A Multi-Source Sensing-Based Anomaly Identification System and Method for Sponge City Facilities

Abstract

This invention discloses a sponge city facility anomaly identification system and method based on multi-source sensing. The system includes a multi-source sensing unit, a telemetry terminal unit, a wireless communication unit, an AIoT urban sensing platform, an anomaly identification module, and a smart platform for comprehensive water environment management. The multi-source sensing unit collects water level, flow rate, video images, and meteorological data; the telemetry terminal unit aggregates and performs preliminary data processing; the wireless communication unit uploads the data to the AIoT urban sensing platform; the AIoT urban sensing platform performs time synchronization, data tagging, and rainfall event identification; the anomaly identification module determines the anomaly type and suspected location, such as blockage, overflow, stagnation, and insufficient drainage capacity, based on the water level-flow response relationship and video recognition results; the smart platform for comprehensive water environment management displays the anomaly information and pushes early warnings. This invention reduces false alarms based on single thresholds through multi-source data correlation analysis, detects facility operation problems in advance, and improves the operation and maintenance efficiency of sponge cities.

Description

Technical Field

[0001] This invention relates to the fields of sponge city facility monitoring, urban drainage facility operation and maintenance, and intelligent water environment management, and particularly to a sponge facility anomaly identification system and method based on multi-source sensing. Background Technology

[0002] With the acceleration of urbanization, the area of ​​impermeable surfaces around urban roads, squares, and buildings is constantly increasing, leading to increased surface runoff and faster runoff velocity during rainfall. This easily causes problems such as road flooding, overloaded pipe networks, and localized waterlogging. The concept of sponge cities aims to improve the city's ability to cope with rainfall and stormwater events through measures such as infiltration, retention, regulation, purification, and discharge.

[0003] Roadside storm drains, sunken green spaces, stormwater storage facilities, drainage outlets, overflow points, and adjacent pipe networks are important components of sponge city stormwater infrastructure. However, existing sponge city stormwater infrastructure still faces the following limitations in operation monitoring and maintenance management: 1. Insufficient data correlation: Water level, flow rate, video and meteorological data are mostly collected and displayed independently, lacking correlation judgment of multi-source data under the same rainfall process.

[0004] 2. Limited anomaly detection: Existing systems rely heavily on alarms triggered by water level exceeding limits, abnormal flow rates, or video surveillance, which can easily lead to false alarms or missed alarms.

[0005] 3. Delayed blockage detection: Problems such as blocked rainwater inlets, clogged pipes and channels, and poor water flow are often only discovered after water accumulation becomes obvious, lacking the ability to detect them in advance.

[0006] 4. Insufficient on-site verification: Video monitoring results are usually not linked with water level, flow rate and meteorological data for analysis, and the cause of anomalies still needs to be judged manually.

[0007] 5. Low operation and maintenance efficiency: Existing platforms mostly focus on data display and threshold alarms, making it difficult to directly provide anomaly types, suspected locations, and handling suggestions.

[0008] Therefore, how to integrate multi-source sensing data such as water level, flow rate, video, and meteorology to identify, locate, and warn of abnormal operation of sponge city facilities is a technical problem that needs to be solved by those skilled in the art. Summary of the Invention

[0009] This invention provides a sponge city facility anomaly identification system and device based on multi-source sensing, which can solve the problems of fragmented multi-source data, single basis for anomaly judgment, delayed blockage identification, insufficient on-site verification, and low operation and maintenance efficiency in existing sponge city facility monitoring technologies.

[0010] The first aspect of this invention provides a sponge city facility anomaly identification system based on multi-source sensing, comprising: Multi-source sensing units are used to collect water level data, flow data, video image data, and meteorological data during the operation of sponge city facilities; The telemetry terminal unit, connected to the multi-source sensing unit, is used to aggregate, cache, and perform preliminary processing of water level data, flow data, video image data, and meteorological data; The wireless communication unit is used to upload the data processed by the telemetry terminal unit to the cloud. The AIoT city sensing platform is used to receive data uploaded by wireless communication units and to perform time synchronization, missing data marking, abnormal change marking and state transition of water level data, flow data, video image data and meteorological data. At the same time, it can identify rainfall events based on meteorological data. The anomaly identification module is used to determine the degree of matching between water level changes and flow response based on the correlation between water level data, flow data, video image data and meteorological data, and to determine the anomaly type and suspected location of the sponge city facility by combining the visual anomaly status of video recognition. The intelligent platform for comprehensive water environment management is used to display the anomaly type, suspected location, anomaly level, triggering basis, and handling suggestions output by the anomaly identification module, and to push early warning information to management personnel.

