Intelligent water level monitoring and rainwater collecting and processing system for underground garage

Abstract

The invention relates to the technical field of underground space water resource management, in particular to an intelligent water level monitoring and rainwater collecting and processing system for an underground garage. And the data acquisition module is used for acquiring water accumulation rate data subjected to temperature compensation, pollutant residue data based on dual verification of a bionic surface contact angle and a conductive film resistor, and drainage state data including flow, pressure and pipeline vibration frequency. According to the invention, multi-dimensional accurate data is collected through the environment adaptive monitoring unit, meteorological data and a scene feature library are accessed in combination with the dynamic scene decision module, three types of rainfall scenes are distinguished through a dynamic weight iteration algorithm, and an adaptive collection strategy is generated. The problems that in the prior art, water level monitoring and rainwater collection are lack of cooperation, and water accumulation scenes cannot be distinguished, so that polluted rainwater is collected mistakenly or high-quality rainwater is wasted are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of underground space water resource management, in particular to an intelligent water level monitoring and rainwater collection and treatment system for underground garage. BACKGROUND

[0002] The underground garage undertakes the function of vehicle parking in urban buildings. Its underground space characteristics make it prone to water accumulation during rainfall. If the rainwater can be reasonably collected and treated, it can be used for garage cleaning, landscaping irrigation, etc., which not only relieves the pressure of waterlogging but also realizes resource recycling. At present, in the industry, underground garage water level monitoring focuses on waterlogging early warning, and rainwater collection relies on fixed pipe systems. Although both are important links in the management of underground space water resources, they have not yet formed a deep collaborative technical system.

[0003] The existing disclosed invention patent, with the publication number CN116448206A, an underground garage waterlogging early warning monitoring system and method, constructs a monitoring system including a buried water level sensor, a LoRa communication module and a Web server, which can collect water level data in real time and realize early warning and drainage control, effectively solving the problem of waterlogging monitoring delay. However, this system is designed only for waterlogging scenarios and does not involve rainwater collection function, nor does it consider the accurate application of water level monitoring data in the rainwater collection link, which cannot provide adaptive decision support for rainwater collection.

[0004] In the existing technology, even if some schemes try to combine water level monitoring and rainwater collection, they can only trigger the collection action through a simple water level threshold, and cannot distinguish between "instant heavy rain waterlogging" and "continuous stable rainwater". When there is instant heavy rain waterlogging, rainwater will carry pollutants such as mud and oil on the ground of the garage, and direct collection will increase the difficulty of subsequent treatment and the water quality will be difficult to meet the standards. While continuous stable rainwater has less impurities and is a high-quality recycled water source. The existing scheme lacks judgment of the stability of water accumulation and the correlation characteristics of pollutants, resulting in either mis-collection of contaminated rainwater leading to a sharp increase in treatment cost, or missing the opportunity to collect high-quality rainwater causing resource waste, and cannot dynamically adjust the collection strategy according to the change of water accumulation, resulting in difficulty in balancing collection efficiency and water quality guarantee. SUMMARY

[0005] In view of the deficiencies of the prior art, the present application provides an intelligent water level monitoring and rainwater collection and treatment system for underground garage, which solves the problem of lack of deep collaboration between water level monitoring and rainwater collection in the prior art, and cannot distinguish between contaminated rainwater mis-collection or high-quality rainwater waste caused by water accumulation scenarios.

[0006] To achieve the above purpose, the present application is implemented by the following technical scheme: an intelligent water level monitoring and rainwater collection and treatment system for underground garage, comprising: An environmental adaptive monitoring unit for collecting temperature-compensated water accumulation rate data, pollution residue data based on dual verification of biomimetic surface contact angle and conductive film resistance, and drainage state data including flow rate, pressure, and pipe vibration frequency; A dynamic scenario decision module for receiving data from the environmental adaptive monitoring unit, accessing external real-time weather data, combining a built-in scenario feature library containing parameter thresholds corresponding to three types of rainfall scenarios, i.e., initial rainfall pollution period, stable rainfall period, and post-rainfall period, determining the current water accumulation scenario type through a dynamic weight iterative algorithm, and generating a collection strategy matching the current water accumulation scenario type, which includes a collection rate upper limit, a filtration accuracy level, and purification agent injection parameters; A self-maintenance collection control module for controlling the opening degree of the collection valve and the working mode of the filtration assembly according to the collection strategy matching the current water accumulation scenario type, and performing anti-backflow operation; A spectral linkage processing module for performing ultraviolet-visible absorption spectrum analysis on collected rainwater to identify pollutant composition and concentration, and dynamically adjusting the type and amount of purification agent based on the analysis results. The processed rainwater is stored after meeting the detection standards, or is returned and feedback adjustment signals to the dynamic scenario decision module if it does not meet the standards; A full-link feedback module for integrating edge computing, device health monitoring, and user interaction functions, converting raw monitoring data into feature parameters for use by the dynamic scenario decision module, and monitoring the operating status of key system components.

