A sewer network water quality multi-parameter online monitoring system and method

Abstract

The application discloses a kind of water quality multi-parameter online monitoring system and method of water supply and drainage pipe network, the system includes water supply and drainage pipe network system, drainage sampling module, detection center module, data transmission module, cloud center module, monitoring center module and system management and maintenance module;Drainage sampling module is set at the key node of water supply and drainage pipe network, and the water sample to be detected is automatically drained to detection center module by control valve;Detection center module carries out the multi-parameter detection of pH value, turbidity, heavy metal ion concentration, organic pollutant content, dissolved oxygen and conductivity to water sample;Data transmission module carries out encrypted transmission to cloud center module with the detection data of time stamp and positioning information;Cloud center module carries out data cleaning, denoising, normalization, data mining, deep learning trend prediction and abnormal early warning;Monitoring center module displays water quality state in real time based on dynamic node diagram, generates alarm, pollution route and pollution traceability result when abnormal.

Description

Technical Field

[0001] This invention relates to the field of water quality monitoring technology, and in particular to an online monitoring system and method for multiple parameters of water quality in water supply and drainage networks. Background Technology

[0002] Water supply and drainage networks are a crucial component of urban water circulation systems, typically comprising water supply networks, drainage networks, pumping stations, storage facilities, inspection wells, and multiple pipe sections connecting to users or treatment facilities. The water quality of these networks directly impacts the safety of domestic water use, wastewater collection efficiency, the management of combined sewer overflows, and overall water environment safety. Therefore, continuous, accurate, and visualized water quality monitoring of water supply and drainage networks is a vital foundation for network operation management and the handling of sudden pollution incidents.

[0003] Current water quality monitoring methods for water supply and drainage networks still largely rely on manual fixed-point sampling and offline laboratory analysis. While this method can obtain highly accurate single-test results, it has limited sampling frequency, inflexible sampling point scheduling, and long result return cycles, making it difficult to promptly capture short-term changes such as sudden discharges, pipeline backflow, rainwater and sewage mixing, or localized pollution spread. For pollution incidents, data lag directly reduces emergency response efficiency.

[0004] Existing online monitoring equipment typically focuses only on a few conventional physicochemical indicators such as pH and turbidity, lacking integrated and continuous detection capabilities for indicators closely related to pollution risk, such as heavy metal ion concentration, organic pollutant content, dissolved oxygen, and conductivity. Furthermore, data from individual monitoring points are usually presented in isolated form, lacking the ability to dynamically display nodes, extrapolate pollution routes, and reverse-locate pollution sources in conjunction with the pipeline network topology.

[0005] Furthermore, existing systems often rely on fixed thresholds or simple curves for data processing, failing to fully utilize historical water quality data, node locations, time-series characteristics, and multi-parameter correlations to build trend prediction models. Consequently, when water quality changes are in their early stages or when multiple indicators exhibit abnormal linkages, the system struggles to identify potential risks and issue early warnings in a timely manner.

[0006] In complex environments such as high temperature and humidity, heavy rainfall, saline and humid coastal air, short-term thunderstorms, and rapid fluctuations in pipe network liquid levels, detection equipment may experience problems such as sensor drift, condensation on the casing, backflow in sampling branches, sediment blockage, communication interruption, or power supply fluctuations. If the online monitoring system does not implement sampling protection, threshold compensation, and equipment protection for the above environmental factors, it is prone to false alarms, missed alarms, or monitoring interruptions. Therefore, there is a need for an online multi-parameter water quality monitoring system and method for water supply and drainage pipe networks that can achieve automatic sampling, multi-parameter detection, cloud analysis, dynamic display, pollution source tracing, and environmental adaptability protection. Summary of the Invention

[0007] To at least partially address the aforementioned problems, an exemplary embodiment of the present invention provides a multi-parameter online monitoring system and method for water quality in water supply and drainage networks. This system automatically diverts water samples to a monitoring center module by setting up diversion and sampling modules at key nodes in the water supply and drainage network system, and synchronously detects the water quality status using multi-parameter sensors. A data transmission module then encrypts and transmits the detection data, carrying timestamps, location information, and equipment status, to a cloud-based central module. The cloud-based central module preprocesses, mines, and performs deep learning predictions on the data. The monitoring center module maps the predictions and abnormal results to a dynamic node graph of the network, and generates pollution routes and pollution source tracing results based on the abnormal nodes.

[0008] In some embodiments, the diversion sampling module includes a diversion pipeline module and a control valve module. The diversion pipeline module can import water samples from key nodes of the pipeline network into the detection center module, and the control valve module can be opened or closed according to remote control commands, predetermined sampling cycles, or abnormal triggering conditions, thereby realizing on-demand sampling, timed sampling, and event-triggered sampling.

[0009] In some embodiments, the detection center module includes a pH detection module, a turbidity detection module, a heavy metal ion concentration detection module, an organic pollutant content detection module, a dissolved oxygen detection module, and a conductivity detection module. Each detection module may share a sampling chamber or have its own independent detection channel. The detection center module may also include a temperature compensation and calibration module to compensate, calibrate, or correct drift in the sensor output.

[0010] In some embodiments, the data transmission module includes a wireless communication module, a positioning module, a data encryption module, a timestamp module, and an edge caching module. The wireless communication module can be adapted to different communication networks, the positioning module is used to associate detection points with geographical locations or pipeline node identifiers, the data encryption module is used to improve transmission security, and the edge caching module is used to cache data when the network is unstable and retransmit it after recovery.

[0011] In some embodiments, the cloud-based central module includes a data preprocessing module, a data mining module, a deep learning algorithm module, a predictive analysis module, a model update module, and a threshold configuration module. The data preprocessing module is used to clean, denoise, and normalize the raw data; the data mining module is used to extract water quality change patterns and multi-parameter correlation patterns; the deep learning algorithm module is used to establish a water quality change trend prediction model; and the predictive analysis module is used to provide early warnings for abnormal trends.

[0012] In some embodiments, the monitoring center module includes a dynamic node graph construction module, a visualization interaction module, an anomaly detection module, an alarm signal generation module, a pollution route generation module, a pollution source tracing module, and a remote control module. The dynamic node graph can overlay and display network nodes, pipe segment connectivity, water flow direction, and water quality status at each node. Management personnel can quickly locate abnormal nodes based on the graphical interface and remotely execute valve control or adjust detection parameters.

[0013] In some embodiments, the system further includes an environmental adaptability module, a sampling protection module, and a power supply and communication protection module. The environmental adaptability module can adjust the sampling frequency, abnormal threshold, and compensation coefficient according to factors such as ambient temperature and humidity, rainfall intensity, pipeline liquid level, and humidity inside the casing; the sampling protection module can perform backflow prevention, flushing, drainage, and sample disposal when there is heavy rainfall or increased risk of backflow; the power supply and communication protection module can maintain continuous operation of the equipment under conditions of high humidity, salt spray, thunderstorms, or short-term mains power outages.

