and large-diameter branch pipes, due to the large water flow inertia, the actual response speed after flow regulation is slow, resulting in a high inertia decay coefficient; while in relatively short and small-diameter pipes, the water flow inertia is relatively small, and the inertia decay coefficient is relatively low.

[0095] Step 205: Dynamically adjust the initial path sorting result according to the user-end demand deviation degree coefficient to generate a multi-objective water transmission path optimization scheme.

[0096] In this step, the multi-objective water conveyance path optimization scheme refers to the optimal water supply path planning obtained by comprehensively considering various factors (such as the urgency of water source replenishment, the deviation degree of user water demand, water flow inertia, etc.).

[0097] In this embodiment, first, the deviation degree coefficient of the user end demand is combined with the relevant parameters of each branch pipeline (such as the hydraulic coupling degree, inertia attenuation coefficient, etc.) to evaluate whether the current path sorting can effectively meet the actual water use demand of users. For example, during some peak water use periods or special events, the water use volume of users in a specific area may increase significantly, resulting in an increase in the deviation degree coefficient. At this time, the system will re-evaluate and adjust the priority and flow distribution ratio of the branch pipelines related to the affected area according to these coefficients. Next, the system uses an optimization algorithm (such as a genetic algorithm or a simulated annealing algorithm) to globally optimize the adjusted path sorting to ensure that the entire water supply network can achieve the best resource utilization rate and the lowest energy consumption on the premise of meeting the needs of all users. The finally generated multi-objective water conveyance path optimization scheme not only considers short-term demand changes but also takes into account long-term water supply stability and system sustainability.

[0098] To solve the problem of rigid path sorting caused by insufficient dynamic adjustment mechanism in traditional water conveyance path optimization schemes and to improve the response accuracy of the system to real-time demand changes and the utilization rate of redundant resources, as another embodiment, according to what is described in

[0098] , the initial path sorting result is dynamically adjusted according to the deviation degree coefficient of the user end demand to generate a multi-objective water conveyance path optimization scheme, including:

[0099] Step 301, generate a path sorting correction factor based on the ratio of the deviation degree coefficient of the user end demand to the inertia attenuation coefficient, and use the path sorting correction factor to scale the priority weights of the branch pipelines in the initial path sorting result proportionally to obtain the scaled priority weights;

[0100] In this step, the path sorting correction factor is a parameter used to dynamically adjust the priority weights of branch pipelines. It is calculated by combining the deviation degree coefficient of the user end demand and the inertia attenuation coefficient, reflecting the balance relationship between the urgency of each branch pipeline in meeting user needs and the adjustment difficulty caused by water flow inertia. A higher path sorting correction factor means that this pipeline needs a higher priority adjustment to meet the actual water use demand of users.

[0101] In this embodiment, first, the demand deviation coefficient of each branch pipeline is divided by its inertia decay coefficient to obtain a value reflecting the relationship between the demand urgency and the adjustment difficulty of the pipeline. Then, these path sorting correction factors are used to proportionally scale the priority weights of the branch pipelines in the initial path sorting result. For example, if a branch pipeline has a high demand deviation coefficient and a low inertia decay coefficient, its path sorting correction factor will be large, which means that this pipeline should be given a higher priority weight and thus be assigned a higher importance in the scheduling strategy. Conversely, if a pipeline has a high inertia decay coefficient (i.e., large water flow inertia), even if its demand deviation coefficient is also high, its path sorting correction factor will be relatively small to avoid over-adjustment resulting in system instability. Finally, through this dynamic adjustment method, a set of optimized and scaled priority weights is generated to ensure that the entire water supply network can meet user demands while maintaining system stability and efficient operation.

[0102] Step 302, update the flow distribution ratio of each branch pipeline according to the scaled priority weights to obtain the updated flow distribution ratio, generate a set of flow distribution ratios based on the updated flow distribution ratio, and calculate the change rate of the node pressure variance of the water transmission pipeline network in combination with the pressure balance degree index;

[0103] In this step, the flow distribution ratio refers to the share of each branch pipeline in its total water supply, reflecting the specific arrangement of the system for the water supply of each branch pipeline; the set of flow distribution ratios is the set of flow distribution ratios of all branch pipelines, used to guide the actual operation and scheduling of the entire water supply network; the change rate of the node pressure variance of the water transmission pipeline network is an index measuring the pressure fluctuation of each node in the water supply network, reflecting the stability of the system pressure distribution by calculating the change of the standard deviation or variance of the pressure values of each node. The lower the value, the more stable the pressure distribution.

[0104] In this embodiment, first, the flow ratio to be allocated for each branch pipeline is recalculated according to the new priority weight of each branch pipeline. For example, if the priority weight of a certain branch pipeline is significantly increased, its flow distribution ratio will also increase accordingly to ensure that the demand in this area can be met. Then, the new flow distribution ratios of all branch pipelines are summarized into a set of flow distribution ratios as the basis for subsequent scheduling operations. Then, based on this new set of flow distribution ratios, in combination with the pressure balance degree index of the water transmission pipeline network, the change rate of the pressure variance of each node is calculated, usually involving the use of historical pressure data and real-time monitoring data, and evaluating the impact of different flow distribution schemes on the node pressure of the pipeline network through statistical analysis methods (such as standard deviation or variance calculation). For example, at some key nodes, if it is found that the pressure fluctuation is large, the flow distribution ratio needs to be further adjusted to reduce the change rate of the pressure variance of these nodes, thereby improving the stability and reliability of the entire system.

[0105] Step 303: According to the water flow inertia factor in the dynamic load data of the water supply pipeline network, based on the proportion change range of each branch pipeline concentrated by the dynamically adjusted flow distribution ratio, calculate the path redundancy attenuation coefficient of the corresponding branch pipeline.

[0106] In this step, the water flow inertia factor is a parameter that describes the ability of water flow in a pipeline to resist changes due to changes in its mass and velocity, reflecting the response characteristics of water flow when facing sudden changes in flow rate or pressure; the proportion change range refers to the degree of difference between the flow distribution ratio of each branch pipeline after dynamic adjustment and its original flow distribution ratio, indicating the degree of change in the flow distribution strategy; the path redundancy attenuation coefficient is an index that measures the degree of weakening of the system redundancy ability after the flow distribution of a branch pipeline changes, reflecting the degree of decline in the system's ability to handle emergencies due to the reallocation of flow.

[0107] In this embodiment, first, by comparing the new and old flow distribution ratios of each branch pipeline, the proportion change range is obtained. Next, the proportion change range is combined with the water flow inertia factor, and a formula or model (such as an empirical formula or a machine learning model) is used to estimate the path redundancy attenuation coefficient of each branch pipeline. For example, in some relatively long and large-diameter branch pipelines, due to the large water flow inertia, even if the proportion change range is small, it may result in a relatively high path redundancy attenuation coefficient; while in relatively short and small-diameter pipelines, the water flow inertia is relatively small, and the path redundancy attenuation coefficient is relatively low. Through this series of calculations, the redundancy ability of the system under the new flow distribution scheme and the ability to handle emergencies can be evaluated, providing a basis for subsequent optimization.

[0108] Step 304: When the pressure variance change rate does not reach the preset optimization threshold and the path redundancy attenuation coefficient is higher than the preset redundancy critical value, use the product of the reciprocal of the water source replenishment urgency coefficient and the pressure variance change rate as the correction factor adjustment amount.