[0007] Further, the environmental adaptive monitoring unit comprises: A temperature-compensated water accumulation rate monitoring component containing a pair of corrosion-resistant ceramic probes, the surface of which is covered with a hydrophobic coating and integrated with a temperature sensing chip for real-time correction of temperature interference on water level sensing signals; A biomimetic pollutant sensing component with a micrometer-scale protrusion and a nanometer-scale coating forming a biomimetic lotus leaf hydrophobic structure, the biomimetic pollutant sensing component integrated with an optical sensor and a conductive film, the optical sensor for capturing changes in surface contact angle caused by pollutant attachment, and the conductive film for detecting resistance value changes; A dual-parameter drainage state monitoring component containing a flow meter, a pressure sensor, and a pipe vibration sensing module for analyzing pipe vibration frequency to determine pre-clogging signs.

[0008] Further, the dynamic scenario decision module comprises: A weather data interface unit for accessing real-time rainfall intensity and rainfall duration prediction data for the area where the underground garage is located; A scene feature library unit is configured to store a parameter threshold set corresponding to each of three scenes, i.e., a "initial rain pollution period", a "steady rain period", and an "end of rain period"; A dynamic weight calculation unit is configured to assign initial weights of the waterlogging rate data, the pollutant residue data, and the drainage state data based on the predicted scene type, and iteratively update the weight coefficients of each data through a self-learning mechanism.

[0009] Further, in the dynamic weight iteration algorithm executed by the dynamic weight calculation unit, the weight of each monitoring parameter at the nth decision is determined by the following formula: ; wherein, represents the weight of the mth monitoring parameter at the nth decision, represents the weight of the mth monitoring parameter at the nth decision, is a forgetting factor and , is an error evaluation factor fed back by the system after the nth decision, is a forgetting factor and is a forgetting factor and , is a forgetting factor and is a forgetting factor and is the total number of monitoring parameters.

[0010] Further, the self-maintaining collection control module comprises: A wear self-compensation valve group, whose valve core is made of a shape memory alloy material, and whose inner wall is provided with a replaceable wear-resistant bushing, the opening speed of the wear self-compensation valve group is linked with the received upper limit parameter of the collection rate; A self-cleaning filter assembly, which is provided with a high-pressure airflow channel and an elastic scraper between multiple layers of filter media, and is used to start reverse airflow blowing and simultaneously scrape off impurities when a preset fixed operation time threshold or an increase in filter resistance is detected; An intelligent anti-backflow system, which comprises an elastic sealing valve piece, a liquid level difference sensor, and a pipeline pressure relief valve, the liquid level difference sensor is used to monitor the liquid level difference between the drainage pipe network and the collection pipeline, and trigger the valve piece to close and the pressure relief valve to operate when the liquid level difference is abnormal.

[0011] Further, the spectrum linkage processing module comprises: A spectrum water quality analysis unit is configured to emit ultraviolet-visible light and receive transmission spectrum, and identify and quantify the concentration of petroleum organic matter and suspended solids by analyzing the absorption peaks at characteristic wavelengths within the ultraviolet-visible wavelength band of 200-800 nm. ​An adaptive purification unit, comprising at least two purifying agent dispensing devices for petroleum-based organic matter and suspended solids, and an ultrasonic degradation module, for linearly adjusting the dispensing amount of the corresponding purifying agent according to the quantitative results of the spectral water quality analysis unit, and starting the ultrasonic degradation module to process the refractory organic matter identified by the ultraviolet-visible absorption spectrum analysis; A closed-loop buffer unit, comprising a flow buffer chamber and an online water quality detector, for adjusting the outflow rate to a set range when collecting flow fluctuation, and returning rainwater to the pretreatment link when the water quality is not up to standard.

[0012] Further, the dispensing amount of the purifying agent in the adaptive purification unit is determined by the following formula: ; Wherein, is the dispensing amount of the purifying agent, is the concentration of the target pollutant detected by the spectral water quality analysis unit, is the dispensing ratio coefficient, is the baseline dispensing amount.