[0011] Optionally, the multi-source sensing unit includes: Water level monitoring units are deployed at key nodes of sponge city facilities, storage spaces, drainage outlets, overflow control points, or areas prone to water accumulation to collect water level data. Flow monitoring units are installed in rainwater pipes, drainage outlets, or water outlet sections to collect flow data; The video monitoring unit is deployed at locations covering water inlets, waterlogged areas on roads, drainage outlets, or overflow nodes. It is used to collect video image data and identify one or more visual abnormal conditions, such as water accumulation, floating objects, water seepage, overflow, abnormal outflow from outlets, or equipment obstruction. Meteorological monitoring units are deployed near the monitoring area to collect rainfall and rainfall intensity data.

[0012] Optionally, the AIoT city sensing platform is also used for: When the cumulative rainfall within a preset time period exceeds the first rainfall threshold, or the instantaneous rainfall intensity exceeds the second rainfall threshold, a rainfall event is determined to occur. The rainfall event is considered to have ended when the cumulative rainfall within a preset time period is less than the third rainfall threshold.

[0013] Optionally, the anomaly detection module is used for: Calculate the rate of change of water level and the rate of change of flow rate; When the rate of change of water level is greater than the water level rise rate threshold for two or more consecutive monitoring cycles, and the rate of change of flow is less than the flow response threshold or less than 0, it is determined to be insufficient drainage response or abnormal blockage.

[0014] Optionally, the anomaly detection module is also used for: During rainfall events, when the water level continues to rise but the flow rate does not increase synchronously, and the video monitoring unit detects the accumulation of floating debris at the inlet or the expansion of local water accumulation, it is determined to be an abnormal blockage, and the corresponding inlet or its downstream drainage channel is marked as a suspected location.

[0015] Optionally, the anomaly detection module is also used for: When the water level exceeds the preset warning level, the flow rate reaches the preset high level or fluctuates, and the video monitoring unit detects water seepage, overflow, or an expansion of the road water accumulation area, it is determined to be an overflow anomaly.

[0016] Optionally, the anomaly detection module is also used for: After a rainfall event ends, if the water level fails to drop to the normal range within the preset receding time and the flow rate remains low, it is judged as an abnormal discharge stagnation. When the flow rate increases but the water level continues to rise, and the rate of rise in the water level exceeds a preset threshold, it is determined to be an abnormality of insufficient drainage capacity.

[0017] A second aspect of the present invention provides a method for anomaly identification of sponge city facilities based on multi-source sensing, comprising: Acquire water level data, flow data, video image data, and meteorological data during the operation of sponge city facilities; The system performs time synchronization, missing data marking, abnormal change marking, and state transition on water level data, flow data, video image data, and meteorological data. It also identifies rainfall events based on meteorological data and determines the start time, duration, intensity level, and end time of rainfall. Based on water level data and flow rate data, determine the matching relationship between water level changes and flow rate response; Based on video image data, identify visually abnormal conditions around the sponge city facilities. These visually abnormal conditions include one or more of the following: water accumulation, floating debris accumulation, inlet blockage, well overflow, abnormal discharge from outlets, and overflow. By correlating the matching results of water level changes and flow response with visible anomalies, the anomaly types and suspected locations of the sponge city facilities can be determined. Output the anomaly type, suspected location, anomaly level, triggering basis, and handling suggestions.

[0018] A third aspect of the present invention provides a sponge city facility anomaly identification device based on multi-source sensing, comprising: One or more processors; A memory on which one or more programs are stored; When the one or more programs are executed by the one or more processors, the one or more processors enable the one or more processors to perform the functions of the sponge facility anomaly identification system based on multi-source sensing as described above.

[0019] A fourth aspect of the present invention provides a computer storage medium for storing a program, which, when executed, performs the function of the sponge city facility anomaly identification system based on multi-source sensing as described in any of the preceding claims.

[0020] Compared with the prior art, the present invention has the following beneficial effects: Multi-source data correlation analysis improves judgment accuracy: Water level, flow rate, video and meteorological data are organized according to the same time base, and the matching degree of water level-flow rate response is judged in combination with the rainfall process, reducing false alarms and false negatives of single threshold.

[0021] Early detection of blockages and other anomalies: By identifying signs of asynchronous water level rise and flow, the system can detect the trend of obstructed drainage channels before significant water accumulation occurs, thus providing early warning.