[0109] In this step, the correction factor adjustment amount is a dynamic parameter used to further adjust and optimize the water supply system scheduling strategy, calculated based on the water source replenishment urgency coefficient, the pressure variance change rate, and the path redundancy attenuation coefficient.

[0110] In this embodiment, first, it is checked whether the change rate of pressure variance has reached a preset optimization threshold, and at the same time, it is evaluated whether the path redundancy attenuation coefficient exceeds a preset redundancy critical value. If it is found that the change rate of pressure variance fails to meet the optimization standard (i.e., the system pressure fluctuates greatly), and the path redundancy attenuation coefficient is higher than the set safety limit (i.e., the system redundancy capacity is insufficient), further adjustment is required. At this time, the system will calculate the correction factor adjustment amount, take the reciprocal of the water source replenishment urgency coefficient, and multiply it by the change rate of pressure variance to obtain the correction factor adjustment amount. For example, for those regions with a low water source replenishment urgency but a high change rate of pressure variance, the correction factor adjustment amount will be relatively high, indicating that more adjustments are needed to reduce pressure fluctuations and enhance the system redundancy capacity; on the contrary, for regions with a high water source replenishment urgency but a low change rate of pressure variance, the correction factor adjustment amount is small, indicating that the pressure conditions in these regions are relatively ideal and do not require significant adjustments.

[0111] Step 305: Update the path sorting correction factor based on the correction factor adjustment amount, and re-execute the operations of scaling the priority weights, updating the flow distribution ratio, calculating the change rate of pressure variance, and the path redundancy attenuation coefficient until the change rate of pressure variance is less than the preset optimization threshold and the path redundancy attenuation coefficient is lower than the preset redundancy critical value, and generate a multi-objective water conveyance path optimization scheme.

[0112] In this embodiment, first, the correction factor adjustment amount is applied to the calculation formula of the path sorting correction factor to adjust the priority weights of each branch pipeline. Then, the operation of scaling the priority weights is re-executed using the updated path sorting correction factor to obtain a new priority weight distribution. Then, based on these new priority weights, the flow distribution ratio of each branch pipeline is updated again to form a new set of flow distribution ratios. Next, using this new set of flow distribution ratios and combining with the pressure balance index of the water conveyance network, the change rate of pressure variance and the path redundancy attenuation coefficient of each node are recalculated. If the calculation results show that the change rate of pressure variance is still higher than the preset optimization threshold or the path redundancy attenuation coefficient is still higher than the preset redundancy critical value, the above process is repeated to continue adjusting the path sorting correction factor and the priority weights until all set criteria are met. Finally, when the system reaches the preset optimization conditions, a stable multi-objective water conveyance path optimization scheme is generated.

[0113] To address the deficiencies of traditional dynamic balance strategies in cross-time-scale resource matching and to improve the coordination between water source replenishment and user demand in different time dimensions such as seasons and daily time periods, in some embodiments, according to what is described in step 102, a dynamic balance strategy is generated based on the correlation between the precipitation gradient change trend of the precipitation distribution characteristic data and the consumption rate in the user-side water demand fluctuation pattern data, including:

[0114] Step 401: Based on the precipitation gradient change trend, divide the precipitation recharge time intervals of the target area within a preset time period, and calculate the cumulative precipitation deficit of each precipitation recharge time interval to generate a water source recharge urgency index at the seasonal scale.

[0115] In this step, the precipitation recharge time interval refers to dividing a year or a specific time period into several time periods with similar precipitation change patterns according to the change trend and periodic characteristics of precipitation. These intervals help to identify the replenishment of water resources by precipitation in different seasons or periods. The cumulative precipitation deficit is an indicator that measures the difference between the actual precipitation and the ideal or expected precipitation within a specific time period, reflecting the degree of water resource shortage, and represents the total amount of water that fails to meet the demand through natural precipitation within a precipitation recharge time interval.

[0116] In this embodiment, first analyze the historical precipitation data of the target area to determine its precipitation gradient change trend. For example, by using statistical methods or time series analysis techniques, the main precipitation peak and trough periods in a year can be identified. Next, based on this trend information, divide the entire year into several precipitation recharge time intervals (such as spring, summer, autumn, and winter). For each precipitation recharge time interval, calculate its actual precipitation, compare it with the ideal precipitation required to meet the user's demand within this interval to obtain the precipitation deficit. Then, accumulate these deficit values to get the cumulative precipitation deficit of this precipitation recharge time interval. Use historical meteorological data, water demand prediction models, and geographic information system (GIS) tools to accurately calculate the precipitation deficit of each interval, and finally generate a seasonal scale report containing the cumulative precipitation deficit of each precipitation recharge time interval to guide the formulation of subsequent water resource management and scheduling strategies.

[0117] Step 402: According to the user-side water demand fluctuation pattern data, extract the consumption rate correlation relationship of different daily time periods, and calculate the water demand deviation rate of each daily time period.

[0118] In this step, the consumption rate correlation relationship refers to the change trend and mutual connection of the user's water consumption in different time periods, describing the consumption speed and its change law of water resources by the user at different time periods of a day (such as morning, afternoon, evening). The water demand deviation rate is an indicator that measures the difference between the actual water consumption and the expected or standard water consumption within a specific time period, used to reflect the fluctuation of the user's water demand and its impact on the overall supply-demand balance.

[0119] In this embodiment, first, a data analysis tool (such as time series analysis or machine learning model) is used to identify the changing trends of the water usage patterns and consumption rates of users in different time periods. For example, by analyzing the historical water usage data provided by the intelligent water meter system, it can be found that the water consumption in certain time periods (such as from 7:00 to 9:00 in the morning and from 6:00 to 8:00 in the evening) is significantly higher than that in other time periods, forming obvious peak periods. Next, based on these consumption rate correlation relationships, the water demand deviation rate of each daily time period is calculated. The actual water consumption of each daily time period is compared with the expected water consumption of that time period to obtain the deviation degree. The expected water consumption can be determined by the historical average value or the prediction model. For example, during the high-temperature period in summer, due to reasons such as residents increasing the shower frequency, the actual water consumption in the evening peak period may be 20% higher than expected. At this time, the water demand deviation rate of this time period is +20%. In the non-peak period, if the actual water consumption is lower than expected, the deviation rate is negative. Through this series of calculations, the fluctuation of water usage demand in different daily time periods can be comprehensively understood, providing a basis for formulating accurate water supply scheduling strategies in the future.

[0120] Step 403: Spatially and temporally couple the water source replenishment urgency index with the water demand deviation rate to generate dynamic constraint conditions across time scales.

[0121] In this step, spatial-temporal coupling refers to comprehensively analyzing data in different time dimensions (such as daily time periods, seasons) and spatial dimensions (such as different regions or nodes) to reveal the interaction laws of the water resource supply-demand relationship in time and space. The dynamic constraint conditions are a set of restrictive rules or parameters generated based on the results of the spatial-temporal coupling analysis, which are used to guide the formulation of water resource scheduling strategies.

[0122] In this embodiment, first, the water source replenishment urgency index calculated previously is integrated with the water demand deviation rate of each daily time period to form a spatial-temporal coupling model. The specific operation is to combine the cumulative precipitation gap at the seasonal scale and the water usage demand fluctuation pattern at the daily time period using mathematical modeling methods (such as linear programming or machine learning algorithms). For example, within a certain season, if the water source replenishment urgency index of a certain region is relatively high (that is, the insufficient precipitation leads to greater water supply pressure), and the water demand deviation rates of some daily time periods are also significantly high, then these time periods will become the key objects of attention. Next, based on these spatial-temporal coupling data, dynamic constraint conditions across time scales are generated. These constraint conditions usually include the upper and lower limit requirements for the flow distribution ratio within a specific time period, the pressure fluctuation suppression threshold of key nodes, and the minimum standard of redundancy capacity, etc. For example, during the high-temperature period in summer, for the peak water usage periods in the morning and evening, the system may set a higher water supply priority and limit the flow distribution ratio in the non-peak period to ensure that the demand in the peak period is met. Through this series of operations, the system can dynamically adjust the scheduling strategy to adapt to the supply-demand changes on different time scales.

