An adaptive water supply and drainage pipeline flow adjusting method and system

Abstract

The application relates to the technical field of water supply and drainage flow monitoring, and discloses a self-adaptive water supply and drainage pipeline flow adjusting method and system, which comprises the following steps: identifying the positions of devices related to flow monitoring and adjustment through a three-dimensional diagram, obtaining a flow monitoring device position set and a flow adjusting device position set; converting collected flow simulation signals into digital signals and outputting the digital signals; training a self-adaptive flow adjusting model according to historical water supply and drainage pipeline flow data sets; inputting real-time water flow data into the self-adaptive flow adjusting model to obtain the supply and demand of water flow of the main line and branch line of the water supply and drainage pipeline; and adjusting the flow adjusting device according to the supply and demand to obtain the operation parameters of the adjusted flow adjusting device. The system comprises a signal output module, a supply and demand module and a parameter adjusting module. The application realizes dynamic adjustment and optimization of water supply and drainage flow, and enhances the self-adaptive ability and controllability of the system.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of water supply and drainage flow monitoring, and particularly relates to a self-adaptive water supply and drainage pipeline flow adjusting method and system. BACKGROUND

[0002] In a water supply and drainage system, flow regulation is an important means to ensure efficient operation of the system and meet user demand. By adjusting the opening of the valve, the flow of water can be controlled to meet different water demand. For example, during peak or low water consumption periods, the flow of water can be controlled by adjusting the valve to ensure stable water pressure and water supply demand.

[0003] Prior art one, Chinese patent, application number: 202121479212.4 discloses a water supply and drainage pipe with flow monitoring function, which comprises a meter and a soft layer. The right side of the meter is provided with a main body, and the upper outer side of the main body is provided with an upper cover body. The upper cover body is fixedly connected with a connecting rod, and the outer side of the connecting rod is provided with a shaft sleeve. The lower side of the shaft sleeve is fixedly connected with a handle. The lower outer side of the main body is provided with a lower cover body, and the surface of the lower cover body is fixedly connected with a clamping groove. The right end of the main body is fixedly connected with a connecting ring, and the surface of the connecting ring is provided with a clamping strip. The right end of the connecting ring is provided with a clamping ring. The soft layer is arranged in the inner part of the lower cover body. The inner part of the clamping strip is provided with a screw, and the inner side of the clamping strip is fixedly connected with a filter screen. Although the water supply and drainage pipe with flow monitoring function is provided with a soft layer, which has good softness, reduces the pressure between the lower cover body and the main body, effectively prevents damage to the surface of the main body, ensures the integrity of the main body, and is convenient to use, but the degree of intelligence is low, and the real-time monitoring and early warning capability is insufficient.

[0004] Prior art two, Chinese patent application, application number CN202410278739.2 discloses a flow regulating device and a flow regulating method, which comprises a main flow path, a side flow path, a large flow regulating valve and a high-frequency quick-response valve. The two ends of the side flow path are connected with the main flow path. The large flow regulating valve is arranged on the main flow path, and the high-frequency quick-response valve is arranged on the side flow path and is connected in parallel with the large flow regulating valve. Although the large flow regulating valve can quickly regulate the flow, and the high-frequency quick-response valve can accurately regulate the flow, by connecting the large flow regulating valve and the high-frequency quick-response valve in parallel, the flow regulating device can quickly regulate and accurately regulate the flow, but the valve regulating performance is limited by its own structure, and the cost is high, and the maintenance cost is high.

[0005] The prior art three, Chinese patent, application number CN202211463071.6 discloses a flow regulating valve and a flow regulating system, wherein the flow regulating valve comprises a valve body, an inlet cavity, a regulating cavity and a first outlet cavity are sequentially arranged in the valve body along the water flow direction; a first outlet is arranged in the first outlet cavity; a rotating body is rotatably arranged in the regulating cavity; a regulating ring is arranged in the regulating cavity and connected with the rotating body; a stepped groove is arranged on the outer wall of the regulating ring; when the regulating ring is rotated to partially block the first outlet under the driving of the rotating body, the water flow can flow out of the first outlet through the stepped groove; during the process that the regulating ring partially blocks the first outlet and continuously rotates to close the first outlet, the water flow can flow out of the first outlet through the stepped groove in a step-by-step decreasing manner. Although the flow regulation is realized by the step-by-step decreasing of the stepped groove, in the drainage system, the accuracy of the flow measurement has errors, which affects the accuracy of the flow regulation.

[0006] At present, the prior art one, the prior art two and the prior art three have the problems of low intelligentization degree, insufficient real-time monitoring and early warning capability, limited valve regulating performance by the structure itself, high cost and large error in the accuracy of flow measurement. Therefore, the present application provides a self-adaptive water supply and drainage pipeline flow adjustment method and system. SUMMARY

[0007] The main purpose of the present application is to provide a self-adaptive water supply and drainage pipeline flow adjustment method and system to solve the problems of low intelligentization degree, insufficient real-time monitoring and early warning capability, limited valve regulating performance by the structure itself, high cost and large error in the accuracy of flow measurement in the prior art.

[0008] To achieve the above purpose, the present application provides the following technical solutions:

[0009] A self-adaptive water supply and drainage pipeline flow adjustment method, the self-adaptive water supply and drainage pipeline flow adjustment method comprising:

[0010] Obtaining a three-dimensional diagram of the water supply and drainage pipeline, identifying the positions of devices related to flow monitoring and adjustment through the three-dimensional diagram, obtaining a set of flow monitoring device positions and a set of flow adjustment device positions; the flow monitoring device is a flow monitoring device, which collects the water flow of the water supply and drainage pipeline in real time; converting the collected flow simulation signal into a digital signal and outputting the digital signal;

[0011] Training the self-adaptive flow adjustment model according to the historical flow data set of the water supply and drainage pipeline; inputting the real-time water flow data into the self-adaptive flow adjustment model to obtain the supply and demand of the water flow of the main road and branch road of the water supply and drainage pipeline;

[0012] According to the supply and demand, the flow adjusting device is adjusted to obtain the operation parameter of the adjusted flow adjusting device; and the water pressure of the branch of the water supply and drainage pipeline after the operation parameter is adjusted is measured through the water pressure monitoring device, and if the trigger pressure threshold is reached, a warning is given.

[0013] As a further improvement of the application, the device positions related to flow monitoring and adjustment are identified through three-dimensional stereogram, including:

[0014] A plurality of view scanning devices are arranged in the target water supply and drainage pipe network area, 360-degree rotation scanning is performed, and a plurality of scanning views are obtained; meanwhile, a plurality of view high-resolution cameras are used for synchronous shooting to obtain texture information of the devices and pipe surfaces;

[0015] The plurality of scanning views are integrated into a single three-dimensional surface model to preliminarily generate point cloud data; the preliminarily generated point cloud data is post-processed to remove noise points and outliers beyond the distance threshold, the point cloud data is cleaned, and pre-processed point cloud data is obtained;

[0016] The pre-processed point cloud data is analyzed, the surfaces of the curved pipes and valve sets and pumps and other devices are geometrically fitted, and the rough shape of the device is obtained; the point cloud data is converted into a raster image, the significant edges are extracted from the raster image, the straight line direction edge part is selected, the same direction and similar edge points are gradually connected, and a plurality of independent straight line segments are formed;

[0017] The extracted straight line segments are analyzed, similar lengths and included angle ranges are set for combination, the straight line segments are aggregated, and a closed body representing the outline of the pipe and the device is formed; a preliminary structure diagram is constructed according to the aggregated line segments, which is used as the outline feature description of the device; the newly generated outline feature is compared with the device model in the database, the shape similarity and direction consistency are used to evaluate the similarity of the shape, and the position, angle and installation direction of the device are determined according to the matching result.

[0018] As a further improvement of the application, the plurality of scanning views are integrated into a single three-dimensional surface model, including:

[0019] At each scanning position, the view scanning device generates a point cloud data set containing three-dimensional coordinates and intensity information of objects around the position; a reference point is selected as the origin of the global coordinate system, and the directions of X, Y and Z coordinate axes are defined according to the far point;

[0020] Detecting feature points of edges and corner points in each point cloud; selecting feature points of overlapping regions in each data set to align two adjacent point clouds; registering adjacent point cloud data through rigid transformation; calculating a rotation matrix and a translation vector of each point cloud relative to a reference coordinate system through an iterative method based on least squares, using overlapping parts of multiple point clouds for multiple iterations, compared with a set tolerance threshold, until the error reaches the tolerance threshold;

[0021] Merging the registered point cloud data set according to the rotation matrix and the translation vector, transforming the coordinates of each point cloud data into a global coordinate system, and forming a point cloud data set; performing triangular meshing processing on the merged point cloud data set to convert the point cloud data set into a three-dimensional surface model.

