Intelligent city water supply pipe network emergency supervision Internet of Things large model system and method

Abstract

The invention provides a smart city water supply pipe network emergency supervision Internet of Things large model system and method, and relates to the field of city water pipe monitoring, and the system comprises a government supervision management platform, a government supervision sensing network platform, a government supervision object platform, a water affair company sensing network platform, and a smart water affair equipment object platform. The government supervision and management platform is configured to determine time sequence flow corresponding to the multiple water pipe node groups based on water flow characteristics of the multiple water pipe nodes in the water supply network; based on the time sequence flow, determining water and soil loss coefficients corresponding to the one or more target areas; based on the water and soil loss coefficient, temporary control parameters are generated through a parameter generation model; and on the basis of the temporary control parameters, the valve opening degrees of the corresponding valves in the one or more target areas are controlled. The system can determine the control parameters of the adjusting device according to the water and soil loss coefficient of the target area, and realizes informatization and intelligence of urban water pipe emergency monitoring.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of urban water pipe monitoring, in particular to a smart city water supply pipe network emergency supervision Internet of Things large model system and method. BACKGROUND

[0002] Urban water supply pipe networks may have leakage due to aging corrosion, external damage, etc. Since the leakage may be hidden and the damage in a short period of time is low, it is easy to be ignored. In some areas, long-term continuous leakage will cause soil erosion, and then cause cavities in the soil under the ground to collapse, which may also cause traffic congestion, damage to urban lifeline related equipment, and other secondary disasters, posing a safety hazard.

[0003] Therefore, it is necessary to provide a smart city water supply pipe network emergency supervision Internet of Things large model system and method, which realizes real-time monitoring and intelligent control of urban water supply pipe networks, and reduces soil erosion and water resource waste. SUMMARY

[0004] The summary includes a smart city water supply pipe network emergency supervision Internet of Things large model system, the system comprising: a government supervision management platform, a government supervision sensing network platform, a government supervision object platform, a water company sensing network platform, and a smart water management equipment object platform, the government supervision object platform comprising a water company management platform; the smart water management equipment object platform comprising a monitoring device and at least one adjusting device; the government supervision management platform is configured to: determine time series flow corresponding to a plurality of groups of water pipe nodes based on water flow characteristics of the plurality of water pipe nodes in the water supply pipe network; determine a soil erosion coefficient corresponding to one or more target areas based on the time series flow; generate temporary control parameters through a parameter generation model based on the soil erosion coefficient, the parameter generation model being a machine learning model; and control the valve opening degree of a corresponding valve in the one or more target areas based on the temporary control parameters.

[0005] The summary includes a smart city water supply pipe network emergency supervision method, which is executed by a government supervision management platform in a smart city water supply pipe network emergency supervision Internet of Things large model system, the method comprising: determining time series flow corresponding to a plurality of groups of water pipe nodes based on water flow characteristics of the plurality of water pipe nodes in the water supply pipe network; determining a soil erosion coefficient corresponding to one or more target areas based on the time series flow; generating temporary control parameters through a parameter generation model based on the soil erosion coefficient, the parameter generation model being a machine learning model; and controlling the valve opening degree of a corresponding valve in the one or more target areas based on the temporary control parameters.

[0006] The summary of the invention comprises a computer readable storage medium, which stores computer instructions, and when the computer reads the computer instructions in the storage medium, the computer executes the smart city water supply network emergency supervision method.

[0007] The beneficial effects of the present invention include but are not limited to: (1) Through the smart city water supply network emergency supervision Internet of Things large model system, an information operation closed loop can be formed between each functional platform, and coordinated and regularly operated under the unified management of the government supervision and management platform, realizing the informatization and intelligence of city water pipe emergency monitoring. (2) By analyzing the time sequence flow of multiple water pipe node groups, the soil erosion coefficient of the target area is determined, and then the control parameters of the adjusting device are determined, realizing real-time monitoring and intelligent control of the water supply network, improving monitoring accuracy, response speed and operation efficiency, reducing soil erosion and water resource waste, and providing effective technical support for water pipe management of smart city. (3) By fusing time sequence flow and soil characteristics, combined with machine learning large model, the soil erosion coefficient can be more accurately and quickly determined, avoiding evaluation deviation caused by single index. BRIEF DESCRIPTION OF DRAWINGS

[0008] The present specification will be further illustrated in the form of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting, and in these embodiments, the same numbers represent the same structures, wherein: Figure 1 is a platform structure schematic diagram of a smart city water supply network emergency supervision Internet of Things large model system according to some embodiments of the present specification; Figure 2 is an exemplary flowchart of a smart city water supply network emergency supervision method according to some embodiments of the present specification; Figure 3 is an exemplary schematic diagram of a coefficient generation model according to some embodiments of the present specification; Figure 4 is an exemplary schematic diagram of generating temporary control parameters according to some embodiments of the present specification. DETAILED DESCRIPTION

[0009] In order to more clearly illustrate the technical solutions of the embodiments of the present specification, the drawings needed in the embodiment description will be briefly introduced below. Obviously, the drawings in the following description are only some examples or embodiments of the present specification, and for those skilled in the art, the present specification can be applied to other similar scenarios without creating creative labor. Unless it is obvious from the language environment or otherwise stated, the same reference numbers in the figures represent the same structures or operations.