[0123] Step 404: Based on the coupling constraint conditions, establish an iterative optimization model, and use the iterative optimization model to minimize the weighted sum of the total water source allocation within the precipitation recharge time interval and the deviation rate of water demand in each daily time period, so as to generate a dynamic balance strategy that meets the cross-time-scale resource matching.

[0124] In this step, the iterative optimization model is a mathematical model that gradually approaches the optimal solution through repeated iterative calculations. It is used to find the best solution in a complex multi-variable system. This model can handle multiple constraint conditions and continuously adjust parameters according to feedback to achieve the optimization goal. The minimization of the weighted sum of the deviation rates of water demand in each daily time period means that by weighted summing the deviation rates of water demand in different daily time periods and taking it as part of the optimization objective function, the total deviation amount is made as small as possible. Among them, the weighting coefficient here can be adjusted according to the actual situation (such as the importance of peak periods or the degree of water supply tension) to ensure that the needs of key periods are given priority.

[0125] In this embodiment, first, based on the spatio-temporal coupling constraint conditions generated in the early stage, an iterative optimization model is established. The specific operation is to select a suitable optimization algorithm (such as linear programming, genetic algorithm or simulated annealing algorithm), and set the weighted sum of the total water source allocation within each precipitation recharge time interval and the deviation rate of water demand in each daily time period as the optimization goal. For example, during the high-temperature period in summer, since the water demand in some periods increases significantly, the deviation rate of water demand in these periods is relatively high, so higher weights need to be given to ensure that their needs are given priority. Next, the model will perform multiple iterative calculations according to the set objective function and constraint conditions. Each iteration will adjust the water source allocation plan and evaluate its impact on the overall supply-demand balance. In this way, an optimal water source allocation plan that can not only meet the water use needs of each period but also minimize the overall deviation amount is gradually found. The finally generated dynamic balance strategy not only considers the seasonal precipitation changes but also takes into account the fluctuations in water use demand in different daily time periods, achieving the best cross-time-scale resource matching.

[0126] To solve the complexity problem of multi-time-scale data integration and constraint condition generation, and to improve the spatio-temporal correlation of dynamic constraint conditions and the robustness of the optimization model, as another embodiment, according to what is described in step 403, the water source recharge urgency index and the water demand deviation rate are spatio-temporally coupled to generate cross-time-scale dynamic constraint conditions, including:

[0127] Step 501: Calculate the time decay factor according to the number of days between preset time periods in the precipitation recharge time interval, and multiply the time decay factor by the water source recharge urgency index to generate the basic weight of the recharge priority distributed by daily granularity.

[0128] In this step, the time decay factor is a coefficient that reflects the gradual weakening of the impact of a specific event (such as precipitation) on the current state over time. It is usually calculated based on the number of days elapsed and is used to quantify the influence of precipitation events within a certain past period on the current water supply system. The basic weight of the recharge priority refers to a basic value calculated based on the time decay factor and the water source recharge urgency index, which is used to measure the priority of water resource replenishment on different dates or time periods. The daily granularity distribution means that these weight values are calculated and allocated on a daily basis, ensuring that there is a corresponding weight value for each day to guide resource scheduling.

[0129] In this embodiment, first, determine each preset time period within the precipitation recharge time interval and calculate the number of days between these time periods. For example, if a precipitation recharge time interval is divided into four seasons: spring, summer, autumn, and winter, there may be several days of interval between each season. Then, calculate the time decay factor based on these number of days using an exponential decay function or other suitable mathematical models to reflect the trend of the influence of earlier time periods decreasing over time. Next, multiply the calculated time decay factor by the water source recharge urgency index corresponding to each time period to generate the basic weight of the recharge priority distributed by daily granularity. For example, during the summer, if it has been five days since the nearest effective precipitation on a certain day, calculate the time decay factor for that day according to the set time decay formula and multiply it by the water source recharge urgency index of that day to obtain the basic weight of the recharge priority for that day. This process needs to be repeated for each day within the entire precipitation recharge time interval to form a complete sequence of basic weights distributed by daily granularity.

[0130] Step 502: Calculate the peak-valley compensation intensity for each daily time period based on the absolute value of the difference between the peak period deviation rate and the valley period deviation rate in the water demand deviation rate, in combination with the water use smoothing coefficient of the same type of daily time period within the historical irrigation cycle.

[0131] In this step, the peak period deviation rate refers to the ratio of the actual water consumption during peak water use periods (such as in the morning and evening) to the expected or standard water consumption, reflecting the fluctuations in water demand during these periods; the valley period deviation rate refers to the ratio of the actual water consumption during low valley water use periods (such as late at night) to the expected or standard water consumption, showing the changing trend of water consumption during off-peak periods; the absolute value of the difference is the absolute value of the difference between these two deviation rates, used to measure the imbalance degree of water demand between peak and valley periods; the historical irrigation cycle refers to the regular water supply cycle for agricultural or other uses over a past period, and its data can be used to analyze water use patterns and predict future demands; the water use smoothing coefficient is a parameter reflecting the smoothness of water demand changes over time, usually obtained through historical data analysis and used to smooth the fluctuations in water consumption; the peak-valley compensation intensity refers to the degree to which the water supply volume or priority needs to be adjusted by the system to balance the water demand differences between peak and valley periods, aiming to reduce the pressure during the peak water use period and make full use of resources during the low valley period.

[0132] In this embodiment, first, calculate the peak period deviation rate and valley period deviation rate for each daily period, and find the absolute value of their difference. Extract the actual water consumption of different daily periods from the historical water use data obtained from the intelligent water meter system, then compare it with the expected water consumption to obtain their respective deviation rates. Next, use these deviation rates to calculate the absolute value of the difference between peak and valley periods to quantify the imbalance of water demand between peak and valley periods. Then, combine the water use smoothing coefficient of the same type of daily period within the historical irrigation cycle to further refine the water use characteristics of each daily period. For example, during the high-temperature period in summer, due to reasons such as residents increasing the shower frequency, the actual water consumption during the evening peak period may be 20% higher than expected, while the actual water consumption during the late-night low valley period is 10% lower than expected. Based on this, the system calculates that the deviation rates of these two periods are +20% and -10% respectively, and the absolute value of their difference is 30%. Then, the system refers to the data of the historical irrigation cycle, especially the water use patterns under similar climate conditions, calculates the corresponding water use smoothing coefficient. Finally, combine the absolute value of the difference with the water use smoothing coefficient to calculate the peak-valley compensation intensity for each daily period, which helps to identify the periods that need special attention and determine how to balance the demand differences between peak and valley periods by adjusting the water supply strategy. For example, the system may increase the water supply volume during the peak period or take measures to encourage users to use water during the low valley period, thereby optimizing the overall water supply efficiency.