[0022] Video verification reduces false alarm rate: Video recognition results are used to verify hydraulic anomalies, avoiding false alarms caused by instantaneous fluctuations in sensors or external interference.

[0023] Clearly define anomaly location and handling recommendations: Output anomaly type, suspected location, triggering basis, and handling recommendations to narrow down the scope of on-site investigation and improve operation and maintenance efficiency.

[0024] Seamless integration with smart platforms: The system architecture supports integration with existing urban water environment management platforms, forming a closed loop of early warning push, on-site handling, and operation and maintenance records. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 A schematic diagram of a sponge city facility anomaly identification system based on multi-source sensing provided in an embodiment of the present invention; Figure 2 A flowchart illustrating an anomaly identification method for sponge city facilities based on multi-source sensing, provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of a device provided in an embodiment of the present invention. Detailed Implementation

[0027] This invention provides a sponge city facility anomaly identification system and method based on multi-source sensing, which is used for the operation monitoring of urban road storm drains, sunken green spaces, storage facilities, drainage outlets, overflow nodes and their adjacent pipes and channels. By performing correlation judgment on water level, flow rate, video and meteorological data under the same time reference, it identifies operational problems such as blockage, water accumulation, overflow, stagnation, insufficient drainage capacity and abnormal monitoring equipment.

[0028] See Figure 1 The figure is a schematic diagram of a sponge facility anomaly identification system based on multi-source sensing provided in an embodiment of the present invention.

[0029] The sponge city facility anomaly identification system based on multi-source sensing provided in this invention includes: a multi-source sensing unit, a telemetry terminal unit, a wireless communication unit, an AIoT urban sensing platform, an anomaly identification module, and a smart platform for comprehensive water environment management.

[0030] The multi-source sensing unit is used to collect water level data, flow data, video image data, and meteorological data during the operation of the sponge city facility.

[0031] In this embodiment of the invention, the multi-source sensing unit includes a water level monitoring unit, deployed at key nodes of the sponge city facility, storage space, drainage outlet, overflow control point, or easily flooded area, for collecting water level data; a flow monitoring unit, deployed at rainwater pipes, drainage outlets, or outlet sections, for collecting flow data; a video monitoring unit, deployed at locations covering water inlets, road flooded areas, drainage outlets, or overflow nodes, for collecting video image data and identifying one or more visually abnormal states such as water accumulation, floating objects, seepage, overflow, abnormal outflow from outlets, or equipment obstruction; and a meteorological monitoring unit, deployed near the monitoring area, for collecting rainfall and rainfall intensity data.

[0032] Specifically, the front-end multi-source sensing units include water level monitoring units, flow monitoring units, video monitoring units, and meteorological monitoring units. Water level monitoring units are deployed in areas prone to flooding, storage spaces, drainage nodes, or warning water level control points; flow monitoring units are deployed at water outlet sections, stormwater pipes, or drainage outlets; video monitoring units cover water inlets, road flooding areas, drainage outlets, or overflow nodes; and meteorological monitoring units are used to acquire meteorological data such as rainfall and rainfall intensity. The water level monitoring units, flow monitoring units, video monitoring units, and meteorological monitoring units are each connected to the telemetry terminal unit.

[0033] Water level monitoring units employ radar, ultrasonic, or pressure level gauges, deployed at key nodes of the sponge city infrastructure, storage spaces, drainage outlets, or areas prone to water accumulation. Flow monitoring units are installed in stormwater pipes, drainage outlets, or outlet sections to acquire flow velocity, flow rate, or flow status data. Video monitoring units utilize high-definition network dome cameras or fixed cameras, covering inlets, drainage outlets, areas prone to water accumulation, and overflow points. Meteorological monitoring units collect at least rainfall and rainfall intensity data, and may also collect wind speed, wind direction, temperature, and humidity data, used to determine the start time, duration, intensity level, and end time of rainfall events.

[0034] The telemetry terminal unit, connected to the multi-source sensing unit, is used to aggregate, cache, and perform preliminary processing of water level data, flow data, video image data, and meteorological data.

[0035] In this embodiment of the invention, the telemetry terminal unit collects, aggregates, caches, and performs preliminary processing on the front-end monitoring data, and uploads it to the AIoT city sensing platform through the wireless communication unit.