[0010] It should be understood that the terms "system," "device," "unit," and / or "module" used herein are a method for distinguishing different components, elements, parts, portions, or assemblies at different levels. However, if other terms can achieve the same purpose, the terms may be replaced by other expressions.

[0011] Flowcharts are used throughout this specification to illustrate the operations performed by systems according to embodiments of this specification. It should be understood that preceding or following operations do not necessarily need to be performed in exact order. Instead, the steps may be processed in reverse order or simultaneously. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0012] Figure 1 It is a platform structure diagram of the smart city water supply network emergency supervision Internet of Things large model system shown in some embodiments of this specification.

[0013] In some embodiments, as Figure 1 As shown, the smart city water supply network emergency supervision Internet of Things large model system 100 can include a government supervision management platform 110, a government supervision sensor network platform 120, a government supervision object platform 130, a water company sensor network platform 140 and a smart water equipment object platform 150.

[0014] The government supervision and management platform refers to a platform for supervising and safely managing the water supply network. The water supply network can be a pipeline network used to transport and distribute water resources.

[0015] In some embodiments, the government regulatory management platform may be configured in a processor and / or server. The government regulatory management platform may include a database. A database is a database used to store regulatory data. For example, the database may be used to store relevant data on the water supply network, parameter generation models, and the like.

[0016] In some embodiments, the government regulatory management platform is configured to determine the time-series flow rates corresponding to multiple groups of water pipe nodes based on water flow characteristics at multiple water pipe nodes in the water supply network, and determine the soil and water loss coefficients corresponding to one or more target areas based on the time-series flow rates. The government regulatory management platform may also be configured to generate temporary control parameters based on the soil and water loss coefficients using a parameter generation model, and to control the valve openings of corresponding valves in one or more target areas based on the temporary control parameters.

[0017] The government-regulated sensor network platform refers to a functional platform for managing government sensor communications. In some embodiments, the government-regulated sensor network platform can be configured as a communication device and / or a gateway.

[0018] In some embodiments, the government regulation sensing network platform can interact with the government regulation management platform and the government regulation object platform.

[0019] The government regulation object platform refers to an information processing platform for regulating the objects related to the water supply network. In some embodiments, the government regulation object platform can include the water company management platform 131.

[0020] The water company management platform refers to a comprehensive management platform for water company information. In some embodiments, the water company management platform can be configured as a processor and / or a server, etc.

[0021] The water company sensing network platform refers to a comprehensive management platform for water company sensing information. In some embodiments, the water company sensing network platform can be configured as a communication device and / or a gateway, etc.

[0022] In some embodiments, the water company sensing network platform can interact with the water company management platform and the smart water equipment object platform.

[0023] The smart water equipment object platform refers to a functional platform for generating sensing information and executing control information. In some embodiments, the smart water equipment object platform at least includes monitoring devices and at least one regulating device arranged in the water supply network.

[0024] The monitoring device refers to a device for monitoring and recording the operating state of the water supply network. In some embodiments, the monitoring device can include a water flow sensor, etc. The water flow sensor can obtain the water flow characteristics of multiple water pipe nodes, etc. The monitoring device can be installed at any feasible position of the pipe corresponding to the water pipe node, such as the starting end of the pipe or inside the pipe, etc.

[0025] The regulating device refers to a device for controlling and regulating the water flow in the water supply network, etc. In some embodiments, the regulating device can include a valve, a water supply pump, a user device, etc. The user device refers to a device for controlling the water use at the user end. For example, a green irrigation system in a city, etc.

[0026] In some embodiments, the smart water equipment object platform further includes a detection robot. The detection robot refers to a machine for detecting the soil information at the location of the water pipe node group. For example, the detection robot can detect the soil characteristics (such as soil density, soil moisture content, and soil viscosity, etc.), etc.

[0027] For more information about the above-mentioned various platforms, please refer to Figures 2-4 and related descriptions.

[0028] In some embodiments of the present specification, through the smart city water supply pipe network emergency supervision Internet of Things large model system 100, an information operation closed loop can be formed between various functional platforms, and coordinated and regularly operated under the unified management of the government supervision and management platform, realizing the informatization and intelligence of city water pipe emergency monitoring.

[0029] It should be noted that the above description of the smart city water supply pipe network emergency supervision Internet of Things large model system and its platform is for convenience of description only, and cannot limit the present specification to the scope of the embodiments. It can be understood that for those skilled in the art, after understanding the principle of the system, various platforms can be combined or connected with other platforms to form a subsystem without departing from the principle.

[0030] Figure 2 is an exemplary flowchart of the smart city water supply pipe network emergency supervision method according to some embodiments of the present specification. As shown in Figure 2 The smart city water supply pipe network emergency supervision method flowchart 200 includes the following steps. In some embodiments, the smart city water supply pipe network emergency supervision method flowchart 200 can be executed by the government supervision and management platform 110.

[0031] Step 210, based on the water flow characteristics of the plurality of water pipe nodes in the water supply pipe network, determining the time sequence flow corresponding to the plurality of water pipe node groups.