[0133] Step 503: Align the replenishment priority base weight with the peak-valley compensation intensity according to daily periods to generate a spatio-temporal correlation matrix;

[0134] In this step, the spatio-temporal correlation matrix is a mathematical tool for integrating data in the time and space dimensions. It forms a comprehensive matrix that can reflect the supply-demand relationship of the water supply system in terms of time (daily time periods) and space (different regions or nodes) by aligning and combining the basic weights of recharge priorities and peak-valley compensation intensities for different daily time periods. Aligning by daily time periods means precisely matching different indicators or parameters according to specific time periods of each day (such as morning, afternoon, evening, etc.) to ensure the consistency and accuracy of the time dimension in data analysis.

[0135] In this embodiment, first, the basic weights of recharge priorities and peak-valley compensation intensities calculated previously are respectively sorted into data sequences distributed by daily time periods. According to each daily time period (such as morning peak, evening peak, low valley period, etc.), the basic weights of recharge priorities and peak-valley compensation intensities for the corresponding time periods are matched one by one. For example, during the peak period from 7 pm to 9 pm in summer evenings, the system will extract the basic weight of recharge priority (calculated based on the time decay factor and the urgency index of water source recharge) and the peak-valley compensation intensity (calculated based on the absolute value of the difference in peak-valley deviation rate and the horizontal sliding coefficient) for this period. Then, these data are arranged one by one according to the daily time periods to form a preliminary two-dimensional matrix. Next, this matrix is extended to each day within the entire precipitation recharge time interval and combined with the spatial characteristics of different regions to generate a complete spatio-temporal correlation matrix, which requires the use of a Geographic Information System (GIS) tool or data analysis software to ensure the accuracy of the spatial distribution and time alignment of the data. The finally generated spatio-temporal correlation matrix not only reflects the changing trend of water supply demand within different daily time periods but also reveals the water use differences among regions and their impact on the overall water supply system.

[0136] Step 504: Calculate the pressure fluctuation suppression coefficient for each branch pipeline according to the node pressure balance degree index in the dynamic load data of the water supply pipeline network, and embed the pressure fluctuation suppression coefficient as a weight into the column vector of the spatio-temporal correlation matrix to obtain a weighted spatio-temporal correlation matrix.

[0137] In this step, the node pressure balance degree index is an indicator for measuring the evenness of pressure distribution among various nodes in the water supply network. The higher the value, the more balanced the pressure distribution. The operating state of the entire pipeline network system is evaluated by analyzing the pressure data of each node. The pressure fluctuation suppression coefficient is a parameter used to quantify the ability of each branch pipeline to control pressure fluctuations during the adjustment process, reflecting the ability of the pipeline system to maintain a stable pressure in response to flow changes. A higher pressure fluctuation suppression coefficient means that the pipeline can better suppress pressure fluctuations. The weighted spatio-temporal correlation matrix is an extended matrix obtained by further adding the pressure fluctuation suppression coefficient as a weight based on the original spatio-temporal correlation matrix, aiming to comprehensively consider the supply-demand relationship in the time and space dimensions and the pressure stability of the pipeline network.

[0138] In this embodiment, first, the pressure balance degree index of each node is extracted from the dynamic load data of the water supply network, and the pressure fluctuation suppression coefficient of each branch pipeline is calculated based on these indices. Statistical methods or models (such as time series analysis, machine learning algorithms) are used to process the historical pressure data to determine the pressure fluctuation situation and its suppression ability of each branch pipeline in different time periods. For example, during the high-temperature period in summer, due to the increase in water demand, the pressure fluctuations at some key nodes are relatively large. The system will calculate the pressure balance degree index based on the historical pressure data of these nodes, and combine the flow distribution and scheduling strategies to deduce the pressure fluctuation suppression coefficient of each branch pipeline; then, these pressure fluctuation suppression coefficients are embedded as weights into the column vectors of the previously generated spatio-temporal correlation matrix, adding a new weight value to each column (representing a specific daily time period), and this weight value reflects the pressure fluctuation suppression ability of the corresponding branch pipeline. Finally, a weighted spatio-temporal correlation matrix containing time, space, and pressure fluctuation suppression information is generated. This matrix not only shows the demand differences in different daily time periods and regions, but also reveals the contributions of each branch pipeline in maintaining the pressure stability of the system.

[0139] Step 505: Perform principal component analysis on the weighted spatio-temporal correlation matrix, extract the eigenvectors that satisfy the coupling relationship between the water source replenishment priority and the peak-valley periods of water use at the user end, and generate dynamic constraint conditions across time scales;

[0140] In this step, principal component analysis is a statistical method used to transform a set of potentially correlated variables into a set of linearly uncorrelated variables (called principal components) to reduce the data dimension and extract the main features. Through PCA, the most important patterns and trends can be identified from the complex weighted spatio-temporal correlation matrix; the water source replenishment priority refers to the water supply priority order in different regions or time periods determined according to factors such as the urgency of water source supply and the deviation rate of water demand; the coupling relationship between the peak-valley periods of water use at the user end refers to the water demand difference between the peak water use periods (such as morning and evening) and the valley water use periods (such as late at night) and its coordination relationship with the water supply strategy.

[0141] In this embodiment, first, use the PCA algorithm package in statistical software or programming languages (such as the scikit-learn library in Python) to perform dimensionality reduction on the matrix. The PCA process will calculate the covariance matrix and find its eigenvalues and eigenvectors, thereby extracting the most important principal components. These principal components represent the key information and patterns in the original data. Next, based on the extracted eigenvectors, analyze which components can best reflect the coupling relationship between the water source recharge priority and the peak and valley periods of water use at the user end. For example, during the high-temperature period in summer, due to the increased water demand of residents, the peak and valley differences in water use in certain daily periods are significant. The system can identify those eigenvectors that have the greatest impact on the overall supply-demand balance through PCA. Then, combine these eigenvectors to generate dynamic constraint conditions across time scales. These constraint conditions can include adjustment suggestions for the water supply priority within different daily periods, flow allocation limits within specific periods, and target values for suppressing node pressure fluctuations. For example, if a certain eigenvector shows that the water demand during the peak period in the evening in summer is significantly higher than expected, the system may set a higher water supply priority and limit the water supply volume during other low-demand periods to ensure that the demand during the peak period is met.

[0142] To solve the problems of spatio-temporal alignment and parameter matching during the conversion of the multi-objective water conveyance path optimization scheme into actual scheduling instructions, and to improve the execution accuracy of the pump station valve control instructions and the stability of the pipe network pressure, in some embodiments, according to step 104, convert the multi-objective water conveyance path optimization scheme into a set of scheduling instructions for controlling the pump station valves in the geospatial topology, including:

[0143] Step 601, generate an initial pump station valve opening and closing timing parameter set based on the branch pipeline flow allocation ratio and the path redundancy attenuation coefficient in the multi-objective water conveyance path optimization scheme;

[0144] In this step, the branch pipeline flow allocation ratio refers to the share of each branch pipeline in its total water supply, reflecting the specific arrangement of the water supply volume for each branch pipeline in the system. This ratio ensures that different regions or user groups can obtain the required water volume while maintaining the balance and stability of the entire water supply network. The initial pump station valve opening and closing timing parameter set is a data set containing the operation schedules of the pump stations and valves, used to guide when the pump stations start or stop supplying water and when the valves open or close to achieve optimal flow allocation and path selection. This parameter set takes into account the flow requirements of the branch pipelines, the path redundancy capabilities, and their dynamic changes.