[0036] The telemetry terminal unit is equipped with a local cache module and an abnormal trigger upload module; local caching is performed when communication is interrupted, and uploading continues after communication is restored; the upload frequency is increased when the monitored value is abnormal.

[0037] The wireless communication unit is used to upload the data processed by the telemetry terminal unit to the cloud.

[0038] In this embodiment of the invention, the wireless communication unit may adopt 4G / 5G communication, or NB-IoT, LoRa or wired network depending on the site conditions.

[0039] The AIoT city sensing platform is used to receive data uploaded by wireless communication units and to perform time synchronization, missing data marking, abnormal change marking, and state transition of water level data, flow data, video image data, and meteorological data. At the same time, it identifies rainfall events based on meteorological data.

[0040] In this embodiment of the invention, when the cumulative rainfall within a continuous preset time is greater than a first rainfall threshold, or the instantaneous rainfall intensity is greater than a second rainfall threshold, a rainfall event is determined to have occurred; when the cumulative rainfall within a continuous preset time is less than a third rainfall threshold, the rainfall event is determined to have ended.

[0041] Specifically, the AIoT city sensing platform accesses multi-source data, synchronizes time, marks outliers, performs state transitions and identifies correlations, and sends the anomaly identification results to the smart platform for comprehensive water environment management.

[0042] The AIoT city sensing platform includes a data preprocessing module, a time synchronization module, a rainfall event recognition module, a water level-flow response judgment module, a video verification module, and a result output module.

[0043] After receiving water level data, flow rate data, video recognition results, and meteorological data, the system first performs time synchronization and data status marking. For data with different sampling frequencies, the system organizes them according to a unified time window, ensuring that changes in water level, flow rate, rainfall intensity, and video visual anomalies are mapped to the same analysis period. For missing data, communication interruption data, short-term sudden changes, or long-term unchanged data, the system marks them and, if necessary, will not use them as direct evidence for anomaly determination.

[0044] The time synchronization error of multi-source data is controlled within 60 seconds. The sampling interval for water level and flow rate data is 10-30 minutes under non-rainfall conditions, 1-5 minutes under rainfall conditions, and can be adjusted to 10-60 seconds under abnormal conditions. Through this processing method, the originally scattered monitoring data is converted into water level rise status, water level fall status, flow rate response status, rainfall status, video visual abnormal status, and equipment abnormal status.

[0045] Based on water level and flow rate data, the water level-flow response relationship is determined. This relationship includes the rate of change of water level and the rate of change of flow rate. The system constructs the water level change rate as follows:

[0046] in, This represents the current water level. The water level at the previous monitoring time. For monitoring time intervals, This represents the rate of change in water level.

[0047] System build traffic change rate:

[0048] in, The current flow rate. The flow rate at the previous monitoring time. For monitoring time intervals, This represents the rate of change in flow rate.

[0049] When the rate of change of water level For two or more consecutive monitoring periods, the water level rise rate exceeds the threshold Vh0, and the flow rate change rate... Less than the flow response threshold If the value is less than 0, it is determined to be insufficient drainage response or abnormal blockage. The rate of change of water level and the rate of change of flow are compared with their respective thresholds, without performing direct ratio calculations due to inconsistencies in dimensions.

[0050] The anomaly identification module is used to determine the degree of matching between water level changes and flow response based on the correlation between water level data, flow data, video image data and meteorological data, and to determine the anomaly type and suspected location of the sponge city facility by combining the visual anomaly status identified by video recognition.

[0051] In this embodiment of the invention, the water level change rate and the flow rate change rate are calculated; when the water level change rate is greater than the water level rise rate threshold for two or more consecutive monitoring cycles, and the flow rate change rate is less than the flow response threshold or less than 0, it is determined that the drainage response is insufficient or the blockage is abnormal.

[0052] During rainfall events, when the water level continues to rise but the flow rate does not increase synchronously, and the video monitoring unit detects the accumulation of floating debris at the inlet or the expansion of local water accumulation, it is determined to be an abnormal blockage, and the corresponding inlet or its downstream drainage channel is marked as a suspected location.

[0053] When the water level exceeds the preset warning level, the flow rate reaches the preset high level or fluctuates, and the video monitoring unit detects water seepage, overflow, or an expansion of the road water accumulation area, it is determined to be an overflow anomaly.

[0054] After a rainfall event ends, if the water level fails to drop to the normal range within a preset receding time and the flow rate remains low, it is judged as an abnormality of stagnant drainage; if the flow rate increases but the water level continues to rise, and the rate of rise in water level exceeds a preset threshold, it is judged as an abnormality of insufficient drainage capacity.