[0014] Through the above technical solutions, the present invention can achieve at least the following technical effects: First, it can realize automatic water sample collection and flexible scheduling by diversion sampling and control valves, reducing the error of manual sampling; second, it can expand the coverage of water quality risk factors by multi-parameter detection; third, it can improve the real-time performance and accuracy of anomaly early warning by encrypted transmission, cloud analysis and deep learning prediction; fourth, it can improve the efficiency of pipeline anomaly handling by dynamic node diagrams, pollution routes and pollution source tracing; and fifth, it can improve the reliability of continuous operation in complex climates and underground humid environments by environmental adaptability modules and equipment protection structures. Attached Figure Description

[0015] Figure 1 A schematic diagram of the overall structure of an online monitoring system according to some embodiments of the present invention is shown; Figure 2 A schematic diagram of the structure of a detection center module according to some embodiments of the present invention is shown; Figure 3 A schematic diagram of the structure of the data transmission module and the cloud center module according to some embodiments of the present invention is shown; Figure 4 A schematic diagram of the structure of a monitoring center module according to some embodiments of the present invention is shown; Figure 5 A schematic diagram of the structure of a dynamic node graph construction module according to some embodiments of the present invention is shown; Figure 6 A schematic diagram of the structure of a visual interaction module according to some embodiments of the present invention is shown; Figure 7 A schematic diagram of the structure of a system management and maintenance module according to some embodiments of the present invention is shown; Figure 8A flowchart illustrating an online monitoring method according to some embodiments of the present invention is shown; and Figure 9 A schematic diagram of an environmental adaptability and sampling protection structure according to some embodiments of the present invention is shown.

[0016] Explanation of reference numerals in the attached figures: 1. Water supply and drainage network system; 2. Drainage and sampling module; 3. Detection Center Module; 31. pH Value Detection Module; 32. Turbidity Detection Module; 33. Heavy Metal Ion Concentration Detection Module; 34. Organic Pollutant Content Detection Module; 35. Dissolved Oxygen Detection Module; 36. Conductivity Detection Module; 37. Sampling Chamber; 38. Temperature Compensation and Calibration Module; 4. Data transmission module; 41. Wireless communication module; 42. Positioning module; 43. Data encryption module; 44. Timestamp module; 45. Edge buffer module; 5. Cloud Center Module; 51. Data Preprocessing Module; 52. Data Mining Module; 53. Deep Learning Algorithm Module; 54. Predictive Analysis Module; 55. Model Update Module; 56. Threshold Configuration Module; 6. Monitoring Center Module; 61. Dynamic Node Graph Construction Module; 611. Node Data Integration Module; 612. Graphics Rendering Module; 613. Status Marking Module; 62. Visual Interaction Module; 621. User Interface Module; 622. Data Display Module; 623. Interactive Operation Module; 63. Anomaly Detection Module; 64. Alarm Signal Generation Module; 65. Pollution Route Generation Module; 66. Pollution Source Tracing Module; 67. Remote Control Module; 7. System Management and Maintenance Module; 71. User Management Module; 72. Device Management Module; 73. Log Management Module; 74. Database Module; 75. Data Backup and Recovery Module; 76. System Maintenance Module; 8. Environmental adaptability module; 81. Environmental sensor; 82. Rainfall and liquid level recognition module; 83. Shell temperature and humidity monitoring module; 84. Threshold compensation and sampling scheduling module; 9. Sampling protection module; 10. Power supply and communication protection module. Detailed Implementation

[0017] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be understood that the following embodiments are only used to more clearly and completely illustrate the technical solutions of the present invention, and are not intended to limit the scope of protection of the present invention. After reading this invention, those skilled in the art can substitute, combine, or modify the structures, modules, processes, algorithms, thresholds, communication methods, sampling methods, and device arrangements in the various embodiments without departing from the concept of the present invention. All such substitutions, combinations, or modifications should fall within the scope of protection of the present invention.

[0018] The term "comprising" and its variations, as used herein, imply open inclusion, meaning "including but not limited to". The term "connection" can refer to a direct connection or an indirect connection via intermediate components, networks, pipelines, valves, interfaces, or protocols. The term "data connection" can include wired communication connections, wireless communication connections, local area network connections, cellular communication connections, Internet of Things (IoT) communication connections, private network connections, and data exchange connections via cloud servers. The term "fluid connectivity" means that a water sample can flow from one location to another through pipelines, valves, sampling chambers, filters, or other fluid components, without requiring continuous flow of the water sample.

[0019] The "water supply and drainage network system 1" in this invention can be a water supply network, drainage network, rainwater network, sewage network, reclaimed water network, or combined sewer system, or a composite network system composed of multiple sub-networks. For the water supply network, key monitoring targets may include water source outlets, pump station outlets, zone metering points, network junctions, nodes near end users, and areas with high incidence of water quality complaints. For the drainage network, key monitoring targets may include manholes, main manifold inlets, industrial park discharge inlets, nodes before and after pump stations, discharge outlets, points at risk of combined sewer overflows, and locations prone to backflow or siltation. In the above different application scenarios, the hardware layout, sampling frequency, and anomaly thresholds of the system may differ, but all can utilize the online monitoring system and method proposed in this invention to achieve multi-parameter water quality detection, anomaly early warning, and pollution source tracing.

[0020] The "key nodes" in this invention can be determined based on pipeline topology, historical anomaly records, population density, industrial emission risks, terrain elevation, risk of combined sewer overflows, pump station operating status, and maintenance experience. Key nodes can be either fixed online monitoring points or dynamically adjusted key monitoring points based on historical data and cloud-based analysis results. For example, when a certain area frequently experiences increased turbidity, sudden changes in conductivity, or increased organic pollutant content over a period of time, the system can temporarily upgrade upstream or adjacent nodes in that area to key monitoring nodes and increase the sampling frequency through remote control commands.

[0021] The "multi-parameter water quality detection" in this invention refers to the online detection of at least two water quality indicators within the same detection center module 3, the same sampling chamber 37, or the same detection process. For example, multi-parameter water quality detection may include pH value, turbidity, heavy metal ion concentration, organic pollutant content, dissolved oxygen, conductivity, water temperature, redox potential, residual chlorine, ammonia nitrogen, total phosphorus, total nitrogen, chemical oxygen demand, or characteristic ion concentration, etc. For ease of explanation, the following embodiments use pH value, turbidity, heavy metal ion concentration, organic pollutant content, dissolved oxygen, and conductivity as the main detection indicators; however, the scope of protection of this invention is not limited by the specific number and combination of detection indicators.

[0022] The "dynamic node diagram" in this invention refers to a visual object that graphically associates and presents pipeline network nodes, pipe sections, detection equipment, flow direction, liquid level, detection data, abnormal states, and early warning information. The dynamic node diagram can be presented using topology wireframes, geographic information layers, 3D pipeline network models, or 2D planar schematic diagrams. The dynamic node diagram can be updated in real time as sampling data is refreshed, and can distinguish normal, risk, and abnormal states through colors, icons, line widths, flashing, pop-ups, list sorting, or audio-visual prompts. The dynamic node diagram also supports timeline playback, allowing managers to view the entire process of an abnormal event from its occurrence and spread to its resolution.

[0023] In this invention, "pollution route" refers to a possible pollution diffusion path inferred based on the location of abnormal nodes, the type of abnormal parameters, the connectivity of the pipeline network topology, the flow direction of the pipe section, liquid level changes, the operating status of pumping stations, and the sequence of anomalies. "Pollution source tracing" in this invention refers to the process of determining suspected pollution sources or pollution source areas by performing reverse analysis along the pollution route or pipeline network topology, combined with the time difference of anomaly occurrence, the similarity of abnormal parameters, the upstream and downstream relationships of nodes, historical discharge records, and hydraulic conditions. The results of pollution routes and pollution source tracing can be used to guide on-site inspections, valve isolation, evidence collection and sampling, emergency response, and subsequent accountability.