[0145] In this embodiment, first, an optimization algorithm (such as linear programming or genetic algorithm) is used to calculate the best flow allocation scheme that meets the system requirements. For example, during the high-temperature period in summer, some key branch pipelines need a higher flow allocation ratio due to connecting to major residential areas. Next, based on these flow allocation ratios and the path redundancy attenuation coefficient, an initial set of timing parameters for opening and closing the pump station valves is generated. A simulation tool or the scheduling software of the control system is used to simulate the flow requirements during different time periods, and a detailed operation plan for the pump station and valves is formulated accordingly. For example, for a branch pipeline with high flow requirements, the system may set the pump station to start in advance during the morning peak period and maintain a high output until the end of the evening peak period; at the same time, the relevant valves will be adjusted in a timely manner according to the real-time requirements to ensure the smooth flow of water and the stability of pressure, which can not only meet the water use needs of users but also effectively avoid pipeline network failures caused by overload and improve the overall operation efficiency.

[0146] Step 602: According to the water flow inertia factor in the dynamic load data of the water transmission pipeline network and in combination with the water source replenishment priority weight in the dynamic balance strategy, calculate the inertia compensation coefficient during the flow regulation process of each branch pipeline.

[0147] In this step, the inertia compensation coefficient is a parameter that quantifies the influence brought by water flow inertia during the flow regulation process of the branch pipeline. It is determined by combining the water flow inertia factor and the water source replenishment priority weight to ensure that the system can quickly respond to demand changes and maintain stable operation.

[0148] In this embodiment, the water flow inertia factor is extracted from the dynamic load data of the water transmission pipeline network and in combination with the water source replenishment priority weight in the dynamic balance strategy, the inertia compensation coefficient of each branch pipeline is calculated, and the inertia effects that need to be additionally considered during the regulation process of different branch pipelines are identified. For example, in a long and large-diameter branch pipeline, due to the large water flow inertia, its inertia compensation coefficient is high; while in a short and small-diameter pipeline, the water flow inertia is relatively small, and the inertia compensation coefficient is low. Through these calculations, the system can fully consider the challenges brought by water flow inertia when formulating scheduling instructions, ensuring the efficient operation and stability of the water supply system.

[0149] Step 603: Based on each monitoring node of the water transmission pipeline network corresponding to the pressure equilibrium degree index, calculate the pressure fluctuation suppression threshold of each monitoring node.

[0150] In this step, the pressure fluctuation suppression threshold refers to the maximum allowable pressure fluctuation range set to maintain the pressure stability of each monitoring node in the pipeline network. It is calculated based on the pressure equilibrium degree index of each monitoring node and is used to guide the system to adjust the operations of the pump station and valves to avoid pipeline network failures caused by excessive pressure fluctuations.

[0151] In this embodiment, the pressure equilibrium degree index in the dynamic load data of the water supply pipeline network is used to analyze the historical pressure data of each monitoring node to determine the pressure fluctuation range during its normal operation. Then, based on this data, the pressure fluctuation suppression threshold of each monitoring node is calculated. For example, for monitoring nodes located in key areas or user-dense areas, the system will set a more stringent suppression threshold to ensure the stability of the water supply pressure.

[0152] Step 604: Align the initial pump station valve opening and closing timing parameter set with the inertia compensation coefficient in accordance with the pipeline connection relationship in the geospatial topology to generate a dynamic adjustment instruction template, where the flow rate adjustment parameter of each pump station valve in the dynamic adjustment instruction template is the product of the corresponding parameter in the opening and closing timing parameter set and the inertia compensation coefficient.

[0153] In this step, the dynamic adjustment instruction template is a data set integrating the operation plan of the pump station valves in the time and space dimensions. By aligning the initial pump station valve opening and closing timing parameter set with the inertia compensation coefficient in accordance with the pipeline connection relationship in the geospatial topology, a detailed flow rate adjustment plan is generated to optimize the operation efficiency of the entire water supply network.

[0154] In this embodiment, the initial pump station valve opening and closing timing parameter set is matched with the previously calculated inertia compensation coefficient to form a comprehensive operation plan. For example, for a branch pipeline connecting a major residential area, assuming its initial opening and closing timing parameter is 100 cubic meters per hour and the inertia compensation coefficient is 1.2, the new flow rate adjustment parameter is 120 cubic meters per hour. Then, these new parameters are arranged in chronological order and combined with the geospatial topology structure to generate a dynamic adjustment instruction template.

[0155] Step 605: Segmentally correct the dynamic adjustment instruction template according to the pressure fluctuation suppression threshold until the pressure fluctuation suppression thresholds of all monitoring nodes are within the preset safe range, for controlling the dispatching instruction set of the pump station valves in the geospatial topology, where the dispatching instruction set includes the opening and closing timing parameters, flow rate adjustment parameters of the pump station valves, and the corresponding pressure fluctuation suppression threshold alarm levels.

[0156] In this embodiment, the dynamic adjustment instruction template is segmented and corrected according to the previously calculated pressure fluctuation suppression threshold. For example, if the pressure fluctuation of a certain monitoring node approaches the upper limit of its suppression threshold, the system will automatically adjust the flow regulation parameters of the relevant branch pipelines, reduce the water supply volume or increase the water supply volume of other paths to reduce the pressure fluctuation of this node. At the same time, the system sets different alarm levels for each node to issue warnings in a timely manner when the pressure fluctuation exceeds the safe range. Through a series of correction operations, the system ensures that the pressure fluctuation suppression thresholds of all monitoring nodes are within the safe range, improving the stability and reliability of the entire water supply network. The finally generated scheduling instruction set not only includes detailed opening and closing timing parameters and flow regulation parameters, but also provides the corresponding pressure fluctuation suppression threshold alarm levels, providing comprehensive support for daily management and emergency response.

[0157] To solve the problems of adjustment amplitude control and convergence in the segmented correction of the pressure fluctuation suppression threshold, and to improve the accuracy and anti-interference ability of the scheduling instruction set, as another embodiment, according to step 605, the dynamic adjustment instruction template is segmented and corrected according to the pressure fluctuation suppression threshold until the pressure fluctuation suppression thresholds of all monitoring nodes are within the preset safe range, generating a scheduling instruction set for controlling the pump station valves in the geospatial topology, including:

[0158] Step 701, based on the flow regulation parameters in the dynamic adjustment instruction template, calculate the predicted pressure fluctuation values of each monitoring node;

[0159] In this step, the flow regulation parameters refer to the specific values generated according to the dynamic adjustment instruction template and used to guide the operation of the pump stations and valves. These parameters determine the water supply volume and water flow velocity of each branch pipeline in different time periods, thus directly affecting the pressure changes of each monitoring node in the pipe network. The predicted pressure fluctuation value is a value calculated based on the flow regulation parameters and reflecting the possible pressure change range that each monitoring node may experience in the future for a period of time, and is an important basis for evaluating the stability and safety of the water supply system.

[0160] In this embodiment, by utilizing the flow rate adjustment parameters in the dynamic adjustment instruction template and combining with the physical characteristics of the water supply pipeline network (such as pipeline length, diameter, material, etc.) and the water flow dynamics model, the predicted pressure fluctuation values of each monitoring node are calculated. Specifically, first, the flow rate adjustment parameters are input into the pipeline network simulation model to simulate the flow process of water in the pipeline and its influence on the node pressure. For example, for a branch pipeline with a relatively high flow rate adjustment parameter, the monitoring node at its end may experience relatively large pressure fluctuations; while for a pipeline with a relatively low flow rate adjustment parameter, the pressure fluctuations at its nodes are relatively small. Through this process, the pressure change conditions of each monitoring node at different time periods can be predicted, and the corresponding predicted pressure fluctuation values are generated, providing basic data support for subsequent pressure fluctuation analysis and optimization, helping the system to identify potential pressure risks in advance and take corresponding measures.