[0055] Specifically, rainfall events are identified based on meteorological data, and the response relationship between water level and flow rate is determined at different stages of rainfall and flooding. During rainfall, if the rainfall intensity increases and the water level continues to rise while the flow rate does not increase synchronously, it indicates that the inlet or outlet channels may be blocked. After the rainfall ends, if the water level does not return to the normal range within a preset time and the flow rate remains low, it indicates that there may be stagnation in the facilities or poor drainage downstream. If the water level approaches or exceeds the warning level, and the flow rate reaches a high level or fluctuates significantly, it indicates an overflow risk. If the flow rate has increased but the water level is still rising rapidly, it indicates that the existing drainage capacity may be insufficient.

[0056] The system determines that a rainfall event has occurred when the cumulative rainfall exceeds 0.2–0.5 mm within 5–10 minutes, or the instantaneous rainfall intensity exceeds 0.5–1.0 mm / h; the system determines that the rainfall event has ended when the cumulative rainfall is less than 0.2 mm within 30–60 minutes. These values ​​can be adjusted according to local rainfall patterns, facility size, and monitoring accuracy.

[0057] In one implementation of this invention, video results are used to verify hydraulic anomalies and output corresponding handling suggestions. The video monitoring unit is used to identify conditions such as increased road surface water accumulation, accumulation of floating debris at water inlets, water overflowing from wells, abnormal outflow from drainage outlets, overflows, abnormal water turbidity, camera obstruction, or equipment damage. When the water accumulation area, floating debris accumulation area, water overflow area, or overflow area appears consecutively for 2 to 3 recognition cycles in the video recognition results, and the recognition confidence level is greater than 70% to 85%, the system uses this result as the basis for anomaly verification.

[0058] When water level and flow data indicate a risk of blockage, overflow, or stagnation, the system retrieves the video recognition results for the corresponding area for verification. If the video recognition results are consistent with the hydraulic response anomaly, the system raises the anomaly level; if no obvious anomaly is identified in the video, the system maintains its monitoring status and continues to encrypt the monitoring to avoid triggering an alarm directly due to a single sensor error.

[0059] Anomaly locations are determined based on water level monitoring points, flow monitoring points, video coverage, and the spatial relationship of facilities. For example, if the upstream water level continues to rise but the downstream flow does not increase synchronously, and the video detects floating debris accumulation at a storm drain, the system marks that storm drain or its short-distance downstream drainage channel as a suspected blockage location. The platform then outputs the anomaly type, suspected location, warning level, and handling suggestions. Blockage anomalies correspond to cleaning the inlet, inspecting the drainage pipes, or clearing the downstream channel; overflow anomalies correspond to inspecting the overflow outlet, initiating emergency drainage, or taking temporary diversion measures; stagnation anomalies correspond to verifying downstream drainage conditions, clearing silt, or inspecting the outlet of the storage space. In one implementation of this invention, the anomaly level includes a state of concern, a state of warning, and a state of alarm. When a single data source reaches 70% to 80% of the corresponding threshold, it is determined to be in a state of concern. When two or more data sources reach 80% to 90% of the corresponding threshold, it is determined to be in a state of warning. When three or more data sources reach more than 90% of the corresponding threshold, or when video identification shows obvious water seepage, overflow, or expansion of road water accumulation, it is determined to be in a state of alarm.

[0060] The anomaly identification results include anomaly type, suspected location, anomaly level, triggering data source, triggering time, and handling recommendations. The anomaly types include blockage anomaly, overflow anomaly, stagnation anomaly, insufficient drainage capacity anomaly, and monitoring equipment anomaly.

[0061] The intelligent platform for comprehensive water environment management is used to display the anomaly type, suspected location, anomaly level, triggering basis, and handling suggestions output by the anomaly identification module, and to push early warning information to management personnel.

[0062] In this embodiment of the invention, the intelligent platform for comprehensive water environment management is used to display monitoring data, anomaly types, suspected locations, warning levels, triggering criteria, and handling suggestions.

[0063] The intelligent platform for comprehensive water environment management includes modules for anomaly display, early warning push, handling suggestions, and operation and maintenance records, which form a closed-loop management system for anomaly identification, early warning push, on-site handling, and historical records.