[0024] Reference Figure 1 According to some embodiments of the present invention, a multi-parameter online monitoring system for water quality in water supply and drainage networks includes a water supply and drainage network system 1, a diversion and sampling module 2, a detection center module 3, a data transmission module 4, a cloud center module 5, a monitoring center module 6, and a system management and maintenance module 7. In a further embodiment, the system may also include an environmental adaptability module 8, a sampling protection module 9, and a power supply and communication protection module 10. These modules can be interconnected through pipelines, control lines, wireless networks, cloud platform interfaces, edge gateways, or other means to form a closed-loop system from water sample collection, detection, transmission, analysis, early warning, treatment to maintenance and management.

[0025] The water supply and drainage network system 1, as the object to be monitored, can include multiple nodes and pipe sections connecting these nodes. Each node can have a node number, geographical location, pipe diameter information, burial depth information, upstream and downstream relationships, zone of origin, operating conditions, historical anomaly records, and equipment binding information. During system deployment, the key nodes requiring the placement of the diversion sampling module 2 can be determined first based on network drawings, on-site surveys, historical maintenance records, and risk assessment results. For nodes with high pollution risk, large flow variations, or a wide range of user impact, a higher sampling frequency can be set; for regular nodes, a lower sampling frequency can be set to balance monitoring accuracy, equipment lifespan, and maintenance costs.

[0026] The diversion sampling module 2 is installed at at least one key node of the water supply and drainage network system 1. It is used to draw water samples to be tested from the key node and deliver the samples to the testing center module 3. The diversion sampling module 2 may include a diversion pipeline module and a control valve module. The diversion pipeline module connects the key node of the water supply and drainage network system 1 to the testing center module 3, and may include a sampling interface, sampling branch pipe, flow regulation structure, venting structure, filtration structure, sedimentation structure, flushing interface, and discharge interface. The sampling interface may be located on the side wall of the main pipeline, in a relatively stable area of ​​the inspection well, at the front end of the pumping station, or at the rear end of the pumping station to reduce the impact of bubbles, sediment, vortices, and local turbulence on the representativeness of the samples.

[0027] The control valve module may include solenoid valves, electric ball valves, electric gate valves, miniature proportional valves, or other remotely controllable valve structures. The control valve module is configured to open or close in response to remote control commands from the monitoring center module 6, so that the water sample to be tested is automatically collected and diverted to the detection center module 3. The control valve module can also automatically operate according to the sampling plan stored in the local controller, for example, opening the sampling valve at predetermined intervals to introduce a certain volume of water sample into the sampling chamber 37. In scenarios where multiple sampling branches share the same detection center module 3, the control valve module can sequentially open different branches according to time slices to avoid mixing of water samples from different nodes.

[0028] In some embodiments, the diversion sampling module 2 can perform a pre-rinsing step before formal sampling. Specifically, the control valve module first opens the sampling branch, allowing the old water sample retained in the diversion pipeline module to be discharged through the bypass discharge unit, and then fresh water sample is introduced into the sampling chamber 37. Pre-rinsing can improve the representativeness of the detected water sample to the current node's water quality status. After the detection is completed, the diversion sampling module 2 can perform evacuation, backwashing, or combined gas-liquid flushing to reduce the impact of sediment, attached substances, air bubbles, and residual water sample on subsequent detection results. For drainage networks or combined sewer systems, the diversion sampling module 2 can also be equipped with anti-clogging filters, sedimentation chambers, and removable maintenance ports to reduce the risk of particulate matter clogging valves and sampling pipelines.

[0029] The testing center module 3 is fluidly connected to the diversion and sampling module 2 and is configured to perform multi-parameter water quality testing on the water samples to be tested. The testing center module 3 can be located in a ground monitoring cabinet, underground equipment box, pump station auxiliary space, equipment compartment next to a manhole, or pipe gallery near key nodes. The testing center module 3 can adopt a flow-through testing structure, allowing the water sample to pass through multiple testing units sequentially; or it can adopt a distribution testing structure, allowing the water sample to enter the sampling chamber 37 and then flow into different testing pools through diversion channels. The testing center module 3 can also be equipped with a constant flow structure to reduce the impact of flow rate changes on turbidity, dissolved oxygen, and conductivity detection.

[0030] Reference Figure 2 The detection center module 3 includes a pH detection module 31, a turbidity detection module 32, a heavy metal ion concentration detection module 33, an organic pollutant content detection module 34, a dissolved oxygen detection module 35, a conductivity detection module 36, a sampling chamber 37, and a temperature compensation and calibration module 38. The pH detection module 31 is used to detect changes in the acidity or alkalinity of the water sample, and can employ a glass electrode, a solid electrode, or other online pH detection structures. The turbidity detection module 32 is used to detect changes in optical scattering caused by suspended particles, colloids, or sediment in the water sample, and can employ either scattered light detection or transmitted light detection methods. The heavy metal ion concentration detection module 33 is used to detect hazardous metal ions such as lead, cadmium, mercury, chromium, arsenic, copper, and zinc, and can employ electrochemical methods, colorimetric methods, spectroscopic methods, or other methods suitable for online detection. The organic pollutant content detection module 34 reflects changes in organic pollutants in the water sample, and can be based on spectral absorption, electrochemical response, fluorescence response, total organic carbon correlation detection, or chemical oxygen demand correlation detection methods. The dissolved oxygen detection module 35 is used to detect the dissolved oxygen content in water samples, and can be performed using fluorescence, electrochemical, or membrane electrode methods. The conductivity detection module 36 is used to detect changes in the total amount of ions in water samples, which can help determine salinity, industrial emissions, rainwater and sewage mixing, or abnormal exchange in the pipe network.

[0031] The sampling chamber 37 is used to contain the water sample to be tested. The sampling chamber 37 can be a flow-through chamber, with its inlet connected to the drainage sampling module 2 and its outlet connected to the waste liquid discharge interface or return pipeline. The sampling chamber 37 can be equipped with a flow stabilization structure, an venting structure, and a liquid level detection structure. The flow stabilization structure reduces the impact of incoming water, the venting structure removes air bubbles from the water sample, and the liquid level detection structure confirms whether the required liquid level has been reached within the sampling chamber 37. The sampling chamber 37 can also be made of corrosion-resistant materials to adapt to the complex water quality environment in drainage networks.

[0032] The temperature compensation and calibration module 38 is used to compensate for or calibrate at least one detection parameter. Since indicators such as pH, conductivity, and dissolved oxygen are easily affected by water sample temperature, the temperature compensation and calibration module 38 can detect the water sample temperature in real time and correct the detection results according to a preset compensation model. The temperature compensation and calibration module 38 can also store the zero-point calibration value, slope calibration value, drift record, last calibration time, and calibration solution batch information for each detection module. When uploading detection data, the temperature compensation and calibration module 38 can write the compensation status, calibration status, and sensor health status into the data packet, enabling the cloud center module 5 to determine the reliability of the detection data.