[0161] Step 702: Generate a pipeline adjustment sensitivity coefficient according to the ratio between the path redundancy attenuation coefficient and the water source replenishment priority weight.

[0162] In this step, the pipeline adjustment sensitivity coefficient is a parameter used to quantify the influence degree of a branch pipeline on the overall operation state of the system during the flow rate adjustment process. It is calculated by the ratio of the path redundancy attenuation coefficient to the water source replenishment priority weight, reflecting the flexibility and importance of the pipeline when facing adjustment requirements. A higher sensitivity coefficient indicates that the adjustment of this pipeline has a greater impact on the system stability or resource allocation and requires special attention.

[0163] In this embodiment, first, the path redundancy attenuation coefficient and the water source replenishment priority weight of each branch pipeline are extracted, and the ratio of the two is calculated to generate the pipeline adjustment sensitivity coefficient. For example, for a pipeline with a relatively high path redundancy attenuation coefficient but a relatively low water source replenishment priority weight, its sensitivity coefficient will be relatively high, indicating that this pipeline may face relatively large pressure or risks during the adjustment process; while for a pipeline with a relatively low path redundancy attenuation coefficient and a relatively high water source replenishment priority weight, its sensitivity coefficient is relatively low, indicating that the adjustment ability of this pipeline is strong and the influence is small. This process helps to identify which pipelines require more refined operation strategies during the flow rate adjustment, thereby optimizing the operation efficiency and stability of the entire water supply system. The finally generated pipeline adjustment sensitivity coefficient provides an important reference basis for subsequent dynamic adjustment and optimization.

[0164] Step 703: Establish a pressure fluctuation - flow rate adjustment feedback matrix, where the row vectors of the pressure fluctuation - flow rate adjustment feedback matrix represent the predicted pressure fluctuation values of each monitoring node, and the column vectors represent the pipeline adjustment sensitivity coefficients of the corresponding branch pipelines.

[0165] In this step, the pressure fluctuation flow regulation feedback matrix is a mathematical tool used to integrate the predicted pressure fluctuation values of each monitoring node in the water supply network and the pipeline regulation sensitivity coefficients of the corresponding branch pipelines. By combining the pressure fluctuation information of the monitoring nodes with the regulation sensitivity information of the branch pipelines, a matrix that comprehensively reflects the dynamic characteristics of the system is formed. This matrix provides data support for subsequent optimization of the scheduling strategy, helping the system to more accurately adjust the flow distribution and pressure control.

[0166] In this embodiment, first, the predicted pressure fluctuation values of each monitoring node calculated previously are used as row vectors, and the pipeline regulation sensitivity coefficients of the corresponding branch pipelines are used as column vectors to construct the pressure fluctuation flow regulation feedback matrix. Specifically, each row of the matrix represents the predicted pressure fluctuation value of a monitoring node, reflecting the pressure change situation of this node in a future period of time; each column corresponds to the pipeline regulation sensitivity coefficient of a branch pipeline, reflecting the influence degree of this pipeline on the system during the flow regulation process. For example, if the predicted pressure fluctuation value of a certain monitoring node is relatively high, and the pipeline regulation sensitivity coefficient of the branch pipeline connected to it is also relatively high, then this high-risk area will be shown in the matrix. In this way, the pressure fluctuations and pipeline regulation characteristics in the time and space dimensions are combined to form a comprehensive feedback matrix, which not only helps to identify the key nodes and pipelines in the system, but also provides clear data basis for formulating dynamic regulation strategies in the future, ensuring the stability and reliability of the water supply system.

[0167] Step 704: Perform eigenvalue decomposition on the pressure fluctuation flow regulation feedback matrix, and extract the eigenvectors whose eigenvalues are greater than the preset eigenvalue threshold as the key regulation channels.

[0168] In this step, the key regulation channels refer to the branch pipelines or node combinations that have a significant impact on the overall operating state of the water supply network. These channels are extracted from the pressure fluctuation flow regulation feedback matrix through eigenvalue decomposition, and their corresponding eigenvalues are greater than the preset eigenvalue threshold, indicating that they play an important role in flow regulation and pressure control. Identifying the key regulation channels helps to concentrate resources to optimize the scheduling strategies in these areas, thereby improving the overall efficiency and stability of the system.

[0169] In this embodiment, first, perform eigenvalue decomposition on the pressure fluctuation flow regulation feedback matrix to calculate its eigenvalues and eigenvectors. Specifically, use linear algebra methods (such as singular value decomposition or eigenvalue decomposition) to extract the main characteristic information in the matrix. Then, screen out the eigenvectors with larger eigenvalues according to a preset characteristic threshold. For example, if the eigenvalue corresponding to a certain branch pipeline is significantly higher than that of other pipelines, it indicates that the regulation effect of this pipeline in the system is more critical, so it is identified as a key regulation channel, which helps to focus on the pipeline or node combinations that have the greatest impact on pressure fluctuation and flow distribution, avoiding resource dispersion in secondary areas. Finally, the extracted key regulation channels provide clear targets for subsequent dynamic regulation, ensuring that the optimization work of the water supply system can be carried out efficiently and accurately.

[0170] Step 705: Determine the target branch pipelines and the regulation directions based on the element values of the eigenvectors of each key regulation channel, and update the flow regulation parameters according to the maximum allowable regulation amplitude of the inertia compensation coefficient.

[0171] In this step, the target branch pipelines refer to the specific branch pipelines in the key regulation channels identified through eigenvalue decomposition, and these pipelines have a significant impact on the overall pressure fluctuation and flow distribution of the system; the regulation direction refers to the operation direction of increasing or decreasing the flow rate for these target branch pipelines to achieve the purpose of optimizing the system performance.

[0172] In this embodiment, first, determine the target branch pipelines and their regulation directions according to the element values of the eigenvectors of each key regulation channel. Specifically, analyze the element values in the eigenvector. Larger positive values usually indicate the direction of increasing the flow rate, while larger negative values indicate the direction of decreasing the flow rate. For example, if the element value of a certain branch pipeline corresponding to the eigenvector is positive and large, it means that this pipeline needs to increase the water supply to relieve the pressure fluctuation in the current state; conversely, if the element value is negative and large, the water supply needs to be reduced. Next, according to the previously calculated inertia compensation coefficient, determine the maximum allowable regulation amplitude of each target branch pipeline and update the flow regulation parameters accordingly. For example, for a target branch pipeline with a high inertia compensation coefficient, its maximum allowable regulation amplitude may be small to avoid over-regulation caused by water flow inertia. In this way, the system can accurately adjust the flow distribution while ensuring stability, ensuring the efficient operation of the water supply system. Finally, these updated flow regulation parameters will be applied to actual operations to improve the response speed and regulation accuracy of the entire water supply network.

[0173] Step 706: Recalculate the pressure fluctuation suppression threshold of the associated monitoring nodes based on the updated flow regulation parameters.

[0174] In this embodiment, by using the updated flow regulation parameters and combining with the physical characteristics of the water supply pipeline network and the water flow dynamics model, the pressure fluctuation suppression thresholds of each associated monitoring node are recalculated. Specifically, first, the updated flow regulation parameters are input into the pipeline network simulation model to simulate the influence of water flow changes on the node pressures. For example, if the flow rate of a certain target branch pipeline is increased, the pressure of the monitoring node at its end may rise, so it is necessary to re-evaluate whether the pressure fluctuation range of this node is still within the safe range. Then, the pressure fluctuation suppression thresholds are adjusted according to the new flow distribution situation to ensure that they can reflect the actual operating state of the current system, help the system dynamically adapt to the changes brought by flow regulation, and provide more accurate target values for subsequent pressure control, thereby further improving the stability and reliability of the water supply network.