[0013] Further, the full-link feedback module comprises: An edge computing node for pre-processing the raw data of the environmental adaptive monitoring unit, extracting feature parameters and compressing data volume, while locally storing the preset historical data within 72 hours; A device health monitoring node, which is deployed with vibration, temperature and current sensors on the core working components of the self-maintenance collection control module and the spectral linkage processing module, for diagnosing the risk of component loosening or overloading by analyzing sensor data; A user interaction node for providing a visual interface to display the system operation status, receiving user's manual intervention instructions, and generating a periodic rainwater recycling report once a day.

[0014] Further, the device health monitoring node determines the loosening state of the component by analyzing the vibration sensor data, and the judgment basis is the degree of deviation of the vibration frequency from the rated working frequency band When the following formula is satisfied, it is determined that there is a loosening risk: ; Wherein, is the nominal working frequency of the component, is the preset frequency tolerance threshold.

[0015] ​Further, a two-way data channel is established between the edge computing node in the full-link feedback module and the dynamic scene decision module, the edge computing node pushes the characteristic parameters to the dynamic scene decision module, the dynamic scene decision module returns the generated collection strategy matching the current waterlogging scene type and the feedback adjustment signal from the spectrum linkage processing module to the edge computing node, for guiding the real-time response operation of the self-maintenance type collection control module.

[0016] Compared with the prior art, the beneficial effects of the present application are: The present application collects multi-dimensional accurate data through the environment self-adaptive monitoring unit, accesses meteorological data and scene characteristic library through the dynamic scene decision module, distinguishes three types of rainfall scenes through the dynamic weight iteration algorithm and generates adaptive collection strategies, solves the problem of pollution rainwater mis-collection or high-quality rainwater waste caused by lack of cooperation between water level monitoring and rainwater collection and inability to distinguish waterlogging scenes in the prior art. The self-compensation, self-cleaning and anti-backflow functions of the self-maintenance type collection control module ensure the stability of the collection process, the spectrum linkage processing module dynamically adjusts the purification scheme through spectrum analysis to ensure that the water quality meets the standards, the full-link feedback module realizes data preprocessing, equipment health monitoring and user interaction, and the two-way linkage of each module enables the system to dynamically optimize operation, which not only improves the rainwater collection efficiency and water quality guarantee capability, but also realizes resource recycling and utilization, and relieves the waterlogging pressure in underground garage. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 is a system architecture overview diagram of the present application; Figure 2 is an environment monitoring unit workflow diagram of the present application; Figure 3 is a dynamic decision logic diagram of the present application; Figure 4 is a spectrum purification control flowchart of the present application; Figure 5 is a full-link feedback mechanism diagram of the present application. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0019] Embodiment one Please refer to Figures 1-5The application provides an underground garage intelligent water level monitoring and rainwater collection and treatment system, which comprises an environment adaptive monitoring unit, a dynamic scene decision module, a self-maintenance type collection control module, a spectrum linkage processing module and a full-link feedback module.

[0020] The environment adaptive monitoring unit is used for collecting temperature-compensated water accumulation rate data, pollutant residue data based on bionic surface contact angle and double-verified conductive film resistance, and drainage state data containing flow, pressure and pipeline vibration frequency. The unit specifically comprises a temperature-compensated water accumulation rate monitoring piece, a bionic pollutant sensing piece and a double-parameter drainage state monitoring piece. The temperature-compensated water accumulation rate monitoring piece adopts a pair of corrosion-resistant ceramic probes, the surface of the probes is covered with a hydrophobic coating to reduce the influence of water adhesion on the monitoring accuracy, and a temperature sensing chip is integrated to correct the interference of temperature on the water level sensing signal in real time. For example, when the temperature sensing chip detects that the environmental temperature is lower than 5℃ in a low-temperature environment, a signal compensation algorithm is automatically started to offset the sensing signal attenuation caused by temperature, so as to ensure the accuracy of the water accumulation rate data. The surface of the bionic pollutant sensing piece is designed as a bionic lotus leaf hydrophobic structure composed of micron-level protrusions and nanometer-level coatings. This structure can reduce the adhesion of pollutants and facilitate the capture of surface property changes caused by pollutants. The optical sensor integrated in the sensing piece is used to capture the surface contact angle change caused by the adhesion of pollutants, and the conductive film is used to detect the resistance value change. Through double verification of the two detection methods, false judgments that may occur in a single detection method are avoided. For example, when the optical sensor detects that the contact angle decreases from 150° to 120° and the resistance value of the conductive film changes by more than 5%, it is determined that there is pollutant residue. The double-parameter drainage state monitoring piece comprises a flowmeter, a pressure sensor and a pipeline vibration sensing module. The pipeline vibration sensing module analyzes the pipeline vibration frequency to judge the clogging precursor. When a small amount of clogging occurs in the pipeline, the fluid flow state changes will cause the vibration frequency to change. The module collects vibration data in real time and compares them with the vibration frequency in the normal state to identify the clogging risk in time.