[0022] As a further improvement of the application, the adaptive flow adjustment model is constructed, including:

[0023] Obtaining historical flow data containing historical real-time water flow data from each flow monitoring device, including flow information of the main road and branch road, and a timestamp, to form a complete flow time series; at the same time, recording water use events in the corresponding time period, including water use peak, device running state change and weather condition;

[0024] Organizing the historical flow data in time series to form a structured database; each record should contain time, main road flow, branch road flow, related water use events and user demand information; labeling sample data stored in the database, associating each set of water flow data with the corresponding supply demand to form a flow data and event and supply demand pair, and obtaining a sample data set;

[0025] Extracting features including daily, weekly and monthly trends and fluctuation amplitudes of water flow from the sample data set; converting the intensity and type of water use events into numerical features;

[0026] Using the sorted sample data set to construct a supply demand relationship table; associating the change patterns of main road and branch road flow with various water use events; creating a supply demand function to form a supply demand mapping by statistically analyzing the correlation between main flow characteristics and supply demand;

[0027] Training the adaptive flow adjustment model using the training set divided from the sample data set, evaluating the adaptive flow adjustment model using the validation set divided from the sample data set, predicting the accuracy of supply demand under specific conditions, and obtaining the accuracy by comparing the output results of the adaptive flow adjustment model with the results of the supply demand mapping according to the error of the adaptive flow adjustment model parameters.

[0028] As a further improvement of the application, the sample data set is obtained, including:

[0029] Obtain historical data set containing main road and branch water flow from flow monitoring equipment, each record has timestamp, main road flow and branch flow field; Extract information matching corresponding timestamp from inventory water use event data, including water use peak, device state change and weather information;

[0030] Clean up flow data and event data in historical data set, remove missing values and outliers, and convert all timestamps to unified time format; Connect flow data and corresponding water use event data based on timestamp;

[0031] Generate a structured data frame, whose rows represent water flow records at each time point, and columns include time, main road flow, branch flow and corresponding water use event information; For each flow data, create supply demand label according to water use event type and intensity, mark water flow in water use peak period as high demand, and mark water flow in low peak period as low demand; Organize processed flow data and corresponding supply demand label into a sample data set in table form, which contains flow related input fields and output fields.

[0032] As a further improvement of the application, the intensity and type of water use event are converted into numerical features, including:

[0033] For each water use event collected, numerical conversion is carried out according to its original intensity value and classification information; Extract water use in time period, quantify the frequency and intensity of event in time window and number; For events involving multiple users, the proportion of various water use conditions is converted into numerical features;

[0034] A set of numerical features is constructed, each record will contain multiple dimensional numerical information, including timestamp, main road flow, branch flow, water use event type, water use event intensity and proportion of various water use events;

[0035] Integrate the generated numerical features into a unified feature matrix to form the final data output for model construction; Each row of the matrix represents a complete water use event record at a time point, including all related numerical features.

[0036] As a further improvement of the application, a supply demand mapping is formed, including:

[0037] Extract key features related to supply demand from the arranged sample data set, including main road flow, branch flow, type and intensity of water use event, weather data set time factor; Each key feature has field identification;

[0038] The extracted key features are analyzed, the correlation coefficient is used to evaluate the relationship strength between each type of flow feature and supply demand, and the main factors affecting the supply demand are identified; based on the correlation analysis, a supply demand function is constructed, which predicts the output demand by inputting the features;

[0039] The labeled input and output pairs are fitted to form a mapping relationship; the trained supply demand function is converted into a mapping matrix, which represents the relationship between the input features and the output demand, and each row represents the predicted output under different input conditions, providing supply demand prediction for upcoming traffic conditions and water use events.

[0040] As a further improvement of the present application, the adaptive flow adjustment model is trained, including:

[0041] The input layer and output layer are configured to ensure that each main road flow, branch flow and water use event intensity input feature is mapped to the output demand; the sample data set is divided into training set and validation set, the training set is used for parameter learning of the adaptive flow adjustment model, and the validation set is used for evaluating the prediction ability and accuracy of the adaptive flow adjustment model;

[0042] The training set is used to train the adaptive flow adjustment model, and after each training period, the error between the output of the adaptive flow adjustment model and the actual supply demand is calculated; the loss function evaluated by the adaptive flow adjustment model is used to realize error feedback adjustment; the validation set is used to evaluate the accuracy of the adaptive flow adjustment model, and the accuracy and recall rate indicators are obtained by comparing the estimated results with the real supply demand;

[0043] After training is completed, the performance indicators of the adaptive flow adjustment model are automatically adjusted.

[0044] As a further improvement of the present application, the running parameters of the adjusted flow adjustment device are obtained, including:

[0045] After receiving the supply demand, a group of control signals are generated and sent to the flow adjustment device, including the current demand parameter, target flow, target pressure, and operation state monitoring instruction; the flow adjustment device performs real-time state evaluation on the valve and pump, and collects the current flow, pressure and state information;

[0046] The target value of the control signal is obtained, and the current running value of the valve and pump is compared with the target value to calculate the required adjustment amount; for each flow regulating valve, the opening value and conversion speed are generated according to the change of supply demand; the running state of the pump is adjusted in terms of speed or power output;

[0047] The newly generated running parameters of the valve and pump are saved in the configuration file of the device.

[0048] To achieve the above object, the present application further provides the technical scheme as follows.

[0049] An adaptive water supply and drainage pipeline flow adjustment system is applied to the adaptive water supply and drainage pipeline flow adjustment method, and the adaptive water supply and drainage pipeline flow adjustment system comprises:

[0050] A signal output module is configured to acquire a three-dimensional perspective view of the water supply and drainage pipeline, identify the positions of devices related to flow monitoring and adjustment through the three-dimensional perspective view, obtain a flow monitoring device position set and a flow adjustment device position set, collect flow simulation signals, convert the collected flow simulation signals into digital signals, and output the digital signals.

[0051] A supply demand module is configured to train an adaptive flow adjustment model according to a historical flow data set of the water supply and drainage pipeline, input real-time water flow data into the adaptive flow adjustment model, and obtain the supply demand of water flow of the main road and branch road of the water supply and drainage pipeline.

[0052] A parameter adjustment module is configured to adjust the flow adjustment device according to the supply demand, obtain the operating parameters of the adjusted flow adjustment device, measure the water pressure of the branch road of the water supply and drainage pipeline after the operating parameters are adjusted through a water pressure monitoring device, and perform early warning if a pressure threshold is triggered.

[0053] The application obtains a three-dimensional graph of the water supply and drainage pipeline, through the acquisition of the three-dimensional graph, so that the system can comprehensively understand the spatial structure and arrangement of the pipeline; the specific positions of the flow monitoring device and the flow adjusting device are identified to form a device position set; the real-time collection capability of the flow monitoring device is realized to ensure the timely availability of the flow data; the analog signal is effectively converted into a digital signal to lay a foundation for subsequent data processing. The significance achieved is that through detailed three-dimensional spatial information, the management and maintenance of the pipeline device can be more scientific; real and reliable basic data are provided for data analysis and flow adjustment to improve the accuracy and real-time performance of flow monitoring. The training of the adaptive flow adjustment model uses the historical flow data set to train the model to establish the understanding and prediction capability of water flow under different conditions; the real-time water flow data is input, and the model can automatically calculate the water flow supply demand of the main road and branch road to provide dynamic adjustment basis. The significance achieved is that the flow adjustment method has self-adaptive capability and can respond to the real-time changes of the water flow to improve the intelligent level of the system; it helps to optimize the allocation and use of water resources, reduces waste, and improves the economy and sustainability of the system. The adjustment of the flow adjusting device and the water pressure monitoring, according to the supply demand, adjust the flow adjusting device to obtain the operating parameters, so that the water flow meets the actual demand; the water pressure of the adjusted branch is monitored through the water pressure monitoring device, and if the water pressure exceeds the preset pressure threshold, the early warning mechanism is triggered. The significance achieved is that the adjusted water flow meets the actual demand, and the safety of the pipeline is maintained to prevent damage risks caused by system out of control; real-time system monitoring capability is provided to ensure the stable operation of the system through the timely early warning mechanism to enhance the safety and reliability of the water supply and drainage system. BRIEF DESCRIPTION OF DRAWINGS

[0054] Figure 1 A step flow chart for an embodiment of the adaptive water supply and drainage pipeline flow adjustment method of the application;

[0055] Figure 2 A step flow chart for an embodiment of the adaptive water supply and drainage pipeline flow adjustment method of the application through the three-dimensional graph to identify the device positions related to flow monitoring and adjustment;

[0056] Figure 3 A step flow chart for an embodiment of the adaptive water supply and drainage pipeline flow adjustment method of the application to construct the adaptive flow adjustment model;

[0057] Figure 4 A step flow chart for an embodiment of the adaptive water supply and drainage pipeline flow adjustment method of the application to obtain the operating parameters of the adjusted flow adjusting device;

[0058] Figure 5A functional module schematic diagram of an embodiment of the adaptive water supply and drainage pipeline flow adjusting system of the present application;

[0059] Figure 6 A structural schematic diagram of an embodiment of the electronic device of the present application;

[0060] Figure 7 A structural schematic diagram of an embodiment of the storage medium of the present application. DETAILED DESCRIPTION

[0061] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the 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 of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts fall within the scope of the present application.