[0175] Step 707, if the recalculated pressure fluctuation suppression threshold does not reach the preset safe range, optimize the recalculated pressure fluctuation suppression threshold by reducing the adjustment amplitude in geometric progression according to the pipeline regulation sensitivity coefficient until the pressure fluctuation suppression thresholds of the monitoring nodes corresponding to all key regulation channels are within the preset safe range, so as to generate a scheduling instruction set for controlling the pump station valves in the geospatial topology.

[0176] In this embodiment, first, it is checked whether the recalculated pressure fluctuation suppression threshold meets the requirements of the preset safe range. If the pressure fluctuation suppression thresholds of some monitoring nodes still do not meet the standards, the flow regulation amplitude is reduced in geometric progression according to the pipeline regulation sensitivity coefficient. Specifically, for the branch pipelines with a higher sensitivity coefficient, their regulation amplitude will be preferentially reduced to reduce the influence on the overall pressure fluctuation of the system; while for the pipelines with a lower sensitivity coefficient, their regulation amplitude remains relatively large. For example, if the pressure fluctuation of the monitoring node of a certain key regulation channel is still too high, the system will gradually reduce the flow regulation amplitude of this channel and recalculate its corresponding pressure fluctuation suppression threshold until the pressure fluctuations of all nodes meet the safety requirements. This process gradually approaches the target through iterative optimization to ensure that the finally generated scheduling instruction set can effectively control the operations of the pump stations and valves in the geospatial topology. The final scheduling instruction set not only includes the specific opening and closing timings and flow regulation parameters of each pump station and valve, but also clarifies the pressure fluctuation suppression thresholds and their alarm levels of each monitoring node, providing comprehensive support for the daily management and emergency response of the water supply system.

[0177] Figure 2 The structural schematic diagram of a water resource scheduling optimization system based on artificial intelligence is provided for the embodiments of this application, as Figure 2 shown, this system includes:

[0178] An acquisition module 21 for acquiring multi-source heterogeneous data of water resources in a target area, where the multi-source heterogeneous data of water resources includes precipitation distribution characteristic data varying with seasons, dynamic load data of a water pipeline network based on geographical space topology, and user-side water demand fluctuation pattern data;

[0179] A generation module 22 for generating a dynamic balance strategy according to the precipitation gradient change trend of the precipitation distribution characteristic data and the consumption rate correlation relationship in the user-side water demand fluctuation pattern data;

[0180] An output module 23 for outputting an optimized multi-objective water conveyance path plan based on the node pressure distribution characteristics of the dynamic load data of the water pipeline network and in combination with the priority weights in the dynamic balance strategy;

[0181] A conversion module 24 for converting the optimized multi-objective water conveyance path plan into a scheduling instruction set for controlling the pump station valves in the geographical space topology

[0182] Figure 2 The described water resource scheduling optimization system based on artificial intelligence can execute Figure 1 The described water resource scheduling method based on artificial intelligence in the illustrated embodiment, and its implementation principle and technical effects will not be elaborated further. For the water resource scheduling optimization system based on artificial intelligence in the above embodiment, the specific manners in which each module and unit perform operations have been described in detail in the embodiment of the method, and will not be elaborated here.

[0183] In a possible design, Figure 2 The water resource scheduling system based on artificial intelligence in the illustrated embodiment can be implemented as a computing device, such as Figure 3 shown, and this computing device can include a storage component 31 and a processing component 32;

[0184] The storage component 31 stores one or more computer instructions, where the one or more computer instructions are called and executed by the processing component 32.

[0185] The processing component 32 is used for the Figure 1 described water resource scheduling optimization method in the above embodiment.

[0186] Among them, the processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above methods. Of course, the processing component may also be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components for executing the above methods.

[0187] The storage component 31 is configured to store various types of data to support the operation of the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0188] Of course, the computing device may also necessarily include other components, such as input / output interfaces, display components, communication components, etc.

[0189] The input / output interface provides an interface between the processing component and the peripheral interface module, and the above peripheral interface module may be an output device, an input device, etc.

[0190] The communication component is configured to facilitate communication between the computing device and other devices in a wired or wireless manner, etc.

[0191] Among them, the computing device may be a physical device or an elastic computing host provided by a cloud computing platform, etc. At this time, the computing device may refer to a cloud server, and the above processing component, storage component, etc. may be basic server resources leased or purchased from the cloud computing platform.

[0192] The embodiment of the present application also provides a computer storage medium storing a computer program, and when the computer program is executed by a computer, it can implement the above Figure 1 shown embodiment of an artificial intelligence-based water resource scheduling optimization method.

[0193] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the above-described systems, devices, and units can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0194] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0195] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution or the part that contributes to the prior art can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0196] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and are not intended to limit them. Although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features. These modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of each embodiment of the present application.

Claims

1. A water resources scheduling optimization method based on artificial intelligence, characterized in that: include: Acquire multi-source heterogeneous data of water resources in the target area, wherein the multi-source heterogeneous data of water resources includes seasonally varying precipitation distribution characteristic data, dynamic load data of water pipe network based on geographic spatial topology, and user-side water demand fluctuation pattern data; Generate a dynamic balance strategy based on the correlation between the precipitation gradient change trend of the precipitation distribution characteristic data and the consumption rate in the user-side water demand fluctuation pattern data; Based on the node pressure distribution characteristics of the water supply network dynamic load data and the priority weights in the dynamic balance strategy, a multi-objective water supply path optimization scheme is output; The multi-objective water delivery path optimization scheme is converted into a scheduling instruction set for controlling the valves of the pumping stations in the geospatial topology.

2. The method according to claim 1, characterized in that Based on the node pressure distribution characteristics in the dynamic load data of the water transmission network and combined with the priority weights in the dynamic balance strategy, a multi-objective water transmission path optimization scheme is generated, including: Based on the node pressure distribution characteristics in the dynamic load data of the water supply network, extract the pressure fluctuation rate and pressure balance index of each water supply network node in the continuous monitoring period; The priority weight in the dynamic balance strategy is split into a water supply urgency coefficient and a user-side demand deviation coefficient; The product of the pressure fluctuation rate and the water supply urgency coefficient is used as the hydraulic coupling degree of the water supply network node, and the branch pipelines in the geographic space topology are initially sorted according to the hydraulic coupling degree to generate an initial path sorting result; Based on the water flow inertia factor in the dynamic load data of the water supply network, calculating the inertia attenuation coefficient of each branch pipeline during the flow regulation process; The initial path sorting result is dynamically adjusted according to the user-side demand deviation coefficient to generate a multi-objective water delivery path optimization plan.