[0064] Now combined Figure 1 For example: Example 1: Blockage Anomaly Identification Process During a rainfall event, the meteorological monitoring unit collected a cumulative rainfall of 0.6 mm over 10 minutes, and the AIoT city sensing platform determined that a rainfall event had occurred. At the same time, the water level upstream of a road drain inlet rose for two consecutive monitoring cycles, and the rate of change of the water level was greater than the preset threshold for the rate of increase of the water level; the flow rate at the corresponding drain outlet did not increase synchronously with the rise of the water level, or the rate of change of the flow rate was less than 0.

[0065] The system further utilizes video recognition results from the vicinity of the storm drain. If the video monitoring unit detects an accumulation of floating debris such as leaves and plastic bags near the inlet, or if the area of ​​localized water accumulation continues to expand, and if the recognition results show two consecutive recognition cycles with a confidence level greater than 80%, the AIoT city sensing platform determines that the facility has an abnormal blockage and marks the storm drain or its downstream short-distance drainage channel as a suspected blockage location. The smart platform for comprehensive water environment management then sends on-site cleanup and verification prompts to maintenance personnel.

[0066] Example 2: Process for identifying overflow anomalies and insufficient drainage capacity When the water level monitoring unit detects that the water level has reached more than 90% of the preset warning level, the flow monitoring unit detects that the drainage flow has reached a high level or has significant fluctuations, and the video monitoring unit detects water gushing from the wellhead, overflowing from the outlet, or an expansion of the road water accumulation area, the AIoT city sensing platform determines that there is an overflow anomaly in the facility.

[0067] If the flow rate has increased significantly, but the water level continues to rise rapidly, and the video shows that the water accumulation area is still expanding, the system determines that the existing drainage capacity may be insufficient. The intelligent water environment management platform displays the location and affected area based on the identification results and pushes suggestions for handling the situation, such as checking the overflow outlet, clearing downstream pipes, initiating temporary drainage, or taking diversion measures.

[0068] Example 3: The process of identifying stagnation anomalies After rainfall ends, if the meteorological monitoring unit collects data showing a cumulative rainfall of less than 0.2 mm over a continuous period of 30–60 minutes, the AIoT city sensing platform determines that the rainfall event has ended. If, within the preset receding time, the water level monitoring unit collects data showing that the water level in the storage space or submerged green space has not yet dropped to the normal water level range, and the flow monitoring unit collects data showing that the outflow rate remains consistently low, the system determines that there is an abnormality in the facility's drainage.

[0069] Based on the abnormal discharge results, the platform will push disposal suggestions such as checking downstream drainage channels, cleaning inlet and outlet water outlets, checking the siltation of the water outlet of the water storage space, or reviewing the status of relevant valves.

[0070] Compared with the prior art, the present invention has the following advantages: The system enables the establishment of correspondences between multi-source data. It organizes water level, flow rate, video, and meteorological data according to the same time base and makes correlation judgments based on the spatial relationships of facilities, reducing the problems of isolated display of monitoring data and difficulty in interpreting the causes of anomalies.

[0071] The anomaly detection more closely approximates the actual operation of the facility. The system first identifies rainfall events, then determines whether the water level and flow rate have generated reasonable responses, and uses video results for verification, which can reduce false alarms and missed alarms caused by single water level or flow rate thresholds.

[0072] It can identify a variety of common operation and maintenance problems. The system can identify conditions such as inlet blockage, drainage channel obstruction, overflow risk, post-rain drainage stagnation, insufficient drainage capacity, and abnormal monitoring equipment, and is suitable for the daily operation and inspection and management of sponge city facilities during rainfall.

[0073] This helps to narrow down the scope of on-site investigation. The system outputs the anomaly type, suspected location, triggering basis, and warning level, allowing maintenance personnel to prioritize checking the corresponding rainwater inlets, outlets, overflow nodes, or downstream drainage channels based on the platform's prompts.

[0074] It can be integrated with existing smart platforms. The system adopts an architecture of telemetry terminals, wireless communication, AIoT urban sensing platform and water environment comprehensive management smart platform, which facilitates access to existing urban water environment management platforms and forms a closed loop of early warning push, on-site handling and operation and maintenance records.

[0075] See Figure 2 This figure is a flowchart illustrating a method for identifying anomalies in sponge city facilities based on multi-source sensing, provided by an embodiment of the present invention. The method for identifying anomalies in sponge city facilities based on multi-source sensing provided by this embodiment of the present invention can be implemented, for example, through the following steps S201-206.

[0076] S201: Acquire water level data, flow data, video image data, and meteorological data during the operation of sponge city facilities.