[0062] The terms "first", "second", "third" in the present application are only for descriptive purpose, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first", "second", "third" can explicitly or implicitly include at least one of the features. In the description of the present application, the meaning of "multiple" is at least two, such as two, three, etc., unless otherwise explicitly and specifically limited. All directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present application are only used to explain the relative positional relationship, movement condition, etc. between the components in a certain posture (as shown in the drawings), and if the certain posture changes, the directional indications also change accordingly. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units is not limited to the listed steps or units, but can optionally include steps or units not listed, or can optionally include other steps or units inherent to the process, method, product or device.

[0063] In this document, reference to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the application. The appearances of the phrase "in an embodiment" in various places in the specification are not necessarily all referring to the same embodiment, nor are they necessarily mutually exclusive of one another. It is expressly understood that any of the features, structures, or characteristics described in connection with an embodiment can be included in a different embodiment.

[0064] As Figure 1 shown, the present embodiment provides an embodiment of the adaptive water supply and drainage pipeline flow adjusting method, which specifically includes the following steps in the present embodiment:

[0065] Step S1: Obtain a three-dimensional perspective view of the water supply and drainage pipeline, identify the positions of devices related to flow monitoring and adjustment through the three-dimensional perspective view, obtain a set of flow monitoring device positions and a set of flow adjustment device positions; the flow monitoring device is a flow monitoring device that collects the water flow of the water supply and drainage pipeline in real time; convert the collected flow analog signal into a digital signal and output the digital signal;

[0066] Wherein, the expression of converting the flow analog signal into a digital signal is:

[0067] Let the flow analog signal be Q(t), which is expressed as a vector changing at time t, containing the signals of multiple flow sensors:

[0068]

[0069] Wherein, 1, …, m represent the mark serial numbers of the flow monitoring devices, Q0 represents the reference flow, A1 represents the flow fluctuation amplitude of the first flow monitoring device, ω1 represents the angular frequency of the first flow monitoring device, φ1 represents the initial phase of the first flow monitoring device, and N1(t) represents the noise term of the first flow monitoring device.

[0070] Introduce a noise matrix N(t) to represent the influence of applying noise to the flow analog signal:

[0071]

[0072] Wherein, W(t) is a noise matrix related to time, and S is a Gaussian distribution.

[0073] Let the time delay matrix D represent the influence of time delay on the measured flow analog signal:

[0074]

[0075] d1 represents the time delay of the first flow monitoring device.

[0076] The delayed flow analog signal is represented as:

[0077] Q delay (t)=D·Q(t)

[0078] Quantization and conversion process, at the sampling time t(t n =nT)(where (T) is the sampling period), the sample of each flow analog signal is represented as:

[0079] Q[n]=Q delay (t n )=D·Q(t n )+D·N(tn )

[0080] t n represents the flow value of the nth sampling point;

[0081] The sample flow simulation signal is quantized. A quantization matrix is set: (Qquant), which is in the following form:

[0082]

[0083] The overall expression is:

[0084]

[0085] Delta Q represents the quantization step;

[0086] The expression for converting the flow simulation signal from analog to digital signal D[n] is obtained as:

[0087]

[0088] Step S2: According to the historical flow data set of the water supply and drainage pipeline, the adaptive flow adjustment model is trained; the real-time water flow data is input into the adaptive flow adjustment model to obtain the supply and demand of the water flow of the main road and branch road of the water supply and drainage pipeline;

[0089] Step S3: According to the supply and demand, the flow adjustment device is adjusted to obtain the operation parameters of the adjusted flow adjustment device; at the same time, the water pressure of the branch road of the water supply and drainage pipeline after adjusting the operation parameters is measured by the water pressure monitoring device, and if the pressure threshold is triggered, a warning is given.

[0090] Preferably, step S1 of the embodiment acquires a three-dimensional perspective view of the water supply and drainage pipeline. Through the acquisition of the three-dimensional perspective view, the system can comprehensively understand the spatial structure and arrangement of the pipeline; identify the specific positions of the flow monitoring device and the flow adjusting device to form a device position set; realize the real-time collection capability of the flow monitoring device to ensure the timely availability of flow data; and effectively convert the analog signal into a digital signal to lay a foundation for subsequent data processing. The significance achieved is that through detailed three-dimensional spatial information, the management and maintenance of pipeline equipment can be more scientific; real and reliable basic data are provided for data analysis and flow adjustment to improve the accuracy and real-time performance of flow monitoring. Step S2 trains the self-adaptive flow adjustment model. The model is trained using historical flow data sets to establish the understanding and prediction capability of water flow under different conditions; and the model can automatically calculate the water flow supply demand of the main road and branch road by inputting real-time water flow data to provide dynamic adjustment basis. The significance achieved is that the flow adjustment method has self-adaptive capability and can respond to real-time changes in water flow conditions to improve the intelligent level of the system; and it helps to optimize the allocation and use of water resources, reduce waste, and improve the economy and sustainability of the system. Step S3 adjusts the flow adjusting device and monitors the water pressure. The flow adjusting device is adjusted according to the supply demand to obtain operating parameters so that the water flow meets the actual demand; and the water pressure of the adjusted branch is monitored by the water pressure monitoring device. If the water pressure exceeds the preset pressure threshold, the early warning mechanism is triggered. The significance achieved is to ensure that the adjusted water flow meets the actual demand while maintaining the safety of the pipeline to prevent damage risks caused by system out of control; and real-time system monitoring capability is provided to ensure stable operation of the system through the timely early warning mechanism to enhance the safety and reliability of the water supply and drainage system.

[0091] In summary, the embodiment realizes dynamic adjustment and optimization of water supply and drainage flow through accurate data collection, intelligent model training, and real-time monitoring to enhance the self-adaptive capability and controllability of the system. The significance lies in improving the utilization efficiency of water resources, reducing management and maintenance costs, and enhancing the robustness and ability to respond to unexpected situations of the system, which has important contribution to improving the intelligent level of urban infrastructure.

[0092] Further, as shown in Figure 2 the process of identifying the device positions related to flow monitoring and adjustment through the three-dimensional perspective view in step S1 specifically includes the following steps:

[0093] Step S11: multiple view scanning devices are arranged in the target water supply and drainage pipeline network area to perform 360-degree rotation scanning to obtain multiple scanning views; and multiple-view high-resolution cameras are used for synchronous shooting to acquire texture information of the devices and the pipeline surface;

[0094] Step S12: integrate the plurality of scanning views into a single three-dimensional surface model, preliminarily generate point cloud data; post-process the preliminarily generated point cloud data, remove noise points and outliers beyond a distance threshold, clean the point cloud data, and obtain pre-processed point cloud data;

[0095] The expression for removing noise points is:

[0096] Cleaned P oint C loud=(x i ,y i ,z i );|;||(x i ,y i ,z i )-(x mid ,y mid ,z mid )||<∈

[0097] Wherein: (x i ,y i ,z i ) represents the coordinates of the i-th point in the point cloud; (x mid ,y mid ,z mid ) represents the center point coordinates of the point cloud data; ∈ is a distance threshold, points beyond the threshold will be considered as noise points and removed;

[0098] Step S13: analyze the pre-processed point cloud data, geometrically fit the curved surface of the pipeline and the surface of the valve set pump and other equipment, obtain the rough shape of the equipment; convert the point cloud data into a raster image, extract the significant edges from the raster image, select the edge parts in the straight line direction, and gradually connect the edge points in the same direction and close to each other to form a plurality of independent straight line segments;

[0099] Wherein, the rough shape of the equipment is obtained:

[0100]

[0101] Wherein: S represents the fitted curved surface function;

[0102] The edges are obtained by the following formula:

[0103]

[0104] Wherein: E(x,y) represents the edge strength of point (x,y); I(x,y) represents the pixel value of the image at (x,y); G x and G yThese are Sobel operators, used to detect horizontal and vertical edges respectively; next, they are used to connect edge points that are in the same direction and close to each other, forming multiple independent line segments;

[0105] Step S14: Analyze the extracted straight line segments, set similar lengths and angle ranges for combination, aggregate the straight line segments to form a closed shape representing the outline of the pipe and equipment, construct a preliminary structural diagram based on the aggregated line segments as the outline feature description of the equipment; compare the newly generated outline features with the equipment model in the database, evaluate the similarity of the form using shape similarity and orientation consistency, and determine the position, angle and installation direction of the equipment based on the matching results;

[0106] The following aggregation formula is used:

[0107] Combined L ine S egments=L j ;|;min(θ ij )<θ <max(θ ij );and;|L i |~|L j |

[0108] Where: L j θ represents the combined line segment. ij It is the angle between two line segments, |L i | and | L j | is the straight line segment L i and L j Length;