[0032] Water pipe node refers to a section of water pipe in the water supply pipe network. In some embodiments, the plurality of water pipe nodes in the water supply pipe network can be pre-set by technical personnel based on historical experience and stored in a database.

[0033] Water flow characteristics refer to data related to water flow at the water pipe node. In some embodiments, the water flow characteristics can include the water flow of the water pipe node, etc. The water flow of the water pipe node can be collected and acquired by the water flow sensor in the monitoring device, etc. For the description of the monitoring device, see Figure 1 and related description.

[0034] Water pipe node group refers to a set consisting of one or more water pipe nodes. In some embodiments, the government supervision and management platform can divide one or more adjacent water pipe nodes on the same water flow path into a water pipe node group. Water flow path refers to the path of water flow in the water supply pipe network.

[0035] In some embodiments, the water flow path can be a path from the starting point (such as the water supply end, etc.) to the branch point (such as the pump station, etc.) or the terminal point (such as the water use end, etc.). The water supply pipe network includes a plurality of water flow paths.

[0036] In some embodiments, when the flow difference between multiple water pipe nodes on the same water flow path is large, the government regulatory management platform can perform gradient division on the multiple flows of the multiple water pipe nodes to obtain multiple flow gradients, and divide one or more water pipe nodes within the same flow gradient into a water pipe node group.

[0037] For example only, one water flow path includes four water pipe nodes. The water flows of the four water pipe nodes are 100 L / s, 80 L / s, 60 L / s, and 40 L / s, respectively. The government regulatory management platform can divide the water pipe nodes corresponding to 100 L / s and 80 L / s into a water pipe node group based on the flow gradient, and divide the water pipe nodes corresponding to 60 L / s and 40 L / s into a water pipe node group.

[0038] The time series flow refers to data representing the water flow of the water pipe node group at multiple time points. In some embodiments, the time series flow can be represented in a sequence form, and the sequence includes the water flow of the water pipe node group at each time point. One water pipe node group corresponds to one time series flow. The multiple time points can be pre-set based on historical experience, such as one day.

[0039] In some embodiments, the government regulatory management platform determines the time series flow corresponding to multiple water pipe node groups based on the water flow characteristics of the multiple water pipe nodes in the water supply pipe network. For example, the government regulatory management platform can calculate the average water flow of the multiple water pipe nodes in the water pipe node group at each time point, and group the average water flows at the multiple time points to obtain the time series flow corresponding to the water pipe node group.

[0040] At step 220, the water and soil loss coefficient corresponding to one or more target areas is determined based on the time series flow.

[0041] The target area refers to an area in the water supply pipe network that needs to be concerned. For example, the target area can include the water supply pipe network area corresponding to a community or a factory building. One target area can include one or more water pipe node groups. It should be noted that the water pipe nodes in one water pipe node group are in one target area.

[0042] The water and soil loss coefficient refers to data representing the water and soil loss condition of the target area. One target area corresponds to one water and soil loss coefficient.

[0043] In some embodiments, the government regulation management platform determines the soil erosion coefficient of one or more target areas based on the time series flow. For example, for a target area, the government regulation management platform can calculate the time series flow difference between two adjacent water pipe node groups in the target area, and determine the soil erosion coefficient of the target area based on the time series flow difference. If the target area includes more than two water pipe node groups, one target area can correspond to multiple soil erosion coefficients, and each soil erosion coefficient corresponds to a group of two adjacent water pipe node groups. The time series flow difference refers to a sequence of water flow difference values at each time point in the two time series flows.

[0044] In some embodiments, the government regulation management platform can regard two water pipe node groups as adjacent two water pipe node groups if the distance between the two water pipe node groups is less than a distance threshold. The distance threshold is pre-set based on historical experience. The distance between the two water pipe node groups can be represented by the straight-line distance between the physical centers of the two water pipe node groups. The physical center of a water pipe node group can include the middle point of the pipe.

[0045] In some embodiments, for each group of two adjacent water pipe node groups, the government regulation management platform can filter out the target time points at which the water flow difference value in the time series flow difference is greater than a first difference threshold based on the time series flow difference between the two water pipe node groups, and determine the soil erosion coefficient based on the target time points and the water flow difference values corresponding to the target time points. The first difference threshold can be pre-set based on historical experience.

[0046] For example, the soil erosion coefficient can be positively correlated with the number of target time points and the average of the water flow difference values corresponding to the target time points, and negatively correlated with the total number of time points corresponding to the time series flow difference. The government regulation management platform can determine the soil erosion coefficient by the following formula (1): S=k* (m / M) *g (1) where S represents the soil erosion coefficient, k represents the flow difference trend, m represents the number of target time points, M represents the total number of time points corresponding to the time series flow difference, and g represents the average of the water flow difference values corresponding to the target time points.

[0047] The flow difference trend refers to data representing the change of the water flow difference values corresponding to the target time points. In some embodiments, the government regulation management platform can obtain a linear function based on the target time points and the water flow difference values corresponding to the target time points, such as linear fitting, and take the slope of the obtained linear function as the flow difference trend. The horizontal axis of the linear function is the target time point, and the vertical axis is the water flow difference value corresponding to the target time point.