[0033] In some embodiments, the detection center module 3 may also include an automatic cleaning structure. The automatic cleaning structure may include a clean water inlet, a cleaning fluid inlet, an air washing inlet, a drain inlet, and a cleaning control valve. The system can automatically initiate a cleaning procedure based on the number of tests, sensor drift, abnormal turbidity, the degree of contamination in the sampling chamber, or a maintenance plan. For example, when turbidity is consistently high and there is a risk of sediment blockage in the drainage scenario, the system can first perform bypass discharge and sampling chamber flushing before conducting formal testing, thereby reducing the impact of sediment on the test results.

[0034] Data transmission module 4 is connected to the detection center module 3 and is configured to encrypt and transmit detection data, timestamp information, and location information. (Refer to...) Figure 3 The data transmission module 4 may include a wireless communication module 41, a positioning module 42, a data encryption module 43, a timestamp module 44, and an edge buffer module 45. The wireless communication module 41 can support cellular networks, narrowband IoT, wireless LAN, private network communication, LoRa communication, Ethernet communication, or a combination of multiple communication methods. For underground spaces or inspection well scenarios, the wireless communication module 41 can work with ground relay devices, edge gateways, or antenna extension structures to improve communication stability.

[0035] The positioning module 42 is used to attach positioning information to the detection data. The positioning information can be latitude and longitude coordinates, elevation information, or pre-bound pipeline node numbers, equipment numbers, or site numbers. In underground pipeline network scenarios, the positioning module 42 can determine the sampling location by binding the equipment installation file with the node number, without relying on real-time satellite positioning. The timestamp module 44 is used to write at least one of the sampling time, detection time, upload time, and reception time into the detection data. Through the timestamp information, the cloud center module 5 can perform time alignment of data from different nodes, with different parameters, and with different sampling periods, thereby improving the accuracy of anomaly analysis and pollution diffusion judgment.

[0036] The data encryption module 43 is used to encrypt the detection data. The data encryption module 43 can employ methods such as symmetric encryption, asymmetric key negotiation, digital signature, data digest, or access token verification to improve the security of data transmission. The encrypted data packet may include node number, sampling sequence number, detection parameters, unit, timestamp, location information, device status, calibration status, anomaly marker, and data signature. Through encrypted encapsulation, the system can reduce the risk of detection data being tampered with, forged, or illegally read.

[0037] The edge caching module 45 is configured to cache detection data when the wireless link is abnormal and retransmit the detection data in the order of data collection after the wireless link is restored. Wireless link abnormalities can include weak signal, base station failure, network congestion, communication interruption in underground spaces, device restart, or temporary unavailability of cloud services. The edge caching module 45 can use circular storage, hierarchical caching, or event-priority caching methods to store data. For ordinary data, it can be cached according to the sampling order; for abnormal data or red alert data, a higher priority can be set so that it can be uploaded first after communication is restored. During retransmission, the edge caching module 45 can sort the data according to the sampling sequence number and timestamp to avoid affecting the trend curve and pollution route inference due to out-of-order retransmission data.

[0038] The cloud center module 5 is connected to the data transmission module 4 and is configured to perform preprocessing, data mining, deep learning analysis, and predictive analysis on the received data. (Refer to...) Figure 3 The cloud-based central module 5 includes a data preprocessing module 51, a data mining module 52, a deep learning algorithm module 53, a predictive analysis module 54, a model update module 55, and a threshold configuration module 56. The cloud-based central module 5 can be deployed in a public cloud, private cloud, industry-specific network cloud, or local data center. Alternatively, a cloud-edge collaborative architecture can be adopted, allowing some data preprocessing and initial anomaly screening to be completed at the edge.

[0039] The data preprocessing module 51 is configured to clean, denoise, and normalize the raw data. Data cleaning may include missing value identification, duplicate data removal, unit conversion, abnormal timestamp correction, offline data marking, and invalid sample removal. For example, when the liquid level in the sampling chamber 37 is insufficient, or when the sensor is in cleaning or calibration mode, the corresponding detection value can be marked as informal detection data and will not be included in the anomaly judgment. Data denoising may employ a moving average filter, median filter, exponential smoothing, Kalman filter, or other time series processing methods. For example, the moving average filter satisfies: y[n] = (1 / N)∑_{k=0}^{N-1}x[nk], Where x[n] is the original data sequence, y[n] is the filtered data sequence, and N is the sliding window size. The sliding average can reduce the impact of short-term random noise on early warning judgment. For sudden pollution events, to avoid excessive smoothing leading to delayed anomaly detection, the data preprocessing module 51 can dynamically adjust the sliding window size according to parameter type and node risk level. For example, a smaller window can be used for scenarios sensitive to sudden changes in pH and conductivity; a larger window can be used for scenarios with large turbidity fluctuations but where persistence is more important.

[0040] Data normalization can scale water quality parameters with different dimensions and value ranges to a uniform range, facilitating joint analysis of multiple parameters. For example, the normalization formula can be: X = (x-min(x)) / (max(x)-min(x)), Where x represents the original data, X represents the normalized data, and min(x) and max(x) can be the minimum and maximum values ​​within the historical time window, or they can be the upper and lower limits of parameters preset by the system according to the node type. After normalization, the deep learning algorithm module 53 can more stably learn the correlation between multiple water quality parameters.

[0041] The data mining module 52 is used to extract water quality change patterns, inter-node correlations, and anomaly patterns from the preprocessed data. The data mining module 52 can analyze the time delay relationships between different nodes, the coupling relationships between different parameters of the same node, and the correlations between abnormal events and rainfall, pump station operation, and industrial emission periods. For example, if the conductivity of a node increases, and downstream nodes show a similar increasing trend after a certain period, the system can identify this as a possible pollution migration path. The data mining module 52 can also mine long-term trends, such as a gradual decrease in dissolved oxygen at the end of the pipe network, a periodic increase in turbidity, or an increase in organic pollutant content at certain nodes after rain.

[0042] The deep learning algorithm module 53 is used to build a water quality change trend prediction model based on historical water quality data, node topology data, environmental data, and equipment status data. The deep learning algorithm module 53 can employ recurrent neural networks, long short-term memory networks, gated recurrent units, temporal convolutional networks, Transformer time series models, graph neural networks, or multi-model fusion structures. For pipe networks with obvious topological relationships, the deep learning algorithm module 53 can construct a graph structure of nodes and pipe segments, combining it with graph neural networks to learn the influence relationships between upstream and downstream nodes. For parameters with obvious time periods, the deep learning algorithm module 53 can incorporate daily, weekly, seasonal variations, and rainfall status features to improve prediction accuracy.

[0043] The predictive analysis module 54 generates parameter trends, risk probabilities, and anomaly warnings based on the output of the deep learning algorithm module 53. The predictive analysis module 54 can not only determine whether the current value is abnormal, but also predict whether there will be anomaly risks in the future. For example, if the conductivity of a node does not exceed a fixed threshold, but its rate of increase is significantly higher than the historical baseline, and an anomaly has already occurred in the upstream node, the predictive analysis module 54 can output a risk warning in advance. The model update module 55 updates the model parameters based on newly added detection data, manual confirmation results, handling feedback results, and historical event review results. The threshold configuration module 56 configures anomaly judgment thresholds based on node category, historical baseline, or operating conditions. Node categories can include main water supply nodes, terminal nodes, drainage main pipe nodes, nodes before and after pumping stations, industrial emission risk nodes, and stormwater and sewage mixing risk nodes, etc. Different nodes have different normal fluctuation ranges; therefore, the threshold configuration module 56 can configure differentiated thresholds for different nodes, thereby reducing false alarms and false negatives.