[0021] The dynamic scene decision-making module receives data from the environmental adaptive monitoring unit and accesses external real-time meteorological data. It combines this data with a built-in scene feature library containing parameter thresholds for three rainfall scenarios: the initial rain pollution period, the stable rainfall period, and the end of the rainfall period. Through a dynamic weighted iterative algorithm, it determines the current waterlogging scenario type and generates a collection strategy matching this type. This matching strategy includes a collection rate limit, filtration accuracy level, and purification agent dosage parameters. Specifically, this module includes a meteorological data interface unit, a scene feature library unit, and a dynamic weight calculation unit. The meteorological data interface unit accesses real-time rainfall intensity and duration prediction data for the underground parking garage area via a wireless communication module. The communication protocol uses TCP / IP to ensure data transmission stability and real-time performance. For example, it accesses the local meteorological department's API interface, updating rainfall data every 5 minutes. The scene feature library unit stores parameter threshold sets for three scenarios: "initial rain pollution period," "stable rainfall period," and "end of rainfall period." The parameter thresholds are set based on historical rainfall data, pollutant concentration monitoring data, and drainage system capacity for the area where the underground parking garage is located. For example, the water accumulation rate threshold for the initial rain pollution period is set to 5 mm / min, the pollutant residue data threshold is set to a contact angle ≤120° and a resistivity change rate ≥5%, and the flow rate threshold in the drainage status data is set to 2 m³ / min. 3 / h; the water accumulation rate threshold during stable rainfall periods is set at 3 mm / min, the pollutant residue data threshold is set at a contact angle ≥135° and resistivity change rate ≤2%, and the flow rate threshold is set at 3 m³ / min. 3 / h; the water accumulation rate threshold at the end of the rainfall period is set to 1 mm / min, and other parameters remain consistent with those during the stable rainfall period. The dynamic weight calculation unit is used to assign initial weights based on scenario type differences to the water accumulation rate data, pollutant residue data, and drainage status data according to the predicted scenario type, and iteratively updates the weight coefficients of each data point through a self-learning mechanism. In its executed dynamic weight iteration algorithm, the first... The weights of each monitoring parameter in the next decision are determined by the following formula:

[0022] in, Indicates the first The second decision-making time The weight of each monitoring parameter, Indicates the first The second decision-making time The weight of each monitoring parameter, Forgetting factor and , here The value is set based on the system's reliance on historical data and response speed requirements. For example, a value of 0.7 can be used when the system needs to respond more quickly to real-time data changes. Adjust to 0.5; For the first The error evaluation factor fed back by the system after the first decision is calculated as follows: After the collection strategy generated by this decision is implemented, the absolute value of the deviation between the actual and expected compliance rates of the collected rainwater is taken. If the actual compliance rate is higher than expected, then... Take the reciprocal of the deviation value; if it is lower than expected, take the deviation value to ensure that the error factor accurately reflects the decision-making effect. To represent the total number of monitored parameters, here These correspond to data on water accumulation rate, pollutant residue, and drainage status, respectively. The initial weights are assigned differently based on the scenario type. For example, during the initial rain pollution period, the initial weight for pollutant residue data is set to 0.5, the initial weight for water accumulation rate data is 0.3, and the initial weight for drainage status data is 0.2; during the stable rainfall period, the initial weight for water accumulation rate data is 0.4, the initial weight for pollutant residue data is 0.3, and the initial weight for drainage status data is 0.3; and during the end of the rainfall period, the initial weight for drainage status data is 0.4, the initial weight for water accumulation rate data is 0.3, and the initial weight for pollutant residue data is 0.3.