3. The method according to claim 2, characterized in that The initial path sorting result is dynamically adjusted according to the user-side demand deviation coefficient to generate a multi-objective water delivery path optimization plan, including: Generate a path sorting correction factor based on the ratio of the user-side demand deviation coefficient to the inertia attenuation coefficient, and use the path sorting correction factor to proportionally scale the priority weights of the branch pipelines in the initial path sorting result to obtain scaled priority weights; updating the flow distribution ratio of each branch pipeline according to the scaled priority weight to obtain an updated flow distribution ratio, generating a flow distribution ratio set based on the updated flow distribution ratio, and calculating the pressure variance change rate of the water supply network node in combination with the pressure balance index; According to the water flow inertia factor in the dynamic load data of the water pipe network, based on the proportional change amplitude of each branch pipe in the dynamically adjusted flow distribution ratio set, the path redundancy attenuation coefficient of the corresponding branch pipe is calculated; When the pressure variance change rate does not reach the preset optimization threshold and the path redundancy attenuation coefficient is higher than the preset redundancy critical value, the product of the inverse of the water source replenishment urgency coefficient and the pressure variance change rate is used as the correction factor adjustment amount; The path sorting correction factor is updated based on the correction factor adjustment amount, and the priority weight scaling, flow distribution ratio updating, pressure variance change rate and path redundancy attenuation coefficient are re-executed until the pressure variance change rate is less than a preset optimization threshold and the path redundancy attenuation coefficient is lower than a preset redundancy critical value, thereby generating a multi-objective water delivery path optimization plan.

4. The method according to claim 1, characterized in that: Generate a dynamic balance strategy based on the precipitation gradient change trend of the precipitation distribution characteristic data and the consumption rate correlation relationship in the user-side water demand fluctuation pattern data, including: Based on the precipitation gradient change trend, the target area is divided into precipitation recharge time intervals within a preset period, and the cumulative precipitation gap in each precipitation recharge time interval is calculated to generate a water source recharge urgency index on a seasonal scale; According to the user-side water demand fluctuation pattern data, the consumption rate correlation relationship of different daily time periods is extracted, and the water demand deviation rate of each daily time period is calculated; The water supply urgency index is coupled with the water demand deviation rate in time and space to generate dynamic constraint conditions across time scales; Based on the coupling constraints, an iterative optimization model is established. The iterative optimization model is used to constrain the weighted sum of the total amount of water source allocation within the precipitation recharge time interval and the water demand deviation rate in each daily period to minimize the dynamic balance strategy that satisfies cross-time scale resource matching.

5. The method according to claim 4, characterized in that The water supply urgency index is coupled with the water demand deviation rate in time and space to generate dynamic constraints across time scales, including: Calculate the time decay factor according to the number of days between the preset time periods in the precipitation recharge time interval, and multiply the time decay factor by the water source recharge urgency index to generate a recharge priority basic weight distributed by daily granularity; Based on the absolute value of the difference between the peak time deviation rate and the valley time deviation rate in the water demand deviation rate, combined with the water use smoothing coefficient of the same type of daily time period in the historical irrigation cycle, the peak-valley compensation intensity of each daily time period is calculated; Aligning the replenishment priority basic weight with the peak-valley compensation intensity by daily time period to generate a spatiotemporal correlation matrix; Calculate the pressure fluctuation suppression coefficient of each branch pipeline according to the node pressure balance index in the dynamic load data of the water supply network, and embed the pressure fluctuation suppression coefficient into the column vector of the spatiotemporal correlation matrix as a weight to obtain a weighted spatiotemporal correlation matrix; The weighted spatiotemporal correlation matrix is ​​subjected to principal component analysis to extract characteristic vectors that satisfy the coupling relationship between the water source replenishment priority and the peak and valley periods of water use at the user end, thereby generating dynamic constraint conditions across time scales.

6. The method according to claim 1, characterized in that Converting the multi-objective water delivery path optimization scheme into a scheduling instruction set for controlling the valves of the pumping station in the geospatial topology includes: Based on the branch pipeline flow distribution ratio and the path redundancy attenuation coefficient in the multi-objective water delivery path optimization scheme, an initial pump station valve opening and closing timing parameter set is generated; According to the water flow inertia factor in the dynamic load data of the water supply network, combined with the water supply priority weight in the dynamic balance strategy, the inertia compensation coefficient in the flow regulation process of each branch pipeline is calculated; Based on each monitoring node of the water pipe network corresponding to the pressure balance index, calculating the pressure fluctuation suppression threshold of each monitoring node; The initial pump station valve opening and closing timing parameter set and the inertia compensation coefficient are time-space aligned according to the pipeline connection relationship in the geographic space topology to generate a dynamic adjustment instruction template, wherein the flow adjustment parameter of each pump station valve in the dynamic adjustment instruction template is the product of the corresponding parameter in the opening and closing timing parameter set and the inertia compensation coefficient; The dynamic adjustment instruction template is modified in sections according to the pressure fluctuation suppression threshold until the pressure fluctuation suppression threshold of all monitoring nodes is in a preset safety interval, which is used to control the scheduling instruction set of the pump station valves in the geographic space topology, wherein the scheduling instruction set includes the opening and closing timing parameters of the pump station valves, flow regulation parameters and the corresponding pressure fluctuation suppression threshold alarm level.

7. The method according to claim 6, characterized in that The dynamic adjustment instruction template is modified in sections according to the pressure fluctuation suppression threshold until the pressure fluctuation suppression threshold of all monitoring nodes is within a preset safety interval, and a scheduling instruction set for controlling the pump station valve in the geographic space topology is generated, including: Calculating the predicted pressure fluctuation value of each monitoring node based on the flow adjustment parameter in the dynamic adjustment instruction template; Generate a pipeline regulation sensitivity coefficient according to the ratio between the path redundancy attenuation coefficient and the water source replenishment priority weight; Establishing a pressure fluctuation flow regulation feedback matrix, wherein the row vector of the pressure fluctuation flow regulation feedback matrix represents the predicted pressure fluctuation value of each monitoring node, and the column vector represents the pipeline regulation sensitivity coefficient of the corresponding branch pipeline; Performing eigendecomposition on the pressure fluctuation flow regulation feedback matrix, and extracting eigenvectors whose eigenvalues ​​are greater than a preset eigenthreshhold as key regulation channels; Determine the target branch pipeline and the adjustment direction based on the characteristic vector element value of each key adjustment channel, and update the flow adjustment parameter according to the maximum allowable adjustment range of the inertia compensation coefficient; Based on the updated flow regulation parameters, recalculate the pressure fluctuation suppression threshold of the associated monitoring node; If the recalculated pressure fluctuation suppression threshold does not reach the preset safety interval, the recalculated pressure fluctuation suppression threshold is optimized by reducing the adjustment range proportionally according to the pipeline adjustment sensitivity coefficient until the pressure fluctuation suppression threshold of the monitoring nodes corresponding to all key adjustment channels is within the preset safety interval, so as to generate a scheduling instruction set for controlling the pump station valves in the geographic spatial topology.

8. A water resources scheduling optimization system based on artificial intelligence, characterized in that: include: An acquisition module is used to acquire multi-source heterogeneous data of water resources in a target area, wherein the multi-source heterogeneous data of water resources include seasonal precipitation distribution characteristic data, dynamic load data of water pipe network based on geographic space topology, and user-side water demand fluctuation pattern data; A generation module, configured to generate a dynamic balance strategy according to a correlation between a precipitation gradient change trend of the precipitation distribution characteristic data and a consumption rate in the user-side water demand fluctuation pattern data; An output module, configured to output a multi-objective water transmission path optimization scheme based on the node pressure distribution characteristics of the water transmission network dynamic load data and in combination with the priority weights in the dynamic balance strategy; A conversion module is used to convert the multi-objective water delivery path optimization plan into a scheduling instruction set for controlling the pump station valves in the geographic space topology.

9. A computing device, characterized in that It comprises a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement an artificial intelligence-based water resource scheduling optimization method as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a computer, an artificial intelligence-based water resources scheduling optimization method as described in any one of claims 1 to 7 is implemented.

Images