[0044] The monitoring center module 6 is connected to the cloud center module 5 and is configured to construct a dynamic node graph based on predictive analysis results, perform anomaly detection, generate alarm signals, generate pollution routes, and perform pollution source tracing. (Refer to...) Figure 4 The monitoring center module 6 may include a dynamic node graph construction module 61, a visualization interaction module 62, an anomaly detection module 63, an alarm signal generation module 64, a pollution route generation module 65, a pollution source tracing module 66, and a remote control module 67.

[0045] The dynamic node graph construction module 61 includes a node data integration module 611, a graphics rendering module 612, and a status marking module 613. The node data integration module 611 integrates water quality data from various monitoring nodes and associates the water quality data with the network topology, node locations, pipe flow direction, monitoring equipment numbers, equipment status, and historical anomaly records. The graphics rendering module 612 renders the nodes and pipe segments of the water supply and drainage network system 1 into a visual graphic. This visual graphic can be a wireframe diagram or overlaid on a base map, network asset map, or 3D model. The status marking module 613 assigns normal, risk, or abnormal status markings to nodes based on water quality monitoring results. For example, green can represent normal, yellow can represent potential risk, and red can represent abnormal. The status marking module 613 can also highlight high-risk nodes through node flashing, bold borders, label prompts, or list placement.

[0046] The visualization interaction module 62 may include a user interface module 621, a data display module 622, and an interactive operation module 623. The user interface module 621 provides access to system login, permission entry, node retrieval, layer switching, alarm lists, and device management. The data display module 622 displays water quality data and analysis results in the form of line charts, bar charts, trend curves, dashboards, parameter tables, or multi-parameter radar charts. The interactive operation module 623 supports clicking on nodes to view details, dragging the topology map, zooming layers, selecting areas, switching timelines, replaying abnormal processes, and exporting handling reports. Through the visualization interaction module 62, managers can quickly locate abnormal nodes and view their upstream and downstream relationships, historical curves, and handling suggestions.

[0047] Anomaly detection module 63 is used for real-time monitoring and analysis of water quality data. Anomaly detection module 63 can use a combination of fixed thresholds, dynamic thresholds, statistical rules, model prediction bias, and manual verification rules for judgment. For example, anomaly detection module 63 can calculate the mean μ and standard deviation σ of the target water quality parameter within a predetermined time window, and when the target data x satisfies |x-μ|>3σ, the detection node corresponding to the target data x is determined as an anomaly node. Furthermore, when the target data x does not exceed the 3σ rule, but there is a continuous deviation between it and the predicted value output by the prediction analysis module 54, or when the abnormal trend is temporally correlated with the anomaly of the upstream node, the anomaly detection module 63 can also determine the node as a risk node or an anomaly node. By combining statistical rules and model prediction bias, the system's ability to identify sudden and slow pollution can be improved.

[0048] The alarm signal generation module 64 generates alarm signals based on the output of the anomaly detection module 63. Alarm signals can be categorized into different levels based on the type of abnormal parameters, duration of the anomaly, number of abnormal nodes, node importance, affected area, and predicted spread range. For minor anomalies, the system can generate risk alerts; for persistent or multi-parameter anomalies, the system can generate critical alarms; for anomalies that may affect water supply safety, drainage safety, or public environmental safety, the system can generate high-level alarms and trigger coordinated response measures. Alarm signals can be sent to relevant personnel via platform pop-ups, SMS, telephone, mobile applications, email, audible and visual alarms, or the emergency command system interface.

[0049] The pollution path generation module 65 is used to determine the pollution diffusion path based on the topological connectivity of the water supply and drainage network system 1, the flow direction of pipe segments, the order of occurrence of abnormal nodes, and the types of abnormal parameters. Specifically, the pollution path generation module 65 can first determine the earliest abnormal node, and then combine the upstream and downstream relationships and the flow direction of pipe segments to determine the possible direction of pollution movement. For water supply networks, pollution diffusion is usually related to the water supply flow direction, pressure zoning, and valve status; for drainage networks, pollution diffusion is usually related to water level, slope, pump station operation, and rainwater inflow. The pollution path generation module 65 can use the type of abnormal parameters as an auxiliary judgment condition. For example, a sudden change in conductivity may correspond to salinity or industrial discharge, an increase in organic pollutant content may correspond to the entry of organic wastewater, and an increase in turbidity may correspond to silt erosion or pipeline disturbance.

[0050] The pollution source tracing module 66 is used to search upstream related nodes along the pollution diffusion path and determine suspected pollution sources by combining the abnormal time difference and parameter similarity of each node. Parameter similarity can be calculated based on the distance between the normalized vectors of multiple water quality parameters, such as comparing the combined characteristics of pH value, turbidity, conductivity, organic pollutant content, and heavy metal ion concentration. When the abnormal combination of an upstream node is similar to the abnormal combinations of multiple downstream nodes, and the time of the abnormal occurrence matches the water flow propagation time, then the upstream node or its vicinity can be identified as a suspected pollution source. The pollution source tracing module 66 can also combine historical emission risks, industrial park locations, pump station operation records, and rainwater and sewage mixing risk points to sort suspected pollution sources and output a source tracing report to the monitoring center module 6.

[0051] The remote control module 67 is configured to remotely control the opening and closing of the control valve module and adjust the detection parameters of the detection center module 3. For example, when an anomaly occurs at a certain node, the remote control module 67 can control the control valve modules of that node and its upstream and downstream nodes to increase the sampling frequency to obtain more intensive water quality data; it can also control the detection center module 3 to retest a certain parameter to confirm whether the anomaly persists. In drainage scenarios, the remote control module 67 can also initiate the flushing, emptying, and sample discarding processes of the sampling protection module 9. In equipment maintenance scenarios, the remote control module 67 can issue calibration, cleaning, restart, pause sampling, or resume sampling commands.

[0052] The system management and maintenance module 7 is data-connected to the monitoring center module 6 and is configured to manage user permissions, device status, log recording, data storage backup, and system maintenance. The system management and maintenance module 7 may include a user management module 71, a device management module 72, a log management module 73, a database module 74, a data backup and recovery module 75, and a system maintenance module 76. The user management module 71 is used to set access permissions and operation permissions for different users. For example, administrators can configure thresholds and models, maintenance personnel can view device status and perform maintenance operations, emergency personnel can view alarms and tracing results, and read-only users can view real-time data but cannot issue control commands.

[0053] The equipment management module 72 is used to establish equipment files for the testing equipment, recording equipment operating status, installation location, equipment number, sensor model, communication module number, valve number, calibration records, fault logs, maintenance records, and spare parts replacement records. The equipment management module 72 can also record sensor drift, valve actuation counts, communication offline counts, battery level, housing humidity, last cleaning time, and last calibration time. The log management module 73 is used to record operation logs for system login, data modification, threshold adjustment, remote control, alarm confirmation, model updates, equipment maintenance, and data recovery. Through log recording, the system operation process and personnel operation process can be traced after abnormal events.