[0048] In step 230, the government regulation management platform generates a temporary control parameter based on the soil erosion coefficient and the parameter generation model.

[0049] In some embodiments, the parameter generation model (also referred to as a parameter generation large model) is a machine learning model. For example, the parameter generation model can include any one or a combination of a Graph Neural Network (GNN) model or a large model, etc., or other custom model structures, etc. The input of the parameter generation model includes one or more soil erosion coefficients corresponding to the target region, and the output of the parameter generation model includes the temporary control parameter.

[0050] The temporary control parameter refers to a parameter used to control the regulating device (such as a valve, etc.) in the water supply network. In some embodiments, the temporary control parameter includes the valve opening degree of the valve in the target region, etc. It can be understood that the temporary control parameter is used to control the regulating device for a period of time, and a new temporary control parameter can be determined subsequently.

[0051] In some embodiments, the opening degree of the valve corresponds to the water flow rate passing through the valve, and the larger the valve opening degree of the upstream water pipe node of the target region corresponds, the larger the water flow rate flowing into the target region. The larger the valve opening degree of the valve corresponding to the water pipe node group, the larger the water flow rate flowing into the water pipe node group.

[0052] In some embodiments, the temporary control parameter output by the parameter generation model can only include the valve opening degree of part of the valve in the target region, etc. The parameter generation model can determine one or more groups of adjacent water pipe node groups whose valve opening degree needs to be adjusted based on at least one soil erosion coefficient corresponding to the target region among the multiple soil erosion coefficients corresponding to the target region, and output the valve opening degree of the valve corresponding to the water pipe node group whose valve opening degree needs to be adjusted.

[0053] In some embodiments, the parameter generation model is obtained by training with a first training sample set. In some embodiments, the first training sample set includes a plurality of first training samples with first labels. The first training sample includes a sample soil erosion coefficient corresponding to one or more target regions, and the first label includes a historical control parameter of the target region. The historical control parameter refers to the temporary control parameter adopted at the historical time.

[0054] In some embodiments, the government regulation management platform determines the first training sample and the first label based on historical data. For example, the government regulation management platform can determine, based on the historical data, a historical control parameter in the historical data that has a good adjustment effect on the second historical time as the first label, and a historical loss coefficient of one or more target regions of a first historical time corresponding to the first label as the first training sample. Wherein, if the government regulation management platform determines that the historical control parameter has a good adjustment effect on the adjustment device based on the historical control parameter, the decline range of the soil and water loss coefficient of the one or more target regions exceeds the range threshold. The range threshold is pre-set based on historical experience. The historical loss coefficient refers to the soil and water loss coefficient in the historical data. The first historical time is earlier than the second historical time.

[0055] In some embodiments, the government regulation management platform can input the plurality of first training samples with the first label into the initial parameter generation model, construct a loss function through the first label and the result of the initial parameter generation model, and update the parameters of the initial parameter generation model through gradient descent and other methods based on the loss function. When the loss function meets the preset condition, the trained parameter generation model is obtained. Wherein, the preset condition can include that the loss function converges, the number of iterations reaches a threshold, etc.

[0056] In some embodiments, the temporary control parameter can also include an output power and an on-off parameter. The government regulation management platform can control the operation of the water supply pump based on the output power, and control the operation of the user equipment based on the on-off parameter. For details of the water supply pump and the user equipment, see Figure 1 and related descriptions.

[0057] The output power refers to the output power of the water supply pump. The on-off parameter refers to the parameter for controlling the on-off of the user equipment. One water supply pump can supply water to one or more target regions.

[0058] In some embodiments, the government regulation management platform can determine the output power and the on-off parameter based on the soil and water loss parameter corresponding to the one or more target regions. For example, if the soil and water loss coefficient corresponding to the one or more regions corresponding to the water supply pump is greater than a first threshold, the government regulation management platform can reduce the output power of the water supply pump. The greater the value of the soil and water loss coefficient greater than the first threshold, the lower the output power. The soil and water loss coefficient corresponding to multiple regions can be represented by the average value of the soil and water loss coefficient.

[0059] For another example, if the soil and water loss coefficient of the target region is greater than a second threshold, the government regulation management platform can determine the on-off parameter of the user equipment corresponding to the target region as off.

[0060] In some embodiments, the government regulation management platform sends the output power and the on-off parameter to the water company management platform through the government regulation sensing network platform, and the water company management platform sends the output power and the on-off parameter to the smart water equipment object platform through the water company sensing network platform. Based on the output power and the on-off parameter, the smart water equipment object platform takes the output power in the temporary control parameter as the output power of the water supply pump, and controls the opening or closing of the user equipment based on the on-off parameter.

[0061] In some embodiments of the present specification, the water pressure of the water supply pipe network can be balanced by adjusting the output power of the water supply pump, and unnecessary water consumption can be limited by closing the user equipment, thereby reducing the load of the water supply pipe network and further reducing the risk of water and soil erosion.

[0062] Step 240, based on the temporary control parameter, control the valve opening degree of the corresponding valve in one or more target areas.

[0063] In some embodiments, the corresponding valve in one or more target areas can include the valve corresponding to the upstream water pipe node of the target area, etc. The greater the valve opening degree of the valve corresponding to the upstream water pipe node of the target area, the greater the water flow into the target area.