[0077] S202: Perform time synchronization, missing data marking, abnormal change marking, and state transition on water level data, flow data, video image data, and meteorological data. Based on the meteorological data, identify rainfall events and determine the start time, duration, intensity level, and end time of rainfall.

[0078] S203: Based on water level data and flow rate data, determine the matching relationship between water level changes and flow rate response.

[0079] S204: Identify visually abnormal conditions around sponge city facilities based on video image data.

[0080] In this embodiment of the invention, the visible abnormal state includes one or more of the following: water accumulation, floating debris accumulation, inlet blockage, wellhead overflow, abnormal outlet flow, and overflow.

[0081] S205: Perform correlation analysis between the matching results of water level changes and flow response and the visible abnormal conditions to determine the abnormal type and suspected location of the sponge city facility.

[0082] S206: Output the anomaly type, suspected location, anomaly level, triggering basis, and handling suggestions.

[0083] Since the multi-source sensing-based sponge facility anomaly identification method corresponds to the multi-source sensing-based sponge facility anomaly identification system provided in the above system embodiments, the specific implementation of each step of the multi-source sensing-based sponge facility anomaly identification method is based on the same concept as the above system embodiments. Therefore, for the specific implementation of each step of the multi-source sensing-based sponge facility anomaly identification method, please refer to the description section of the multi-source sensing-based sponge facility anomaly identification method system in the above method embodiments, and it will not be repeated here.

[0084] This invention also provides a sponge city facility anomaly identification device based on multi-source sensing, the device comprising: a processor and a memory; The memory is used to store instructions; The processor is used to execute the instructions in the memory to perform the functions of the sponge facility anomaly identification system based on multi-source sensing mentioned in the above embodiments.

[0085] It should be noted that the hardware structure of the sponge city facility anomaly identification device based on multi-source sensing provided in the embodiments of the present invention can be as follows: Figure 3 The structure shown, Figure 3 This is a schematic diagram of the structure of a device provided in an embodiment of the present invention.

[0086] Please see Figure 3 As shown, device 300 includes: processor 310, communication interface 320, and memory 330. The number of processors 310 in device 300 can be one or more. Figure 3 Taking a processor as an example, in this embodiment of the invention, the processor 310, communication interface 320, and memory 330 can be connected via a bus system or other means. Figure 3 Taking the connection between China and Israel via bus system 340 as an example.

[0087] Processor 310 may be a central processing unit (CPU), a network processor (NP), or a combination of a CPU and an NP. Processor 310 may further include hardware chips. These hardware chips may be application-specific integrated circuits (ASICs), programmable logic devices (PLDs), or combinations thereof. The PLD may be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), generic array logic (GAL), or any combination thereof.

[0088] The memory 330 may include volatile memory, such as random-access memory (RAM); the memory 330 may also include non-volatile memory, such as flash memory, hard disk drive (HDD) or solid-state drive (SSD); the memory 330 may also include a combination of the above types of memory.

[0089] Optionally, the memory 330 stores an operating system and programs, executable modules, or data structures, or subsets thereof, or extended sets thereof. The programs may include various operation instructions for implementing various operations. The operating system may include various system programs for implementing various basic services and handling hardware-based tasks. The processor 310 can read the programs in the memory 330 to implement the data acquisition method provided in this embodiment of the invention.

[0090] The bus system 340 can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The bus system 340 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 3The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0091] This invention also provides a computer-readable storage medium including instructions that, when executed on a computer, cause the computer to perform the multi-source sensing-based sponge facility anomaly identification method mentioned in the above embodiments.

[0092] This invention also provides a computer program product containing instructions that, when run on a computer, causes the computer to execute the sponge facility anomaly identification method based on multi-source sensing mentioned in the above embodiments.

[0093] Although the invention has been specifically shown and described in conjunction with preferred embodiments, those skilled in the art will understand that various changes in form and detail may be made to the invention without departing from the spirit and scope of the invention as defined in the appended claims, all of which shall be within the scope of protection of the invention.