[0054] Database module 74 stores and manages system data, including raw detection data, preprocessed data, prediction results, alarm records, pollution routes, pollution source tracing results, equipment files, maintenance records, and user operation logs. Data backup and recovery module 75 can back up the database according to daily, weekly, monthly, or event-triggered strategies, and perform recovery in case of database corruption, server migration, accidental deletion, or system failure. System maintenance module 76 is used for software upgrades, hardware replacements, configuration distribution, remote diagnostics, and security policy updates. System management and maintenance module 7 can also generate equipment health scores based on equipment operating status to trigger calibration reminders, maintenance reminders, or backup and recovery operations. The equipment health score can be calculated by comprehensively considering sensor drift amplitude, communication stability, valve action success rate, edge cache occupancy rate, casing humidity, power status, and calibration overdue days. When the equipment health score falls below a preset threshold, the system can automatically generate a maintenance work order and push the equipment location, fault type, and suggested handling measures to maintenance personnel.

[0055] According to some embodiments of the present invention, a method for online monitoring of multiple parameters of water quality in water supply and drainage networks is also provided. This method can be applied to the online monitoring system for multiple parameters of water quality in water supply and drainage networks in any of the above embodiments. The method includes steps such as water sample collection and diversion, multi-parameter water quality detection, data marking and encrypted transmission, cloud data processing and analysis, real-time monitoring and anomaly handling, and system management and maintenance. The steps can be executed in a fixed order, or in parallel or cyclically depending on the actual operating conditions.

[0056] In the water sampling and diversion process, the system performs water sampling and diversion at key nodes of the water supply and drainage network system 1 via the diversion sampling module 2. Specifically, the control valve module opens according to the sampling plan or remote control command, and the diversion pipeline module introduces the water sample from the key node into the detection center module 3. Before formal testing, the system can perform a pre-flushing to drain the stagnant water in the diversion pipeline. If multiple key nodes share the same detection center module 3, the system can sequentially open the corresponding sampling branches according to node number, risk level, or scheduling strategy. For nodes that have shown abnormal trends, the system can temporarily shorten the sampling cycle to improve the temporal resolution of abnormal events.

[0057] In the multi-parameter water quality testing step, the testing center module 3 performs multi-parameter water quality testing on the diverted water sample. After the water sample enters the sampling chamber 37, the pH detection module 31, turbidity detection module 32, heavy metal ion concentration detection module 33, organic pollutant content detection module 34, dissolved oxygen detection module 35, and conductivity detection module 36 output their respective test results. The temperature compensation and calibration module 38 can acquire the water sample temperature and compensate for the test results affected by temperature. If the system detects insufficient liquid level in the sampling chamber 37, excessive sensor drift, or abnormal calibration status, it can mark the corresponding test data with confidence to prevent it from directly participating in the anomaly warning.

[0058] In the data marking, encryption, and transmission steps, the data transmission module 4 timestamps, locates, encrypts, and transmits the detection data. The timestamp module 44 records the sampling time, detection time, and upload time; the location module 42 records the node location or node number; the data encryption module 43 encrypts and encapsulates the data packets; and the wireless communication module 41 uploads the encrypted data to the cloud center module 5. If the wireless link fails, the edge caching module 45 temporarily stores the detection data and retransmits the detection data according to the data acquisition order after the wireless link is restored. Through the edge caching and retransmission mechanism, the system can reduce the risk of data loss caused by communication interruption.

[0059] In the cloud-based data processing and analysis steps, the cloud center module 5 performs data preprocessing, data mining, deep learning analysis, and predictive analysis on the received data. The data preprocessing module 51 first performs integrity checks, noise reduction, and normalization on the raw data. The data mining module 52 extracts inter-node correlations and anomaly patterns from the preprocessed data. The deep learning algorithm module 53 establishes or updates a water quality change trend prediction model based on historical and real-time data. The predictive analysis module 54 outputs future time period parameter trends, anomaly probabilities, and risk levels. The threshold configuration module 56 dynamically configures anomaly judgment thresholds based on node type, historical baseline, operating conditions, and environmental status.

[0060] In the real-time monitoring and anomaly handling process, the monitoring center module 6 displays the water quality status in real time, marks abnormal nodes, generates alarm signals, generates pollution routes, and performs pollution source tracing when anomalies occur. The dynamic node graph construction module 61 maps the detection data to pipeline nodes and pipe segments, and the status marking module 613 updates the node status based on the anomaly detection results. The anomaly detection module 63 can identify abnormal nodes based on the 3σ rule and prediction deviation conditions. The alarm signal generation module 64 generates alarms based on the anomaly level and pushes the alarms to relevant personnel. The pollution route generation module 65 determines the pollution diffusion path based on topological connectivity, pipe segment flow direction, the order of abnormal node occurrence, and the type of abnormal parameters. The pollution source tracing module 66 searches upstream related nodes along the pollution diffusion path and determines suspected pollution sources by combining the anomaly time difference and parameter similarity.

[0061] In the system management and maintenance process, module 7 performs user access control, device status management, log management, data storage backup, and system maintenance. Module 72 continuously updates device profiles and device health scores. When a device health score falls below a threshold, the system can trigger calibration reminders, maintenance reminders, or backup / restore operations. For example, if a pH sensor experiences persistent zero-point drift, the system can prompt maintenance personnel to perform calibration; if a wireless communication module frequently goes offline, the system can prompt to check the antenna, power supply, or communication network; if the edge buffer is chronically over-occupied, the system can indicate an anomaly in the communication link.

[0062] According to a further embodiment of the present invention, the system further includes an environmental adaptability module 8. The environmental adaptability module 8 includes an environmental sensor 81, a rainfall and liquid level identification module 82, a housing temperature and humidity monitoring module 83, and a threshold compensation and sampling scheduling module 84. The environmental sensor 81 can collect environmental temperature, environmental humidity, rainfall intensity, wind and rain conditions, air pressure, wind speed, external water accumulation status of equipment, or air corrosivity-related indicators. The rainfall and liquid level identification module 82 is used to identify heavy rainfall, rapid rise in pipeline liquid level, water accumulation in well chambers, or backflow risks. The housing temperature and humidity monitoring module 83 is used to monitor the temperature and humidity inside the housing of the equipment where the detection center module 3 or data transmission module 4 is located to determine whether condensation, water ingress, or abnormal heat dissipation exists. The threshold compensation and sampling scheduling module 84 is configured to dynamically adjust the sampling frequency, anomaly judgment threshold, and compensation coefficient of at least one water quality parameter in environments with high temperature and humidity, heavy rainfall, strong winds and rain outside typhoons, saline and humid coastal air, or rapid changes in pipeline liquid level.

[0063] In high-temperature environments, elevated water sample temperature may affect dissolved oxygen, conductivity, and some electrochemical detection results. The threshold compensation and sampling scheduling module 84 can compensate for the results of the dissolved oxygen detection module 35 and conductivity detection module 36 by combining the water sample temperature and ambient temperature, and mark the compensation coefficient in the cloud center module 5. In high-humidity environments, condensation may occur inside the equipment casing, which may affect electrical connections, communication stability, and sensor signals. When the casing temperature and humidity monitoring module 83 detects abnormal humidity, it can trigger the anti-condensation heating or dehumidification unit in the power supply and communication protection module 10 and increase the equipment self-test frequency. In heavy rainfall environments, rainwater runoff may cause a rapid rise in the drainage network liquid level, pollutants being washed into the network, or rainwater and sewage mixing. The rainfall and liquid level identification module 82 can identify this condition, and the threshold compensation and sampling scheduling module 84 increases the sampling frequency. At the same time, it dynamically adjusts the abnormal judgment thresholds for turbidity, conductivity, and organic pollutant content to distinguish between normal rainfall scouring effects and abnormal pollution events.