[0064] In some embodiments, the corresponding valve in one or more target areas can also include the valve corresponding to the water pipe node group in the target area, etc. The greater the valve opening degree of the valve corresponding to the water pipe node group, the greater the water flow into the water pipe node group.

[0065] In some embodiments, the government regulation management platform can determine the target valve based on the temporary control parameter, and adjust the valve opening degree of the target valve based on the temporary control parameter, so as to adjust the water flow of the overall target area or one or more water pipe node groups in the target area. The target valve refers to the valve whose valve opening degree needs to be adjusted.

[0066] In some embodiments of the present specification, by analyzing the time sequence flow of multiple water pipe node groups, the water and soil erosion coefficient of the target area is determined, and then the control parameter of the adjusting device is determined, thereby realizing real-time monitoring and intelligent control of the water supply pipe network, improving monitoring accuracy, response speed and operation efficiency, reducing water and soil erosion and water resource waste, and providing effective technical support for water pipe management of smart city.

[0067] It should be noted that the above description of the smart city water supply pipe network emergency regulation method process 200 is only for example and illustration, and does not limit the scope of the present specification. Those skilled in the art can make various modifications and changes to the smart city water supply pipe network emergency regulation method process 200 under the guidance of the present specification. However, these modifications and changes are still within the scope of the present specification.

[0068] Figure 3 is an exemplary schematic diagram of a coefficient generation model according to some embodiments of the present specification.

[0069] In some embodiments, the smart water equipment object platform further comprises a detection robot. The government regulation management platform collects soil features 310 of a water pipe node group in one or more target areas through the detection robot, constructs a water and soil erosion atlas 330 based on the soil features 310 and the time series flow 320, and determines a water and soil erosion coefficient 350 through the coefficient generation model 340 based on the water and soil erosion atlas 330. For descriptions of the detection robot, the water pipe node group, the time series flow, and the water and soil erosion coefficient, please refer to Figure 1 and Figure 2 and related contents thereof.

[0070] The soil feature refers to data related to the soil at the location of the water pipe node group. In some embodiments, the soil feature can include at least one of soil density, soil moisture content, and soil viscosity.

[0071] The water and soil erosion atlas refers to a graph structure for characterizing the water and soil erosion of the water supply pipe network. In some embodiments, the government regulation management platform can construct the water and soil erosion atlas based on the soil features and the time series flow of the water pipe node group in one or more target areas. In some embodiments, the main structure of the water and soil erosion atlas is determined based on the structure of the water supply pipe network. The water and soil erosion atlas includes multiple nodes (such as node 331, etc.) and edges (such as edge 332, etc.).

[0072] In some embodiments, the government regulation management platform takes a water pipe node group in a target area as a node, and there is an edge between two adjacent water pipe node groups. The node feature can include the target area to which the water pipe node group belongs and the corresponding soil feature, etc. The edge feature can include the water flow direction between the nodes, the adjacency distance, and the time series flow difference between the two nodes connected by the edge, etc. For descriptions of the two adjacent water pipe node groups, the adjacency distance, and the time series flow difference, please refer to Figure 2 and related descriptions thereof.

[0073] In some embodiments, the node feature can further include a node hotspot offset.

[0074] The node hotspot offset refers to the distance between the node and the corresponding target control hotspot. For descriptions of the control hotspot, please refer to Figure 4 and related descriptions thereof.

[0075] In some embodiments, one node can correspond to one or more target control hotspots. The government regulation management platform can determine the target control hotspots based on the location of the node and the location of the control hotspots determined when the temporary control parameters were determined last time. For example, the government regulation management platform can calculate the distance between the node and the plurality of control hotspots determined when the temporary control parameters were determined last time, determine one or more control hotspots with a distance less than a node distance threshold as the target control hotspots corresponding to the node, and calculate the distance between each target control hotspot and the node to obtain one or more node hotspot offsets. The node distance threshold can be pre-set based on historical experience.

[0076] In some embodiments of the present specification, the introduction of the node hotspot offset as a node feature can provide valuable information from the perspective of spatial distribution and regulation effect, obtain a more comprehensive soil erosion map, and further improve the accuracy of determining the soil erosion coefficient.

[0077] In some embodiments, the node feature can also include a detection parameter of the detection robot.

[0078] The detection parameter refers to a parameter related to the sampling of the detection robot. In some embodiments, the detection parameter includes the sampling amount and sampling depth at different water pipe node groups, etc. The sampling depth can be the soil depth sampled by the detection robot. The government regulation management platform counts the sampling amount and sampling depth corresponding to the water pipe node group into the node feature of the node corresponding to the water pipe node group.

[0079] In some embodiments of the present specification, the detection parameter of the detection robot can reflect the reliability of the soil feature, so that the soil erosion map is more referable, and further improve the accuracy of determining the soil erosion coefficient.

[0080] In some embodiments, at least one key edge can be included in the soil erosion map. The edge feature of the at least one key edge includes the erosion-related coefficient.

[0081] The key edge refers to an edge in the soil erosion map that needs to be focused on. In some embodiments, the key edge can be pre-set based on historical experience.