[0109] Preferably, the step S11 of the embodiment scans and image acquisition, obtains 360-degree point cloud data without dead angle by arranging multiple view scanning devices in the target area, ensures comprehensive recording of the actual geometric shape of the pipeline and equipment; synchronizes image acquisition of the multi-view high-resolution camera, adds surface texture information to the point cloud data, and helps texture analysis and visual recognition of the equipment. Meaning: Through the combination of laser and high-resolution image, it is ensured that the point cloud data contains rich spatial information and surface features, improving the quality of the overall data and the accuracy of the analysis; the obtained high-quality data lays a solid foundation for the realization of subsequent algorithms, and improves the accuracy of shape recognition and feature analysis. Step S12 point cloud integration and cleaning, multiple scanning views are integrated into a single three-dimensional surface model to form a complete point cloud dataset; the point cloud data is processed to delete noise points and outliers outside the distance threshold to obtain cleaned high-quality point cloud data. Meaning: Through denoising processing, the data in the analysis process is more accurate, and the risk of misidentification is reduced; when constructing the three-dimensional surface model, the authenticity and integrity of the data are ensured, which is crucial for subsequent geometric fitting and shape recognition. Step S13 geometric fitting and edge extraction, geometric fitting is performed on the curved surface of the pipeline and the equipment to generate a rough shape of the equipment; through rasterization processing, key edge information is extracted to provide a basis for subsequent straight line segment connection. Meaning: Through the generation of a rough shape, the appearance features of the equipment can be quickly identified, and the relationship and layout between the equipment are preliminarily defined; accurate edge information provides an important basis for subsequent straight line segment extraction and shape matching, ensuring that the expression of shape features is clearer. Step S14 straight line segment assembly and equipment positioning, the extracted straight line segments are combined according to the set aggregation conditions to form a closed geometric shape; the newly generated contour feature is compared with the equipment model in the database to determine the actual position, angle and installation direction of the equipment. Meaning: The closed body formed by combining the straight line segments can adaptively describe the geometric features of the equipment, thereby extracting the equipment feature information; by comparing with the model in the database, the equipment can be quickly and accurately positioned, and reference basis is provided for flow monitoring and adjustment, improving the intelligent level of system management.

[0110] In summary, the embodiment realizes accurate recognition and positioning of the flow monitoring and adjustment related equipment in the water supply and drainage pipe network. From data acquisition to point cloud integration, to geometric analysis and equipment positioning, all links are closely connected and complementary to each other, and finally a high-efficiency, accurate and intelligent flow monitoring and adjustment system is built. It can effectively improve the efficiency of equipment management, reduce labor cost, and respond more quickly and accurately to potential flow problems.

[0111] Further, the process of integrating multiple scanning views into a single three-dimensional surface model in step S12 specifically includes the following steps:

[0112] Step S121: At each scanning position, the view scanning device generates a point cloud dataset containing three-dimensional coordinates and intensity information of the objects around the position; a reference point is selected as the origin of the global coordinate system, and the directions of the X, Y and Z coordinate axes are defined according to the far point;

[0113] Step S122: The feature points of edges and corners in each point cloud are detected; the feature points of the overlapping regions are selected in each dataset, and the two adjacent point clouds are aligned; the adjacent point cloud datasets are registered through rigid transformation; the rotation matrix and the translation vector of each point cloud relative to the reference coordinate system are calculated through an iterative method based on the least square method, the overlapping parts of the multiple point clouds are used for multiple iterations, and compared with the set tolerance threshold, until the error reaches the tolerance threshold;

[0114] Step S123: The registered point cloud datasets are merged according to the rotation matrix and the translation vector, the coordinates of each point cloud dataset are transformed into the global coordinate system, and a point cloud dataset is formed; the merged point cloud dataset is subjected to triangular meshing processing, and the point cloud dataset is converted into a three-dimensional surface model.

[0115] Preferably, in step S121 of this embodiment, point cloud data generation and global coordinate system definition involve generating a point cloud dataset containing the three-dimensional coordinates and intensity information of surrounding objects at each scanning position using a view scanning device. The point cloud data provides geometric information in space. By selecting a reference point as the origin of the coordinate system and defining the directions of the X, Y, and Z axes, a unified reference coordinate system is established. Significance: Rich point cloud data provides a foundation for the construction of three-dimensional surface models. The point cloud data captures the geometric shape, surface features, and environmental information within the target area. By setting a global coordinate system, a spatial reference is provided for the integration of data from different views, avoiding errors caused by coordinate differences. Step S122 involves feature point detection and point cloud registration. Edges and corners in the point cloud are detected, and key information is extracted to enhance the effect of subsequent registration. By selecting feature points in overlapping areas, more important geometric features in the dataset can be effectively utilized. Using an iterative method based on least squares, adjacent point clouds are effectively aligned by calculating the rotation matrix and translation vector between different point clouds until the registration error of all point clouds is reduced to below the set tolerance threshold. Significance: Accurate feature point selection and extraction improve the point cloud registration effect, making the correspondence between multiple views more reliable. By registering adjacent point clouds based on feature points, the coordination of various data views under a unified reference system is ensured, improving the accuracy and consistency of the integrated 3D surface model. Step S123: Point cloud merging and meshing. The registered point cloud datasets are merged according to the calculated rotation matrix and translation vector to form a single point cloud dataset, and the coordinates of all points are transformed to the global coordinate system. The merged point cloud dataset is then meshed, converting the discrete point cloud data into a continuous 3D surface model to form a complete and visually appealing 3D structure. Significance: Through data merging, the integrated point cloud can form a more comprehensive and detailed 3D surface model, removing redundant information and improving the model's visualization effect. The generated 3D surface model provides a foundation for subsequent analysis, visualization, simulation, and other applications, ensuring practicality and accuracy in engineering applications, flow monitoring, and equipment management.

[0116] In summary, this embodiment realizes the integration process of multi-view scanning views. Through point cloud generation, feature extraction and registration, and final merging and meshing, an accurate and complete three-dimensional surface model is successfully constructed. The three-dimensional surface model lays the foundation for intelligent management, maintenance decisions, and subsequent spatial analysis, and can improve the efficiency and scientific nature of engineering project management.

[0117] Furthermore, such as Figure 3 As shown, the construction process of the adaptive flow adjustment model in step S2 specifically includes the following steps:

[0118] Step S21: Obtain historical flow data containing real-time water flow data from various flow monitoring devices, including flow information of main and branch roads, as well as timestamps, to form a complete flow time series; at the same time, record water usage events within the corresponding time period, including peak water usage, changes in equipment operating status, and weather conditions, etc.

[0119] Step S22: Organize the historical flow data according to the time series to form a structured database; each record should include information such as time, main road flow, branch road flow, related water use events and user demand; label the sample data stored in the database, associate each set of water flow data with the corresponding supply and demand, form input (flow data and events) and output (supply and demand) pairs, and obtain the sample dataset;

[0120] Step S23: Extract features from the sample dataset that include daily, weekly, and monthly trends and fluctuations in water flow; convert the intensity and type of water use events into numerical features, such as the proportion of water use by households, factories, parks, etc. in a certain period of time;

[0121] Step S24: Construct a supply and demand relationship table using the organized sample dataset; associate the change patterns of main and branch road flow with various water use events; create a supply and demand function, and form a supply and demand mapping by statistically analyzing the correlation between main flow characteristics and supply and demand;

[0122] Step S25: Train the adaptive flow adjustment model using the training set divided from the sample dataset, evaluate the adaptive flow adjustment model using the validation set divided from the sample dataset, and predict the accuracy of supply and demand under specific conditions. The accuracy is obtained by comparing the output of the adaptive flow adjustment model with the result of supply and demand mapping. The error of the adaptive flow adjustment model parameters is determined based on the accuracy.

[0123] Preferably, step S21 of the present embodiment collects historical water flow data and time stamps, water usage events, and other information from each flow monitoring device to form a complete time series data. These data provide the basis for subsequent analysis and model construction. The significance achieved: to ensure the comprehensiveness and accuracy of the data, so that the model can obtain the real water usage, thereby improving the accuracy of the model prediction. Historical data is the basis for analysis, which can reflect the trend of supply and demand changes. Step S22 data arrangement, the collected raw flow data is arranged, labeled, and converted into a structured database format, forming input and output pairs, obtaining sample data sets, laying the foundation for data analysis and model training. The significance achieved: structured data facilitates subsequent algorithm processing, enhancing the usability of the data; by labeling sample data, the relationship between input and output can be clearly defined, providing necessary information for establishing an accurate model. Step S23 feature extraction, extracting flow features (such as daily, weekly, monthly trends and fluctuation amplitude) and numerical features of water usage events from the sample data set, and analyzing the water flow changes in different time dimensions. The significance achieved: feature extraction is an important part of machine learning, by capturing key features, the model can better understand and map the relationship between supply and demand, improving the accuracy and efficiency of prediction; by numerically analyzing water usage events, their impact on flow changes can be quantitatively analyzed. Step S24 supply and demand relationship construction, using the arranged sample data set to establish a supply and demand relationship table, analyzing the correlation between main road and branch flow and water usage events, and creating a supply and demand function. The significance achieved: constructing the supply and demand mapping relationship can help understand the reasons and patterns of flow changes; in addition, the supply and demand mapping relationship can be used for real-time monitoring and adjustment of flow, improving the rational allocation and use efficiency of water resources. Step S25 model training and evaluation, using the training set to train the adaptive flow adjustment model, and evaluating the prediction accuracy of the model through the validation set, comparing the output results with the supply and demand mapping results, and adjusting the model parameters. The significance achieved: by evaluating the prediction ability of the model, the model can be continuously optimized to have higher accuracy and adaptability in actual application, and the adaptability is beneficial to flow adjustment under different conditions and flexible response to complex water demand.