[0064] According to a further embodiment of the present invention, the system further includes a sampling protection module 9. The sampling protection module 9 may include an anti-backflow check valve, a self-cleaning flushing unit, a bypass discharge unit, and an anti-clogging filter. The sampling protection module 9 is configured to close the affected sampling branch when there is heavy rainfall or an increased risk of backflow in the pipe network, and to perform flushing, emptying, and sample disposal before resuming sampling, thereby reducing the impact of rainwater and sewage contamination, sediment blockage, or residual water samples on the test results. The anti-backflow check valve may be located near the water supply and drainage pipe network system 1 in the sampling branch, and automatically closes when the pipe network water level is higher than the equipment's safe level or when reverse water pressure occurs, preventing sewage, rainwater, or sediment from flowing into the detection center module 3. The self-cleaning flushing unit may use clean water flushing, air-water mixed flushing, or reverse flushing to clean the sampling pipe and the anti-clogging filter. The bypass discharge unit is used to discharge pre-flushing water, discarded sample water, or abnormally turbid water, preventing it from directly entering the sampling chamber 37. The anti-clogging filter can intercept larger particles and may be designed as a detachable or automatically scraping type to reduce maintenance difficulty.

[0065] In some embodiments, when the rainfall and liquid level identification module 82 detects a rapid rise in the pipe network liquid level and a risk of backflow in the sampling branch, the system first closes the corresponding control valve module and the anti-backflow check valve, and suspends sampling in that branch. After the liquid level stabilizes, the sampling protection module 9 first activates the self-cleaning flushing unit and the bypass discharge unit to discharge residual water samples and sediments in the sampling branch, and then executes the sample discarding process. The sample discarding process may include continuously draining water samples for a predetermined time or discharging a predetermined volume until the sampled water sample reaches a stable state, after which the system introduces the water sample into the sampling chamber 37 for detection. Through the above sampling protection process, the impact of heavy rainfall, backflow, and sediment blockage on the detection accuracy can be significantly reduced.

[0066] According to a further embodiment of the present invention, the system further includes a power supply and communication protection module 10. The power supply and communication protection module 10 may include a sealed housing, an anti-condensation heating or dehumidification unit, a surge protection unit, and a backup power unit. The power supply and communication protection module 10 is configured to maintain the continuous operation of the detection center module 3 and the data transmission module 4 under conditions of high humidity, salt spray, thunderstorms, or short-term mains power outages, and to temporarily store data during the outage in the edge cache module 45. The sealed housing may employ corrosion-resistant materials, waterproof joints, sealing rings, breathable waterproof membranes, and anti-corrosion coatings to improve the long-term stability of the equipment in humid and salty air environments. The anti-condensation heating or dehumidification unit can automatically activate based on the detection results of the housing temperature and humidity monitoring module 83, keeping the humidity inside the housing within a safe range. The surge protection unit may be located at the power input terminal, communication interface, and sensor interface to reduce the impact of transient overvoltages on the equipment during thunderstorms. The backup power unit may include a battery, supercapacitor, solar supplementary power supply unit, or other energy storage unit to maintain system operation during short-term mains power outages.

[0067] During a power outage, the power supply and communication protection module 10 can implement a tiered power supply strategy based on the remaining power. For example, when the power is sufficient, the system maintains the normal operation of the detection center module 3, data transmission module 4, and environmental adaptability module 8; when the power drops to the first threshold, the system reduces the regular sampling frequency but retains anomaly monitoring; when the power drops to the second threshold, the system prioritizes ensuring the edge buffer module 45, key sensors, and alarm communication functions. If the communication link is also interrupted, the edge buffer module 45 continuously saves detection data and device status data, and retransmits them in chronological order after power and communication are restored. Through the power supply and communication protection module 10, the system can maintain continuous monitoring capabilities in complex environments such as high humidity, salt spray, thunderstorms, and short-term power outages.

[0068] In some embodiments, the environmental adaptability module 8, sampling protection module 9, and power supply and communication protection module 10 can be linked with the cloud center module 5 and the monitoring center module 6. Specifically, the environmental adaptability module 8 uploads data on ambient temperature and humidity, rainfall intensity, liquid level changes, and shell humidity to the cloud center module 5; the cloud center module 5 adjusts thresholds and sampling strategies based on the above data; the monitoring center module 6 displays the environmental risk status and allows maintenance personnel to remotely initiate flushing, evacuation, dehumidification, or backup power supply detection. Thus, the system can not only monitor water quality parameters but also monitor and adapt to external environmental factors that affect detection accuracy and equipment reliability.

[0069] In some embodiments, the detection center module 3 can centrally serve multiple adjacent key nodes. In this case, the diversion sampling module 2 includes multiple sampling branches and multiple control valve modules, with water samples from different nodes entering the same sampling chamber 37 in a scheduled order. To avoid cross-contamination, the system can perform flushing and sample discarding steps when switching nodes, and distinguish data from different nodes using timestamps and node numbers. This centralized structure reduces the number of detection sensors, lowers equipment and maintenance costs, and is suitable for pipe network areas with high node density but limited space.

[0070] In some embodiments, some functions of the cloud-centric module 5 can be offloaded to edge computing devices. For example, in scenarios with poor communication conditions or high real-time requirements, the edge computing device can first perform preliminary data cleaning, anomaly screening, threshold judgment, and local alarms before uploading the data to the cloud-centric module 5. The cloud-centric module 5 then performs cross-regional data analysis, model training, contamination route deduction, and long-term trend prediction. Through cloud-edge collaboration, communication bandwidth pressure can be reduced while ensuring real-time performance.

[0071] In some embodiments, the dynamic node map can be linked with a geographic information system, a pipeline asset management system, a pump station control system, a valve control system, or an emergency command system. The pollution path generation module 65 can call upon real-time hydraulic models, pump station operating status, and valve opening / closing status to further improve the accuracy of pollution diffusion path prediction. The pollution source tracing module 66 can associate suspected pollution source results with historical emission records, enforcement records, or enterprise discharge outlet information to assist in on-site verification.

[0072] In some embodiments, the system can be configured with different combinations of detection indicators according to different scenarios. For water supply terminal nodes, the system can focus on detecting pH value, turbidity, residual chlorine, conductivity, dissolved oxygen, and heavy metal ion concentration; for drainage main nodes, the system can focus on detecting turbidity, organic pollutant content, conductivity, dissolved oxygen, and characteristic ion indicators; for nodes around industrial parks, the system can increase the detection priority of heavy metal ion concentration and organic pollutant content; for nodes at risk of rainwater and sewage mixing, the system can dynamically analyze changes in conductivity, turbidity, and organic pollutant content by combining rainfall and liquid level information.