[0082] In some embodiments, the government regulation management platform can determine at least one key edge based on the time series flow difference of the two nodes connected by the edge in the soil erosion map.

[0083] In some embodiments, the government regulation management platform can determine the edge in the soil erosion map as a key edge if the time series flow difference of the two nodes connected by the edge is greater than a second difference threshold. The second difference threshold can be pre-set based on historical experience.

[0084] In some embodiments of the present specification, edges with larger time flow differences can better reflect the dynamic changes of soil and water loss, and setting key edges in the soil and water loss atlas can provide more information for determining the soil and water loss coefficient.

[0085] The loss-related coefficient refers to a coefficient representing the correlation of soil and water loss between the water pipe node groups. In some embodiments, the government supervision and management platform can determine the loss-related coefficient corresponding to each key edge based on historical data through correlation analysis, etc. For example, the government supervision and management platform can obtain the historical loss coefficients corresponding to the nodes connected by the key edge at multiple historical time points, etc., and calculate the loss-related coefficient corresponding to the key edge after preprocessing the obtained data. The preprocessing can include cleaning, denoising, filling missing values, unifying time scale and spatial resolution, standardization, time alignment, etc. The multiple historical time points can be pre-set based on historical experience. The historical loss coefficient refers to the soil and water loss coefficient in the historical data.

[0086] In some embodiments, the government supervision and management platform can calculate the loss-related coefficient corresponding to the key edge based on the following formula (2): P = cos ((x1, …, xn), (y1, …, yn)) (2) Wherein, P is the loss-related coefficient corresponding to the key edge, x1 is the historical loss coefficient of water pipe node group x at historical time point 1, y1 is the historical loss coefficient of water pipe node group y at historical time point 1, xn is the historical loss coefficient of water pipe node group x at historical time point n, yn is the historical loss coefficient of water pipe node group y at historical time point n. Wherein, water pipe node group x and water pipe node group y are two nodes connected by the key edge.

[0087] In some embodiments of the present specification, setting key edges can reduce the amount of data calculation while ensuring sufficient information. Since the target area does not exist alone, considering the correlation of soil and water loss between nodes can further integrate the influencing factors around each target area, thereby improving the accuracy of determining the soil and water loss coefficient.

[0088] In some embodiments, the coefficient generation model (also referred to as the coefficient generation large model) can be a machine learning model. For example, the coefficient generation model can be any one or combination of a graph neural network (GNN) model or a large model, etc. or other custom model structures, etc.

[0089] In some embodiments, the input of the coefficient generation model includes the soil and water loss atlas, and the output includes the soil and water loss coefficient corresponding to each node in the soil and water loss atlas.

[0090] In some embodiments, the coefficient generation model is trained by a second training sample set. The second training sample set includes a plurality of second training samples with second labels. The second training samples include sample soil erosion maps, and the second labels include actual soil erosion coefficients corresponding to each node in the sample soil erosion maps.

[0091] In some embodiments, the second training samples can be obtained based on historical data, and the second labels can be obtained by manual annotation. For example, the government regulatory management platform can take the soil erosion map of a third historical time in the historical data as a second training sample, and take the actual soil erosion coefficient corresponding to each node of a fourth historical time as a second label. The actual soil erosion coefficient can be obtained by manual rainfall simulation method and the like. The third historical time is earlier than the fourth historical time.

[0092] In some embodiments, the training process of the coefficient generation model is similar to that of the parameter generation model, and the training process can refer to that of the parameter generation model.

[0093] In some embodiments of the present specification, soil feature data is automatically collected by a detection robot, which not only improves the efficiency and accuracy of data collection, but also reduces the dependence on manual intervention, and provides technical support for large-scale regional soil erosion monitoring. By fusing the time series flow and soil features, combined with a machine learning large model, the soil erosion coefficient can be determined more accurately and quickly, avoiding evaluation bias caused by a single indicator.

[0094] Figure 4 is an exemplary schematic diagram of generating temporary control parameters according to some embodiments of the present specification.

[0095] In some embodiments, the government regulatory management platform determines at least one control hotspot 420 based on the pipe network pressure map structure 410 and the soil erosion coefficient 350, and generates temporary control parameters 430 based on the at least one control hotspot 420. For descriptions of the soil erosion coefficient and the temporary control parameter, see Figure 2 and related descriptions thereof.

[0096] The pipe network pressure map structure is a statistical map reflecting the water pressure of different regions or positions in the water supply pipe network. In some embodiments, the pipe network pressure map structure can be represented by a heat map or the like. For example, the government regulatory management platform can represent the water pressure of different regions or positions in the water supply pipe network by color (e.g., blue for low pressure, red for high pressure, etc.) in a structural layout map reflecting the water supply pipe network, and represent the water pressure difference between different regions or positions by color gradient, to obtain the pipe network pressure map structure.

[0097] A control hot spot refers to a pipe node in the pipe network pressure map structure that needs or is able to be regulated. A pipe node in the pipe network pressure map structure can be a water pipe node. For the description of the water pipe node, see Figure 2 and related content.