[0124] In summary, the present embodiment forms a complete chain from data collection to model evaluation. Each step contributes to more efficient water resource management and optimized allocation. Ultimately, an adaptive flow adjustment system that can respond to drainage pipeline demand in real time, accurately, and intelligently is obtained, improving the use efficiency of water resources and ensuring water supply safety.

[0125] Further, the process of obtaining the sample data set in step S22 specifically includes the following steps:

[0126] Step S221: Obtain the historical dataset containing the main and branch water flow from the flow monitoring device, each record has fields such as timestamp, main flow and branch flow, etc.; Extract the information matched with the corresponding timestamp from the inventory water use event data, including water use peak, device state change and weather information, etc.

[0127] Step S222: Clean the flow data and event data in the historical dataset, remove missing values and outliers, and convert all timestamps to a unified time format; Connect the flow data with the corresponding water use event data based on the timestamp;

[0128] Step S223: Generate a structured data frame, whose rows represent water flow records at each time point, and columns include time, main flow, branch flow and corresponding water use events, etc. For each flow data, create a supply demand label according to the type and intensity of water use event, mark the water flow in the water use peak period as high demand, and mark the water flow in the low peak period as low demand; Organize the processed flow data and corresponding supply demand label into a sample dataset in table form, which contains flow-related input fields (main flow, branch flow, event type, etc.) and output fields (supply demand label).

[0129] Preferably, step S221 of the present embodiment collects and matches data, extracts water flow history data of the main road and branch road from the flow monitoring device, and extracts water use event information matching the timestamp from the inventory. This process ensures that the data obtained covers multiple dimensions of flow (such as time, location) and key factors causing flow changes (such as water use peak, device change); by integrating flow monitoring data and water use event data, a rich context is established, which can more comprehensively consider various factors affecting supply and demand. Significance: Ensure the accuracy and completeness of the data used in the analysis process; multiple dimensions of data connect flow and water use behavior, allowing analysts to understand changes in water resource demand in different scenarios. Step S222 data cleaning and integration, by removing missing values and outliers, improves the quality of the data set, and the cleaning process makes the data more consistent and reliable, reducing interference factors in subsequent analysis; convert all timestamps to a unified format to ensure that the time information in the data set can be seamlessly connected, facilitating the effective connection of flow data and water use events through timestamps. Significance: Cleaned data can more accurately reflect the real situation and reduce false analysis caused by low data quality; by standardizing the timestamp, accurate matching of flow data and event data becomes possible, laying the foundation for building an effective supply and demand model. Step S223 sample data set construction, the structured data frame formed clearly lists the water flow at each time point, providing a clear framework for data processing and analysis; according to the type and intensity of water use events, supply and demand labels (such as "high demand", "low demand") are assigned to the flow data, and the labels cover different situational states in the flow; the sample data set formed has clear input and output fields, with a clear structure for storage, query and subsequent use. Significance: The structure design of input and output in the sample data set makes subsequent machine learning and prediction analysis simple and clear, supporting model training and effect evaluation; by combining flow data with supply and demand labels, analysts can identify and understand the relationship between supply and demand and their trends, providing data support for subsequent scientific decision-making.

[0130] In summary, the present embodiment establishes a high-quality, structured sample data set to facilitate subsequent model training and analysis. The final output sample data set not only has good functionality, but also can produce real value in practical applications. Through this process, technical personnel can provide more accurate data basis for water resource management, supply and demand prediction, and decision support systems.

[0131] Further, the process of converting the intensity and type of water use events into numerical features in step S23 specifically includes the following steps:

[0132] Step S231: For each water use event collected, numerical conversion is performed according to its original intensity value and classification information; the use of water in the time period is extracted, the frequency and intensity of the event in the time window are quantified and numbered; for events involving multiple users, the proportion of each type of water use is converted into numerical features;

[0133] Step S232: A set of numerical features is constructed, and each record will contain multiple dimensions of numerical information, including timestamp, main road flow, branch flow, water use event type, water use event intensity, and the proportion of each type of water use;

[0134] Step S233: The generated numerical features are integrated into a unified feature matrix to form the final data output for model construction; each row of the matrix represents a complete water use event record at a time point, containing all related numerical features.

[0135] Preferably, step S231 of the present embodiment ensures the uniform comparability of data by numerically converting the original intensity values of each water use event; this is crucial for data analysis and model training, and helps to eliminate consistency problems between data; extracting the use situation in a specific time period and quantifying the frequency and intensity of events within the time window can reveal the changing trend of water use behavior; through the statistics of event frequency, the water use pattern can be better understood; for events involving multiple users, the proportion of various water use situations is calculated and converted, simplifying the complex multi-user water use information into easily manageable numerical features, enhancing the interpretability and practicality of the data. Significance: This lays a solid data foundation for feature generation and integration, ensuring that the features of water use events can be quantified into meaningful numerical values, providing analysis basis for the model; quantifying frequency and proportion provides a depth perspective for water use behavior analysis, supporting the analysis of water use peaks and troughs under different conditions, and providing decision support for supply management optimization. Step S232 constructs a complete set of numerical features, including timestamp, main road flow, branch flow, water use event type, water use event intensity and proportion, etc.; the features not only enrich the data structure, but also provide a diversified perspective for subsequent analysis; integrating various water use information into a single record makes each record have multi-dimensional information, simplifies the complexity of subsequent analysis, and can quickly obtain comprehensive information at each time point. Significance: By constructing multi-dimensional features, it ensures that data can be smoothly input into machine learning or statistical models; at the same time, rich data features improve the prediction ability of the model, which can capture more complex supply and demand relationships; multi-dimensional numerical features help to deeply understand the dynamic changes of water use events and support various analysis needs such as trend analysis and anomaly detection. Step S233 integrates all generated numerical features into a unified feature matrix, which is a highly organized and easy-to-manage structured data output form that facilitates subsequent analysis; it ensures that each row in the feature matrix represents a complete record of a time point, avoiding confusion and inconsistency between data and improving data usability. Significance: The output forms the basis data for adaptive flow adjustment model training, ensuring that the model can learn and predict using these accurately constructed features; through the organization of the feature matrix, subsequent data analysis, feature engineering, model training and evaluation are more efficient, improving the speed and accuracy of operation and ensuring the smoothness of the overall data processing chain.

[0136] In summary, the present embodiment realizes the effective conversion from complex raw water use data to structured numerical features, not only improving the usability of data, but also providing a solid foundation for model construction and analysis, which has important practical value and theoretical significance.

[0137] Further, the process of forming a supply-demand mapping in step S24 specifically includes the following steps:

[0138] Step S241: Extract the key features related to supply demand from the refined sample dataset, including main road flow, branch flow, type of water use event and its intensity, weather dataset time factor (such as hour, day, week), etc.; each key feature has a field identifier;

[0139] Step S242: Analyze the extracted key features, use correlation coefficient to evaluate the relationship strength between each type of flow feature and supply demand, and identify the main factors affecting supply demand; based on correlation analysis, construct a supply demand function, which predicts the output demand by input features;

[0140] The expression of supply demand function is:

[0141] D(X) = β0 + β1Q p + β2Q s + β3E + β4I + β5W + β6T

[0142] In the formula, D(X) is the output of the supply demand function, representing the supply demand under a given set of feature values X, β0 represents the intercept term, β1, β2, β3, β4, β5, β6 represent the weight coefficients, β1 is the coefficient of main road flow Q p , representing the degree of change of supply demand D when the main road flow increases by one unit, β2 is the coefficient of branch flow Q s , representing the influence of branch flow on supply demand, β3 is the coefficient of water use event type E, representing the influence strength of different types of water use events on demand, β4 is the coefficient of event intensity I, representing the influence of water use event intensity on supply demand, β5 is the coefficient of weather data W, reflecting the influence of weather conditions on supply demand, and β6 is the coefficient of time factor T, representing the influence of time variable (such as hour, day, week) on supply demand;

[0143] Step S243: Use the labeled input (flow data and event) and output (supply demand) pairs to fit and form a mapping relationship; convert the trained supply demand function into a mapping matrix, which represents the relationship between input features and output demand, each row represents the predicted output under different input conditions, providing supply demand prediction for upcoming flow conditions and water use events;

[0144] The expression of mapping matrix is:

[0145]

[0146] Preferably, step S241 of the embodiment is characterized by feature extraction, which extracts relevant key features from the sorted sample data set to ensure the comprehensiveness and consistency of the information; the key features include main road flow, branch flow, water use event type, intensity, weather data, and time factor, providing necessary data support for subsequent analysis; each key feature has a field identifier, making the use of each feature in the data processing and analysis process more explicit and reducing the complexity of data management. Significance: Feature extraction ensures that the basic data for subsequent analysis is powerful and rich, so that model training does not miss key influencing factors, improving the rationality of analysis; identifying and providing key features related to supply and demand helps improve the prediction accuracy of subsequent models and provides a basis for making scientific decisions. Step S242 is characterized by feature analysis and correlation evaluation, which uses correlation coefficients and other indicators to evaluate the relationship strength between various flow features and supply and demand, and quantifies the actual effect of influencing factors; the analysis results enable the model to extract useful signals from complex data; based on the correlation analysis, a supply and demand function is constructed, which can directly predict the output demand by input features, realizing the effective conversion from data to model. Significance: Identifying the main factors affecting supply and demand lays the foundation for subsequent quantitative analysis and model optimization, and helps to improve supply strategies in a targeted manner; by constructing a function that can predict demand, water supply managers can plan water supply strategies in advance, improving the responsiveness and flexibility of the water supply system. Step S243 is characterized by fitting and mapping relationship construction, which uses labeled input and output pairs to fit and form a reliable mapping relationship, involving linear or nonlinear regression techniques to ensure that the model can effectively capture the relationship between input features and demand; the trained supply and demand function is converted into a mapping matrix, which clearly defines the correspondence between input features and output demand, and the matrix realizes dynamic prediction under input conditions. Significance: Through the mapping matrix, the supply and demand prediction at a certain time or under certain conditions can be quickly obtained, improving the response speed of flow adjustment and optimizing resource allocation; the structured form of the mapping matrix makes performance evaluation and subsequent improvement of the model more feasible, and also facilitates comparison with actual supply and demand.

[0147] In summary, the formation process of the supply and demand mapping of the embodiment constitutes a complete supply and demand prediction chain by gradually extracting key features, analyzing relationship strength, fitting models, and constructing mapping relationships. The technical effects and significance of each step complement each other, collectively promoting the intelligentization and self-adaptation of flow adjustment and water supply management, and providing a solid foundation for efficient water resource management.

[0148] Further, the training of the adaptive flow adjustment model in step S25 specifically includes the following steps:

[0149] Step S251: configure the input layer and the output layer to ensure that each input feature such as main road flow, branch flow, and water use event intensity is mapped to the output demand; divide the sample data set into a training set and a validation set, the training set is used for parameter learning of the adaptive flow adjustment model, and the validation set is used for evaluating the prediction ability and accuracy of the adaptive flow adjustment model;

[0150] Step S252: train the adaptive flow adjustment model using the training set, calculate the error between the output of the adaptive flow adjustment model and the actual supply demand after each training period; realize error feedback adjustment for the loss function evaluated by the adaptive flow adjustment model; evaluate the accuracy of the adaptive flow adjustment model using the validation set, and obtain indicators such as accuracy and recall rate by comparing the estimated result with the real supply demand;

[0151] Step S253: after training is completed, automatically adjust the hyperparameters according to the performance indicators of the adaptive flow adjustment model.

[0152] Preferably, step S251 of the embodiment is configured to input layer and output layer and data division. By configuring the input layer and the output layer, it is ensured that various types of input features (such as main road flow, branch flow and water use event intensity) can be accurately mapped to the corresponding supply demand. This mapping process is the basis of model learning and prediction; the sample data set is reasonably divided into training set and validation set, which maintains the representativeness of the sample size and distribution, provides sufficient data support in the training and evaluation process, and reduces the risk of overfitting. Practical significance: By reasonably dividing the data, it is ensured that the performance of the model on unseen data can be reliably evaluated, and the generalization ability of the model is improved, which is not only suitable for training data, but also can handle new data in practical application; appropriate input and output design can enhance the efficiency of model learning, ensure that each iteration can effectively improve the understanding and prediction ability of the model to supply demand, and lay a good foundation for the subsequent steps. Step S252 model training and error feedback, after each training period, the error between the model output and the actual demand is calculated, and feedback adjustment is made according to the loss function, so that the model can be dynamically optimized; by using the validation set for evaluation, the accuracy, recall rate and other indicators are calculated to measure the effectiveness of the model; the accuracy and recall rate can comprehensively evaluate the performance of the model in prediction, and the model parameters can be further adjusted. Practical significance: During the entire training process, through continuous monitoring and adjustment, the model gradually reduces the gap between prediction and reality, improves the final prediction performance, and enhances the robustness of the model in complex environments; timely performance evaluation and feedback adjustment enable the model to quickly adapt to data changes during training, reducing uncertainty during training, which has a positive significance for the accuracy and stability of actual operation. Step S253 automatic adjustment of hyperparameters, after the training is completed, the model performance indicators (such as loss, accuracy, etc.) are monitored and evaluated in real time to identify the key hyperparameters that affect the performance of the model; automatically adjust the hyperparameters (such as learning rate, batch size, etc.), and realize the best combination of hyperparameters through intelligent algorithms (such as grid search, random search or Bayesian optimization) without manual intervention. Practical significance: Through automatic adjustment of hyperparameters, the flexibility of the model in practical application is improved, ensuring the adaptability of the model in different scenarios and being able to face changes in flow and demand in real time; simplifies the model optimization process, compared with manual selection of hyperparameters, the automatic way is more efficient and can better guarantee the optimal performance, ensuring that the self-adaptive flow adjustment model can continue to run effectively in complex scenarios.

[0153] In summary, the embodiment gradually shapes an efficient, flexible, and accurate self-adaptive flow adjustment model by configuring a series of steps of input and output, model training, and hyperparameter adjustment. Not only does it ensure a deep understanding of the relationship between flow and supply demand, improving prediction accuracy and response speed, but it also enables the model to adapt to changing water demand and environmental factors, making it widely applicable and potentially commercially valuable.

[0154] Further, the process of dividing the sample data set into a training set and a validation set in step S251 specifically includes the following steps:

[0155] Step S2511: Randomly shuffle the sample data set, and according to the specific situation of water consumption peak and valley, implement stratified sampling in the data set to generate a training set and a validation set;

[0156] Step S2512: After completing the division of the training set and the validation set, confirm that the number of samples in the training set and the validation set meets the set division ratio, and that the feature distribution of each subset matches the sample data set. Record the specific information of the division, including the number of samples, feature distribution, and stratified sampling of the training set and the validation set;

[0157] Step S2513: At different stages of self-adaptive flow adjustment model training, dynamically evaluate the composition of the generated training set and validation set, including the effectiveness of the samples, the representativeness of the features, and the phased effect of the training progress.

[0158] Preferably, step S2511 of the present embodiment randomly shuffles and stratifies the sample data set to ensure randomness of sample order and eliminate sequence effects in time series data by randomly shuffling the sample data set; the learning is not affected by a certain sequence of samples, thereby improving the independence and diversity of learning; stratified sampling is performed according to key features such as water consumption peaks and troughs to ensure that the proportions of different types of samples in the training set and the validation set are consistent with those in the original data set, thereby ensuring that the model can access important change points in the sample data and enhancing the learning ability and robustness of the model under different conditions. Practical significance: Stratified sampling ensures that the training set and the validation set can fully represent the entire data set, making the model more accurate and reliable in predicting actual scenarios after training; effectively utilizing limited data resources ensures that key features are fully learned, enabling the model to have a strong response strategy when facing actual problems. Step S2512 confirms and records the data set, verifies the sample size and feature distribution of the training set and the validation set after division to ensure that they meet the established division ratio and feature distribution requirements, and prevents data imbalance caused by errors in the division process; the system records specific information about the division, including the sample size, feature distribution, and stratified sampling details of the training set and the validation set, which not only provides a basis for model evaluation but also serves as an important reference for subsequent model optimization. Practical significance: A clear record process makes the model development process transparent, which is helpful for communication and cooperation, and also provides a reliable basis for subsequent research and model inheritance; during model training and evaluation, the division of the data set is consistent with the features of the original data set, thereby improving the reliability and effectiveness of the model in the validation phase. Step S2513 dynamically evaluates the composition of the data set, dynamically evaluates the training set and the validation set generated during model training to detect sample effectiveness, feature representativeness, and training effectiveness, and accumulates data for subsequent adjustment and optimization; based on the evaluation feedback during training progress, the performance of the model at a specific stage can be understood in a timely manner, thereby enabling necessary adjustments to the data set to optimize model performance. Practical significance: Dynamic evaluation ensures that the model can make continuous optimization and adjustment according to the actual situation during training, enabling the model to quickly adapt to complex and changing water demand scenarios and improving the practicality of the model; dynamic review of the composition of the model data set can improve the adaptability and predictability of the overall model, thereby achieving higher accuracy and effectiveness in actual applications.