Claims

1. A multi-source perception based sponge facility anomaly identification system, characterized in that, The method comprises the following steps: a multi-source sensing unit is used to collect water level data, flow data, video image data and meteorological data during the operation of the sponge facility; a telemetry terminal unit is connected to the multi-source sensing unit and is used to aggregate, cache and preliminarily process the water level data, the flow data, the video image data and the meteorological data; a wireless communication unit is used to upload the data processed by the telemetry terminal unit to the cloud; an AIoT city sensing middle platform is used to receive the data uploaded by the wireless communication unit, and is used to perform time synchronization, missing data marking, abnormal mutation marking and state conversion on the water level data, the flow data, the video image data and the meteorological data, and is used to identify a rainfall event according to the meteorological data; an abnormality identification module is used to determine the matching degree of water level change and flow response according to the correlation among the water level data, the flow data, the video image data and the meteorological data, and is used to determine the abnormal type and the suspected position of the sponge facility in combination with the visual abnormal state identified by the video. a water environment comprehensive treatment intelligent platform is used to display the abnormal type, the suspected position, the abnormal level, the triggering basis and the disposal suggestion output by the abnormality identification module, and is used to push early warning information to the management personnel.

2. The system of claim 1, wherein, The multi-source sensing unit comprises: a water level monitoring unit is arranged at a key node of the sponge facility, a storage space, a drainage outlet, an overflow control point or an area prone to water accumulation, and is used to collect water level data; a flow monitoring unit is arranged at a rainwater pipe, a drainage outlet or a water outlet section, and is used to collect flow data; a video monitoring unit is arranged at a position covering a water inlet, a road water accumulation area, a drainage outlet or an overflow node, and is used to collect video image data and identify one or more visual abnormal states such as water accumulation, floating objects, water spouting, overflow, abnormal outflow of the drainage outlet or equipment shielding; a meteorological monitoring unit is arranged near the monitoring area, and is used to collect rainfall amount and rainfall intensity data.

3. The system of claim 1, wherein, The AIoT city sensing middle platform is further used to: determine that a rainfall event has occurred when the cumulative rainfall amount in a continuous preset time is greater than a first rainfall threshold value or the instantaneous rainfall intensity is greater than a second rainfall threshold value; determine that the rainfall event has ended when the cumulative rainfall amount in a continuous preset time is less than a third rainfall threshold value.

4. The system of claim 1, wherein, The abnormality identification module is used to: calculate the water level change rate and the flow change rate; determine that there is a drainage response deficiency or a blockage abnormality when the water level change rate is greater than a water level rising rate threshold value for more than two consecutive monitoring periods and the flow change rate is less than a flow response threshold value or less than 0.

5. The system of claim 1, wherein, The abnormality identification module is further used to: determine that there is a blockage abnormality when the water level continues to rise, the flow does not increase synchronously, and the video monitoring unit identifies that there are floating objects accumulated at the water inlet or local water accumulation is expanded during the rainfall event, and mark the corresponding water inlet or the downstream drainage channel thereof as a suspected position.

6. The system of claim 1, wherein, The abnormality identification module is further used to: determine that there is an overflow abnormality when the water level exceeds a preset warning water level, the flow reaches a preset high position or fluctuates, and the video monitoring unit identifies that there is water spouting, overflow or road water accumulation range expansion.

7. The system of claim 1, wherein, The abnormality identification module is further used to: After a rainfall event ends, if the water level fails to drop to the normal range within the preset receding time and the flow rate remains low, it is judged as an abnormal discharge stagnation. When the flow rate increases but the water level continues to rise, and the rate of rise in the water level exceeds a preset threshold, it is determined to be an abnormality of insufficient drainage capacity. 8.A method for identifying anomalies in a sponge facility based on multi-source perception, characterized in that, include: Acquire water level data, flow data, video image data, and meteorological data during the operation of sponge city facilities; The water level data, flow rate data, video image data, and meteorological data are synchronized in time, marked for missing data, marked for abnormal changes, and transitioned in state. Based on the meteorological data, rainfall events are identified, and the start time, duration, intensity level, and end time of rainfall are determined. Based on the water level data and the flow rate data, determine the matching relationship between water level changes and flow rate response; Based on the video image data, identify visual abnormal conditions around the sponge facility. The visual abnormal conditions include one or more of the following: water accumulation, floating debris accumulation, inlet blockage, well overflow, abnormal outlet flow, and overflow. The matching results of water level changes and flow response are correlated with the visible abnormal conditions to determine the abnormal type and suspected location of the sponge city facilities. Output the anomaly type, the suspected location, the anomaly level, the triggering basis, and the handling suggestions.

9. An electronic device, comprising: The device includes: a processor and a memory; The memory is used to store instructions; The processor is configured to execute the instructions in the memory and perform the functions of the system according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, Including instructions that, when executed on a computer, cause the computer to perform the functions of the system described in any one of claims 1-7 above.

Images