[0098] In some embodiments, the government regulatory management platform can determine at least one control hot spot based on the pipe network pressure map structure and the water and soil erosion coefficient. For example, the government regulatory management platform determines a pipe node corresponding to a high-pressure location in the pipe network pressure map structure as a control hot spot, and determines a pipe node corresponding to a physical center of a target region with a water and soil erosion coefficient greater than an erosion coefficient threshold as a control hot spot. The erosion coefficient threshold can be preset according to prior experience. The physical center of the target region can include the most central region or location in the water supply pipe network corresponding to the target region, etc.

[0099] The high-pressure location in the pipe network pressure map structure refers to the region or location with the highest water pressure in the pipe network pressure map structure. For example, the high-pressure location can be the region or location with the deepest red color in the pipe network pressure map structure.

[0100] In some embodiments, if the target region corresponds to multiple water and soil erosion coefficients, the government regulatory management platform can calculate the mean of the multiple water and soil erosion coefficients, and take the mean as the water and soil erosion coefficient of the target region to determine whether the water and soil erosion coefficient is greater than the erosion coefficient threshold.

[0101] In some embodiments, the government regulatory management platform can generate temporary control parameters based on at least one control hot spot. For example, for a control hot spot determined by a high-pressure location in the pipe network pressure map structure, the government regulatory management platform can determine the temporary control parameters (i.e., the valve opening degree of the valve of the pipe corresponding to the control hot spot, etc.) corresponding to the control hot spot according to the water pressure corresponding to the control hot spot by querying the parameter preset table. For a control hot spot determined by the water and soil erosion coefficient, the government regulatory management platform can determine the temporary control parameters corresponding to the control hot spot by the parameter generation model. For the description of the parameter generation model, see Figure 2 and related content.

[0102] In some embodiments, the parameter preset table can be constructed according to prior experience, including multiple water pressures and temporary control parameters corresponding to each water pressure.

[0103] In some embodiments, the government regulatory management platform determines one or more high-sensitive areas in the pipe network pressure map structure based on multiple historical pressure map structures and multiple historical erosion coefficients in a preset historical period, and determines at least one control hot spot based on the one or more high-sensitive areas and the water and soil erosion coefficient.

[0104] In some embodiments, the preset historical time period can be a period of time earlier than the current time point. The preset historical time period can be preset based on historical experience, including a plurality of historical time points.

[0105] The historical pressure map refers to a pipe network pressure map structure in the historical data. The historical pressure map includes one or more target regions corresponding to the historical water pressure at the historical time point. The historical erosion coefficient refers to the water and soil erosion coefficient corresponding to the target region in the historical data.

[0106] In some embodiments, one historical time point can correspond to one historical pressure map and one or more historical erosion coefficients corresponding to the target region at the historical time point.

[0107] The high sensitivity area refers to an area in the pipe network pressure map structure where the water and soil erosion coefficient is highly sensitive to the water pressure.

[0108] In some embodiments, for each target region, the government regulatory management platform can determine a plurality of historical water pressures of the target region based on the historical pressure map corresponding to a plurality of historical time points, and determine the water pressure fluctuation based on the plurality of historical water pressures. The government regulatory management platform can also determine the coefficient fluctuation of the target region based on the historical erosion coefficient corresponding to a plurality of historical time points. The coefficient fluctuation can reflect the change of the historical erosion coefficient.

[0109] In some embodiments, the pressure fluctuation can be represented by the difference between the historical water pressures of two adjacent historical time points. The coefficient fluctuation can be represented by the difference between the historical erosion coefficients of two adjacent historical time points.

[0110] In some embodiments, in response to the pressure fluctuation of the target region being less than the pressure fluctuation threshold and the coefficient fluctuation of the target region being greater than the coefficient fluctuation threshold, the government regulatory management platform determines the target region as a high sensitivity area. The pressure fluctuation threshold and the water and soil erosion fluctuation threshold can be preset according to prior experience.

[0111] In some embodiments, the government regulatory management platform can determine at least one control hotspot based on the high sensitivity area and the water and soil erosion coefficient. For example, the government regulatory management platform can determine the high pressure position of each high sensitivity area as a control hotspot, and determine the pipe node corresponding to the physical center of the target region with a water and soil erosion coefficient greater than the erosion coefficient threshold as a control hotspot. The high pressure position of the high sensitivity area represents the area or position with the highest water pressure in the high sensitivity area. For example, the high pressure position of the high sensitivity area can be the area or position with the deepest red color in the high sensitivity area.

[0112] In some embodiments, for the control hotspot determined by the high pressure position of the high sensitivity area, the government regulatory management platform can determine the temporary control parameter corresponding to the control hotspot by querying the parameter preset table according to the water pressure corresponding to the control hotspot.

[0113] In some embodiments of the present specification, based on the historical pressure map and the historical loss coefficient, determining the high sensitive area in the pipe network pressure map structure can determine the area that may have a water loss risk in advance, and improve the timeliness of risk early warning. By the high sensitive area and the water and soil loss coefficient, the control hot spot and the temporary control parameter corresponding to the control hot spot are determined, so that the area that is more likely to have a water and soil loss risk can be prevented in advance.

[0114] In some embodiments of the present specification, by the pipe network pressure map structure and the water and soil loss coefficient, the control hot spot and the temporary control parameter corresponding to the control hot spot are determined synchronously, so that the area that may have a water and soil loss risk can be prevented in advance. By differentiating the temporary control parameter, energy waste caused by global adjustment can be avoided.