[0159] In summary, the implementation of the present embodiment ensures that the self-adaptive flow adjustment model has a good training foundation and evaluation mechanism through precise data management, enabling the model to still perform supply and demand prediction stably, efficiently, and accurately when facing complexity in actual application. Each step complements each other to promote the improvement of model capability and the success of application.

[0160] Further, asFigure 4 As shown, the process of obtaining the operating parameters of the flow adjustment device in step S3 specifically includes the following steps:

[0161] Step S31: After receiving the supply demand, a set of control signals is generated and sent to the flow adjustment device, including the current demand parameter, target flow, target pressure, and instructions for monitoring the operating state; the flow adjustment device performs real-time state evaluation on the valves and pumps, and collects the current flow, pressure, and state information;

[0162] Step S32: Obtain the target value of the control signal, compare the current operating value of the valve and pump with the target value, and calculate the required adjustment amount; for each flow regulating valve, the opening value and conversion speed generated by the change in supply demand; adjust the speed or power output of the pump;

[0163] Step S33: The newly generated operating parameters of the valve and pump are saved to the configuration file of the device.

[0164] Preferably, step S31 of the present embodiment generates control signals that can be quickly and accurately transmitted to the flow adjustment device after receiving the supply demand, ensuring the immediacy and accuracy of the relevant instructions and assisting the flow adjustment device in comprehensively monitoring the system state in real time. Significance: A communication bridge between the flow adjustment system and the environmental demand is established; through real-time monitoring, the device can respond to changes in external conditions in a timely manner, ensuring the dynamic balance of the water supply and drainage system; the immediate feedback mechanism provides the necessary information basis for subsequent flow regulation, reducing the reaction delay of the system to unexpected events and improving the overall efficiency of the system. In step S32, the required adjustment amount can be accurately calculated by comparing the current operating value with the target value. As a result, the flow regulating valve opening value and the pump speed or power output can be precisely set, enabling the device to quickly respond to changes in supply demand. Significance: Provides clear operational guidelines for flow adjustment, ensuring that the device can make fine adjustments according to actual demand; by calculating the adjustment amount, the system reduces the risk of resource waste, optimizes energy consumption, and improves the accuracy of flow adjustment; precise control helps achieve water and energy saving goals while maintaining system stability and sustainability. In step S33, the newly generated operating parameters of the valve and pump are saved to the configuration file of the device, ensuring the persistence and consistency of the parameters, so that the device can continue to work according to the optimal configuration in subsequent operation. Significance: The step of saving operating parameters provides an important basis for long-term performance monitoring and optimization of the device; the data not only reflects the effective configuration of the device under specific historical conditions, but also lays a foundation for future adjustment and strategy formulation; regular data display and analysis can guide subsequent maintenance decisions and technology upgrades, thereby improving the intelligent level of the entire water supply and drainage system.

[0165] The calculation process for the flow regulating valve

[0166] Calculate the current valve opening and target opening, get the current parameter: first read the actual opening of the current flow regulating valve. For example, the current opening is recorded as C current (%) or other units);

[0167] Get the target parameter: receive the target flow C target from the control signal and the target opening C current obtained from the flow characteristics;

[0168] Calculate the adjustment amount, adjustment amount formula:

[0169] Calculate the required adjustment amount ΔC: ΔC = C target -C current

[0170] Determine the adjustment direction and intensity:

[0171] According to the calculation result, determine whether the valve needs to increase or decrease the opening, and evaluate the intensity of the adjustment;

[0172] Generate opening value and conversion speed, generate opening value:

[0173] Based on the adjustment amount, set a new opening value: C new = C current + ΔC

[0174] Conversion speed setting: determine the conversion speed according to the design and actual working characteristics of the valve. It can be assumed that the maximum conversion speed (V max ) is a known constant, and the final conversion speed V is estimated by the following formula: Where Δt can be set according to the specific application scenario and the change rate of flow demand.

[0175] Pump calculation process

[0176] Calculate the current speed and target speed, get the current speed:

[0177] Read the current speed N current of the pump (unit: RPM).

[0178] Get the target flow and target speed: according to the target flow C target in the control signal and the flow-speed relationship of the pump, calculate the target speed N target .

[0179] Calculate the adjustment amount

[0180] Adjustment amount formula: calculate the required speed adjustment amount ΔN: ΔN = N target -N current

[0181] Determining the adjustment direction and amplitude: according to the positive and negative values of the adjustment amount, confirming whether to increase or decrease the speed adjustment;

[0182] In summary, the present embodiment forms a complete flow adjustment mechanism from information collection to accurate calculation and data storage. The mechanism not only enhances the response capability and efficiency of the water supply and drainage system, but also plays an active role in efficient water resource management, helping to achieve a more intelligent water management solution.

[0183] As shown in Figure 5 The present embodiment also provides an embodiment of an adaptive water supply and drainage pipeline flow adjustment system, in which the adaptive water supply and drainage pipeline flow adjustment system is applied to the adaptive water supply and drainage pipeline flow adjustment method in the above embodiment, and the adaptive water supply and drainage pipeline flow adjustment system comprises a signal output module 1, a supply and demand module 2, and a parameter adjustment module 3 connected in sequence.

[0184] The signal output module 1 is used to obtain a three-dimensional perspective view of the water supply and drainage pipeline, identify the positions of devices related to flow monitoring and adjustment through the three-dimensional perspective view, and obtain a set of flow monitoring device positions and a set of flow adjustment device positions. The flow monitoring device is a flow monitoring device that collects the water flow of the water supply and drainage pipeline in real time. The collected flow simulation signal is converted into a digital signal, and the digital signal is output. The supply and demand module 2 is used to train the adaptive flow adjustment model according to the historical flow data set of the water supply and drainage pipeline. The real-time water flow data is input into the adaptive flow adjustment model to obtain the supply and demand of the water flow of the main road and branch road of the water supply and drainage pipeline. The parameter adjustment module 3 is used to adjust the flow adjustment device according to the supply and demand to obtain the operating parameters of the adjusted flow adjustment device. At the same time, the water pressure of the branch road of the water supply and drainage pipeline after adjusting the operating parameters is measured by the water pressure monitoring device. If the pressure threshold is triggered, a warning is given.

[0185] As shown in Figure 6 The present embodiment provides an embodiment of an electronic device, in which the electronic device 4 comprises a processor 41 and a memory 42 coupled to the processor 41.

[0186] The memory 42 stores program instructions for implementing the adaptive water supply and drainage pipeline flow adjustment method of any of the above embodiments.

[0187] The processor 41 is configured to execute the program instructions stored in the memory 42 to perform adaptive water supply and drainage pipeline flow adjustment.

[0188] The processor 41 can also be referred to as a CPU (Central Processing Unit). The processor 41 can be an integrated circuit chip having a processing capability for signals. The processor 41 can also be a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application-Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0189] Further, Figure 7 For a structural schematic diagram of the storage medium of an embodiment of the present application, the storage medium 5 of the embodiment of the present application stores program instructions 51 capable of implementing all the methods described above, wherein the program instructions 51 can be stored in the storage medium in the form of a software product, including a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the methods described in various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a ROM (Read-Only Memory), a RAM (Random Access Memory), a magnetic disk or an optical disk, and various media capable of storing program codes, or a terminal device such as a computer, a server, a mobile phone, a tablet, etc.

[0190] In several embodiments of the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the device embodiments described above are merely schematic, for example, the division of units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0191] In addition, each functional unit in each embodiment of the present application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit. The above is only an embodiment of the present application, and does not limit the patent scope of the present application, and any equivalent structure or equivalent process conversion using the content of the specification and drawings of the present application, or direct or indirect application in other related technical fields, are also included in the patent protection scope of the present application.

[0192] The foregoing detailed description of the application has been presented for purposes of illustration and description. It is not intended to be exhaustive or to limit the application to the precise form disclosed, and obviously many modifications and variations are possible in light of the above teaching. It is intended that the scope of the application be governed by the claims and their equivalents.