[0073] Through the above embodiments, the online multi-parameter water quality monitoring system for water supply and drainage networks of the present invention can realize automatic diversion of water samples from key nodes, synchronous detection of multiple parameters, encrypted uploading of detection data, cloud-based predictive analysis, dynamic node diagram display, anomaly identification, pollution route generation, pollution source tracing, and equipment maintenance management. Furthermore, through the environmental adaptability module 8, sampling protection module 9, and power supply and communication protection module 10, the system can adapt to complex operating environments such as high temperature and humidity, heavy rainfall, strong winds and rain from the periphery of typhoons, saline and humid air along the coast, rapid changes in pipeline liquid levels, thunderstorms, and short-term power outages, reducing the impact of environmental factors on detection accuracy and the continuous operation capability of the equipment.

[0074] It should be understood that the modules described in the above embodiments can be implemented using hardware, software, firmware, or a combination thereof. The connection and control relationships between the modules can be adjusted according to the actual engineering deployment. The technical features of the diversion sampling module 2, detection center module 3, data transmission module 4, cloud center module 5, monitoring center module 6, system management and maintenance module 7, environmental adaptability module 8, sampling protection module 9, and power supply and communication protection module 10 in the above embodiments can be arbitrarily combined without conflict to form an online monitoring system suitable for different pipe network types, different node densities, different water quality risks, and different environmental conditions.

Claims

1. A multi-parameter online monitoring system for water quality in water supply and drainage networks, characterized in that, include: Water supply and drainage network system (1); A sampling module (2) is installed at at least one key node of the water supply and drainage network system (1) for drawing out water samples to be tested from the key node; The detection center module (3) is fluidly connected to the drainage sampling module (2) and is used to perform multi-parameter water quality detection on the water sample to be tested; The data transmission module (4) is connected to the detection center module (3) for encrypted transmission of detection data, timestamp information and location information; The cloud center module (5) is connected to the data transmission module (4) for preprocessing, data mining, deep learning analysis and predictive analysis of the received data; The monitoring center module (6) is connected to the cloud center module (5) for constructing a dynamic node graph based on the analysis results, performing anomaly detection, generating alarm signals, generating pollution routes and performing pollution source tracing. as well as The system management and maintenance module (7) is connected to the monitoring center module (6) for managing user permissions, device status, log recording, data storage backup and system maintenance.

2. The online monitoring system for multiple parameters of water quality in water supply and drainage networks according to claim 1, characterized in that, The diversion sampling module (2) includes a diversion pipeline module and a control valve module. The diversion pipeline module is connected between the key node of the water supply and drainage network system (1) and the detection center module (3). The control valve module opens or closes in response to the remote control command of the monitoring center module (6) to automatically collect and divert the water sample to be tested to the detection center module (3).

3. The online monitoring system for multiple parameters of water quality in water supply and drainage networks according to claim 1, characterized in that, The detection center module (3) includes a pH detection module (31), a turbidity detection module (32), a heavy metal ion concentration detection module (33), an organic pollutant content detection module (34), a dissolved oxygen detection module (35), and a conductivity detection module (36). The detection center module (3) also includes a sampling chamber (37) for accommodating the water sample to be tested and a temperature compensation and calibration module (38) for temperature compensation or calibration of at least one detection parameter.

4. The online monitoring system for multiple parameters of water quality in water supply and drainage networks according to claim 1, characterized in that, The data transmission module (4) includes a wireless communication module (41), a positioning module (42), a data encryption module (43), a timestamp marking module (44), and an edge cache module (45). The edge cache module (45) caches the detection data when the wireless link is abnormal, and retransmits the detection data in the order of data collection after the wireless link is restored.

5. The online monitoring system for multiple parameters of water quality in water supply and drainage networks according to claim 1, characterized in that, The cloud center module (5) includes a data preprocessing module (51), a data mining module (52), a deep learning algorithm module (53), a predictive analysis module (54), a model update module (55), and a threshold configuration module (56). The data preprocessing module (51) cleans, denoises, and normalizes the raw data. The threshold configuration module (56) configures anomaly judgment thresholds according to node type, historical baseline, or operating conditions.

6. The online monitoring system for multiple parameters of water quality in water supply and drainage networks according to claim 1, characterized in that, The monitoring center module (6) includes a dynamic node graph construction module (61), a visualization interaction module (62), an anomaly detection module (63), an alarm signal generation module (64), a pollution route generation module (65), a pollution source tracing module (66), and a remote control module (67). The dynamic node graph construction module (61) includes a node data integration module (611), a graphics rendering module (612), and a status marking module (613). The node data integration module (611) is used to integrate water quality data from each detection node. The graphics rendering module (612) is used to render the nodes and pipe segments of the water supply and drainage network system (1) into a visual graphic. The status marking module (613) is used to assign normal, risk, or abnormal status markings to the nodes based on the water quality detection results. The remote control module (67) is used to remotely control the opening and closing of the control valve module and adjust the detection parameters of the detection center module (3).

7. The online monitoring system for multiple parameters of water quality in water supply and drainage networks according to claim 1, characterized in that, It also includes an environmental adaptability module (8), which includes an environmental sensor (81), a rainfall and liquid level identification module (82), a shell temperature and humidity monitoring module (83), and a threshold compensation and sampling scheduling module (84). The threshold compensation and sampling scheduling module (84) dynamically adjusts the sampling frequency, anomaly judgment threshold, and compensation coefficient of at least one water quality parameter in environments such as high temperature and humidity, heavy rainfall, strong winds and rain outside typhoons, salty and humid air in coastal areas, or rapid changes in the liquid level of the pipeline network.

8. The online monitoring system for multiple parameters of water quality in water supply and drainage networks according to claim 7, characterized in that, It also includes a sampling protection module (9), which includes an anti-backflow check valve, a self-cleaning flushing unit, a bypass discharge unit and an anti-clogging filter. The sampling protection module (9) closes the affected sampling branch when there is heavy rainfall or the risk of backflow in the pipeline increases, and performs flushing, emptying and discarding of samples before resuming sampling.

9. The online monitoring system for multiple parameters of water quality in water supply and drainage networks according to claim 7, characterized in that, It also includes a power supply and communication protection module (10), which includes a sealed housing, an anti-condensation heating or dehumidification unit, a surge protection unit and a backup power supply unit. The power supply and communication protection module (10) maintains the continuous operation of the detection center module (3) and the data transmission module (4) under conditions of high humidity, salt spray, thunderstorms or short-term mains power outages, and temporarily stores the data during the interruption to the edge cache module (45).

10. A method for online monitoring of multiple parameters of water quality in a water supply and drainage network, applied to the online monitoring system for multiple parameters of water quality in a water supply and drainage network as described in any one of claims 1 to 9, characterized in that, The method includes: Water samples are collected and diverted at key nodes of the water supply and drainage network system (1) through the diversion sampling module (2); Multi-parameter water quality testing is performed on the diverted water sample through the detection center module (3); The detection data is timestamped, located, encrypted, and transmitted through the data transmission module (4); The cloud center module (5) performs data preprocessing, data mining, deep learning analysis and predictive analysis on the received data; The monitoring center module (6) displays the water quality status in real time, marks abnormal nodes, generates alarm signals, generates pollution routes, and performs pollution source tracing when anomalies occur; and The system management and maintenance module (7) performs user permission management, device status management, log management, data storage backup and system maintenance.

Images