[0115] The present specification also provides a computer readable storage medium, which stores computer instructions. When the computer reads the computer instructions in the storage medium, the computer executes the smart city water supply pipe network emergency supervision method in the above embodiments.

[0116] In addition, some features, structures or characteristics in one or more embodiments of the present specification can be appropriately combined.

[0117] Some embodiments use numbers to describe components, attributes, and the like. It should be understood that such numbers used in the description of embodiments are, in some examples, modified by the words "about", "approximately", or "generally". Unless otherwise stated, "about", "approximately", or "generally" indicates that the stated number can vary by ±20%. Accordingly, in some embodiments, numerical parameters in the specification and claims are approximations, and can vary depending on the desired characteristics of the individual embodiments. In some embodiments, numerical parameters should be considered in the context of the number of significant digits and rounding off, as is customary in the art. Although numerical ranges and parameters setting forth the broadest scope of the embodiments described in the present specification are approximations, in specific embodiments, such numerical values are set forth as precisely as possible.

[0118] If the use of descriptions, definitions, and / or terms in the cited materials in the present specification is inconsistent or conflicts with the descriptions, definitions, and / or terms used in the present specification, the use of descriptions, definitions, and / or terms in the present specification shall prevail.

Claims

1. A large-scale IoT model system for emergency monitoring of water supply networks in smart cities, characterized by: The system includes a government supervision management platform, a government supervision sensor network platform, a government supervision object platform, a water company sensor network platform, and a smart water equipment object platform. The government supervision object platform includes a water company management platform; the smart water equipment object platform includes monitoring equipment and at least one regulating device; the government supervision management platform is configured to: Based on the water flow characteristics of multiple water pipe nodes in the water supply network, the time series flow corresponding to multiple groups of water pipe nodes is determined; determining a soil and water loss coefficient corresponding to one or more target areas based on the time series flow; Based on the soil and water loss coefficient, a temporary control parameter is generated by a parameter generation model, wherein the parameter generation model is a machine learning model; Based on the temporary control parameter, the valve opening of the corresponding valve in the one or more target areas is controlled.

2. The system according to claim 1, wherein The smart water equipment object platform also includes a detection robot, and the government supervision and management platform is further configured to: collecting soil characteristics of the water pipe node groups in the one or more target areas by the detection robot; Constructing a soil and water loss map based on the soil characteristics and the time series flow, wherein the soil and water loss map includes a plurality of nodes and a plurality of edges; Based on the soil and water loss atlas, the soil and water loss coefficient is determined by a coefficient generation model, and the coefficient generation model is a machine learning model.

3. The system according to claim 2, wherein: The node characteristics of the multiple nodes include a node hotspot offset, where the node hotspot offset is the distance between the node and a target control hotspot.

4. The system according to claim 2, wherein: The soil and water loss atlas includes at least one key edge, and the edge feature of the at least one key edge includes a loss correlation coefficient.

5. The system according to claim 1, wherein: The government supervision and management platform is further configured to: determining at least one control hotspot based on the pipe network pressure map structure and the soil erosion coefficient; The temporary control parameter is generated based on the at least one control hotspot.

6. The system according to claim 5, wherein: The government supervision and management platform is further configured to: Determining one or more highly sensitive areas in the pipe network pressure map structure based on a plurality of historical pressure maps and a plurality of historical loss coefficients in a preset historical period; Based on the one or more highly sensitive areas and the soil erosion coefficient, the at least one control hotspot is determined.

7. The system according to claim 1, wherein: The temporary control parameters include output power and start / stop parameters. The government supervision and management platform is further configured to: Based on the output power, controlling the operation of the water supply pump; Based on the activation and deactivation parameters, the operation of the user equipment is controlled.

8. A method for emergency supervision of water supply network in smart city, characterized in that: The method is executed by a government supervision and management platform in a large-scale model system of the Internet of Things for emergency supervision of a smart city water supply network, and the method includes: Based on the water flow characteristics of multiple water pipe nodes in the water supply network, the time series flow corresponding to multiple groups of water pipe nodes is determined; determining a soil and water loss coefficient corresponding to one or more target areas based on the time series flow; Based on the soil and water loss coefficient, a temporary control parameter is generated by a parameter generation model, wherein the parameter generation model is a machine learning model; Based on the temporary control parameter, the valve opening of the corresponding valve in the one or more target areas is controlled.

9. The method according to claim 8, wherein The smart water equipment object platform further includes a detection robot, and determining the soil and water loss coefficient corresponding to one or more target areas based on the time series flow includes: collecting soil characteristics of the water pipe node groups in the one or more target areas by the detection robot; Constructing a soil and water loss map based on the soil characteristics and the time series flow, wherein the soil and water loss map includes a plurality of nodes and a plurality of edges; Based on the soil and water loss atlas, the soil and water loss coefficient is determined by a coefficient generation model, and the coefficient generation model is a machine learning model.

10. The method according to claim 8, wherein The method further comprises: determining at least one control hotspot based on the pipe network pressure map structure and the soil erosion coefficient; The temporary control parameter is generated based on the at least one control hotspot.

Images