Claims

1. An adaptive flow adjustment method for water supply and drainage pipelines, characterized in that, The adaptive water supply and drainage pipeline flow adjustment method comprises: Obtain a three-dimensional perspective view of the water supply and drainage pipeline, identify the positions of devices related to flow monitoring and adjustment through the three-dimensional perspective view, obtain a set of flow monitoring device positions and a set of flow adjustment device positions; the flow monitoring device is a flow monitoring device that collects the water flow of the water supply and drainage pipeline in real time; convert the collected flow analog signal into a digital signal and output the digital signal; Train the adaptive flow adjustment model according to the historical flow data set of the water supply and drainage pipeline; input the real-time water flow data into the adaptive flow adjustment model to obtain the supply and demand of the water flow of the main road and branch road of the water supply and drainage pipeline; According to the supply and demand, adjust the flow adjustment device to obtain the operating parameters of the adjusted flow adjustment device; at the same time, measure the water pressure of the branch road of the water supply and drainage pipeline after adjusting the operating parameters through the water pressure monitoring device, and if the pressure threshold is triggered, an early warning is performed; Identify the positions of devices related to flow monitoring and adjustment through the three-dimensional perspective view, comprising: Arrange a plurality of view scanning devices in the target water supply and drainage pipeline network area, perform 360-degree rotation scanning to obtain a plurality of scanning views; at the same time, use a multi-view high-resolution camera to perform synchronous shooting to obtain texture information of the devices and the pipe surface; Integrate the plurality of scanning views into a single three-dimensional surface model to preliminarily generate point cloud data; post-process the preliminarily generated point cloud data to remove noise points and outliers beyond the distance threshold, clean the point cloud data, and obtain pre-processed point cloud data; Analyze the pre-processed point cloud data, geometrically fit the curved surface of the pipe and the surface of the valve set pump to obtain the rough shape of the device; convert the point cloud data into a raster image, extract the significant edges from the raster image, select the edge parts in the straight line direction, and gradually connect the edge points in the same direction and close to each other to form a plurality of independent straight line segments; Analyze the extracted straight line segments, set similar lengths and angle ranges for combination, aggregate the straight line segments to form a closed body representing the outline of the pipe and the device, construct a preliminary structure diagram according to the aggregated line segments as the outline feature description of the device; compare the newly generated outline feature with the device models in the database, evaluate the similarity of the shape according to the shape similarity and the direction consistency, and determine the position, angle and installation direction of the device according to the matching result; Construction of the adaptive flow adjustment model, comprising: Obtain historical flow data containing historical real-time water flow data from each flow monitoring device, including flow information of the main road and branch road, and a time stamp to form a complete flow time sequence; at the same time, record the water use events in the corresponding time period, including water use peak, device operating state change and weather condition; Organize the historical flow data according to the time sequence to form a structured database; each record should contain time, main road flow, branch road flow, related water use event and user demand information; label the sample data stored in the database, associate each set of water flow data with the corresponding supply and demand to form a pair of flow data and event and supply and demand, and obtain a sample data set; Extracting features including daily, weekly, monthly trends and fluctuation amplitudes of water flow from the sample dataset; converting the intensity and type of water usage events into numerical features; Building a supply-demand relationship table using the sorted sample dataset; correlating the change patterns of main and branch flow with various types of water usage events; creating a supply-demand function, and forming a supply-demand mapping by statistically analyzing the correlation between the main flow features and supply-demand; Training the adaptive flow adjustment model using the training set divided from the sample dataset, and evaluating the adaptive flow adjustment model using the validation set divided from the sample dataset to predict the accuracy of supply-demand under specific conditions. The accuracy is obtained by comparing the output results of the adaptive flow adjustment model with the results of the supply-demand mapping, and the error of the adaptive flow adjustment model parameters is determined according to the accuracy.

2. The self-adapting water line flow adjustment method of claim 1, wherein, Integrating multiple scan views into a single three-dimensional surface model, including: At each scan position, the view scanning device generates a point cloud dataset containing the three-dimensional coordinates and intensity information of the objects around the position; selecting a reference point as the origin of the global coordinate system, and defining the directions of X, Y, and Z coordinate axes according to the far point; Detecting feature points of edges and corners in each point cloud; selecting feature points in the overlapping area of each dataset to align two adjacent point clouds; registering multiple adjacent point cloud data through rigid transformation; calculating the rotation matrix and translation vector of each point cloud relative to the reference coordinate system through an iterative method based on the least squares method, and using the overlapping parts of multiple point clouds for multiple iterations until the error reaches the tolerance threshold compared with the set tolerance threshold; Merging the registered point cloud datasets according to the rotation matrix and translation vector, transforming the coordinates of each point cloud data to the global coordinate system, and forming a point cloud dataset; performing triangular meshing processing on the merged point cloud dataset to convert the point cloud dataset into a three-dimensional surface model.

3. The self-adapting water line flow adjustment method of claim 1, wherein, Obtaining a sample dataset, including: Obtaining a historical dataset containing main and branch water flow from the flow monitoring device, each record having a timestamp, main flow, and branch flow fields; extracting information matching the corresponding timestamp from the stored water usage event data, including water usage peak, device state change, and weather information; Cleaning the flow data and event data in the historical dataset to remove missing values and outliers, and converting all timestamps to a unified time format; connecting the flow data with the corresponding water usage event data based on the timestamp; Generating a structured data frame whose rows represent water flow records at each time point and columns include time, main flow, branch flow, and corresponding water usage event information; for each flow data, creating a supply-demand label according to the type and intensity of water usage events, marking water flow during the water usage peak period as high demand and during the low peak period as low demand; arranging the processed flow data and corresponding supply-demand labels into a sample dataset in table form, containing flow-related input fields and output fields.

4. The self-adapting water line flow adjustment method of claim 1, wherein, Converting the intensity and type of water usage events into numerical features, including: For each water use event collected, numerical conversion is carried out according to its original intensity value and classification information; the use of water in the time period is extracted, and the frequency and intensity of the event in the time window are quantified and numbered; for events involving multiple users, the proportion of each type of water use is converted into a numerical feature; A set of numerical features is constructed, and each record will contain multiple dimensions of numerical information, including timestamp, main flow, branch flow, water use event type, water use event intensity, and the proportion of each type of water use; The generated numerical features are integrated into a unified feature matrix to form the final data output for model construction; each row of the matrix represents a complete water use event record at a time point, including all related numerical features.

5. The self-adapting water line flow adjustment method of claim 1, wherein, A supply-demand mapping is formed, including: Extracting key features related to supply and demand from the sorted sample data set, including main flow, branch flow, water use event type and intensity, and weather data set time factors; each key feature has a field identifier; Analyze the extracted key features, use correlation coefficients to evaluate the relationship strength between each type of flow feature and supply and demand, and identify the main factors affecting supply and demand; based on correlation analysis, a supply and demand function is constructed, which predicts the output demand by input features; Use the labeled input and output pairs to fit, form a mapping relationship; convert the trained supply and demand function into a mapping matrix, which represents the relationship between input features and output demand, and each row represents the predicted output under different input conditions, providing supply and demand prediction of upcoming flow conditions and water use events.

6. The self-adapting water line flow adjustment method of claim 1, wherein, Training the adaptive flow adjustment model, including: Configure the input layer and output layer to ensure that each main flow, branch flow, and water use event intensity input feature is mapped to the output demand; divide the sample data set into training set and validation set, the training set is used for parameter learning of the adaptive flow adjustment model, and the validation set is used to evaluate the predictive ability and accuracy of the adaptive flow adjustment model; Use the training set to train the adaptive flow adjustment model, calculate the error between the output of the adaptive flow adjustment model and the actual supply and demand after each training period; realize error feedback adjustment for the loss function evaluated by the adaptive flow adjustment model; use the validation set to evaluate the accuracy of the adaptive flow adjustment model, and obtain the accuracy and recall rate indicators by comparing the estimated results with the real supply and demand; After training is completed, automatically adjust the hyperparameters according to the performance indicators of the adaptive flow adjustment model.

7. The self-adapting water line flow adjustment method of claim 1, wherein, Get the operating parameters of the adjusted flow adjustment device, including: After receiving the supply and demand, generate a set of control signals and send them to the flow adjustment device, including the current demand parameters, target flow, target pressure, and operation state monitoring instructions; the flow adjustment device performs real-time state evaluation on the valve and pump, and collects the current flow, pressure, and state information; The target value of the control signal is obtained, the current operating value of the valve and the pump is compared with the target value, and the required adjustment amount is calculated; for each flow regulating valve, the opening value and the conversion speed are generated according to the supply demand change; the operating state of the pump is adjusted to change the rotating speed or the power output; The newly generated operating parameters of the valve and the pump are saved in the configuration file of the device.

8. An adaptive water supply and sewer pipeline flow adjustment system for use in the adaptive water supply and sewer pipeline flow adjustment method according to any one of claims 1 to 7, characterized by The adaptive water supply and drainage pipeline flow adjustment system comprises: A signal output module is configured to obtain a three-dimensional image of the water supply and drainage pipeline, identify the positions of devices related to flow monitoring and adjustment based on the three-dimensional image, obtain a set of positions of flow monitoring devices and a set of positions of flow adjustment devices, collect flow analog signals, convert the flow analog signals into digital signals, and output the digital signals. A supply demand module is configured to train an adaptive flow adjustment model based on historical flow data sets of the water supply and drainage pipeline, input real-time water flow data into the adaptive flow adjustment model, and obtain the supply demand for the water flow of the main line and branch lines of the water supply and drainage pipeline. A parameter adjustment module is configured to adjust the flow adjustment devices based on the supply demand, obtain the operating parameters of the adjusted flow adjustment devices, and measure the water pressure of the branch lines of the water supply and drainage pipeline after the adjustment of the operating parameters through a water pressure monitoring device, and perform early warning if a pressure threshold is triggered.

Images