[0023] The self-maintaining collection control module controls the opening degree of the collection valve and the working mode of the filter assembly based on the collection strategy generated by the dynamic scenario decision module, and performs anti-backflow operation. This module includes a wear-compensating valve assembly, a self-cleaning filter assembly, and an intelligent anti-backflow system. The valve core of the wear-compensating valve assembly is made of shape memory alloy, specifically Ni-Ti shape memory alloy, with a phase transition temperature set at 25℃. Under normal operating temperature, it maintains a stable structural shape. When wear causes a decrease in sealing performance, it can restore its original shape through temperature adjustment to achieve self-compensation. The inner wall of the valve core is equipped with a replaceable wear-resistant bushing made of polytetrafluoroethylene (PTFE), which has good wear resistance and corrosion resistance. The replacement cycle is set according to the usage frequency, generally once every 6 months. The opening speed of the wear-compensating valve assembly is linked to the upper limit parameter of the received collection rate, using a linear linkage method. For example, when the upper limit of the collection rate is 2m... 3 At / h, the opening speed is 5mm / s, and when the upper limit of the collection rate is increased to 4m 3When t = h, the opening speed is synchronized to increase to 10 mm / s, ensuring that the valve opening speed matches the collection demand. The self-cleaning filter assembly is provided with a high-pressure airflow channel and an elastic scraper between the multi-layer filter medium. The filter medium uses a composite medium of quartz sand and activated carbon. The upper layer is quartz sand with a particle size of 0.5-1 mm, and the lower layer is activated carbon with a particle size of 0.3-0.5 mm. The airflow pressure of the high-pressure airflow channel is set to 0.6 MPa, and the air source is provided by a small air compressor. The elastic scraper is made of nitrile rubber material, which has good elasticity and wear resistance, and its installation position is closely attached to the surface of the filter medium. When the running time of the filter assembly reaches the preset fixed running time threshold (for example, 2 hours), or the filter resistance rises to 10 kPa detected by the pressure sensor, the system automatically starts reverse airflow purging, and the elastic scraper simultaneously scrapes the impurities on the surface of the filter medium. The purging time is set to 30 seconds, and the scraper movement speed is 2 mm / s, ensuring that the impurities can be effectively removed and the filtering efficiency of the filter assembly is restored. The intelligent anti-backflow system includes an elastic sealing valve, a liquid level difference sensor, and a pipeline pressure relief valve. The elastic sealing valve is made of fluororubber material and has excellent sealing performance, which can maintain sealing under high pressure. The measurement accuracy of the liquid level difference sensor is ±0.01 m, which is installed at the connection between the drainage pipe network and the collection pipeline to monitor the liquid level difference between them in real time. The opening pressure of the pipeline pressure relief valve is set to 0.3 MPa. When the liquid level difference sensor detects an abnormal liquid level difference, i.e., the liquid level in the collection pipeline is higher than that in the drainage pipe network and the difference exceeds 0.5 m, the system immediately triggers the elastic sealing valve to close, and at the same time opens the pipeline pressure relief valve for pressure relief operation to prevent the sewage in the drainage pipe network from backflowing into the collection pipeline and polluting the collected rainwater.

[0024] The spectrum linkage processing module is used for ultraviolet-visible absorption spectrum analysis of the collected rainwater to identify the pollutant composition and concentration, and dynamically adjusts the type and dosage of the purifying agent according to the analysis result. The treated rainwater is stored after detection, or is returned and feedback to the dynamic scene decision module. The module includes a spectrum water quality analysis unit, an adaptive purification unit and a closed-loop buffer unit. The spectrum water quality analysis unit is used for emitting ultraviolet-visible light and receiving transmission spectrum, adopts a xenon lamp as a light source, emits a continuous spectrum with a wavelength range of 200-800 nm, and irradiates the spectrum after being split by a grating monochromator into the rainwater sample flowing through a cuvette. The light path of the cuvette is 1 cm, and the material is quartz glass, which ensures good light transmission in the ultraviolet-visible band. The unit identifies and quantifies the concentration of petroleum organic matter and suspended solids by analyzing the absorption peaks at the characteristic wavelengths in the ultraviolet-visible band of 200-800 nm. The characteristic absorption peak of petroleum organic matter is located at 254 nm, and the characteristic absorption peak of suspended solids is located at 450 nm. The quantification process adopts the Lambert-Beer law, measures the absorbance at the characteristic wavelength, and calculates the pollutant concentration by combining the standard curve prepared in advance. The standard curve is obtained by fitting the absorbance after measuring the absorbance of petroleum organic matter standard solution and suspended solids standard solution with different concentrations. The adaptive purification unit includes at least two purifying agent dispensing devices for petroleum organic matter and suspended solids, and an ultrasonic degradation module. PAC is selected as the purifying agent for treating suspended solids, and powdered activated carbon is selected for treating petroleum organic matter. The two kinds of purifying agents are respectively stored in independent medicine tanks and are dispensed by metering pumps. The ultrasonic frequency of the ultrasonic degradation module is set to 20 kHz, and the power is set to 500 W, which is used to treat the refractory organic matter identified by the ultraviolet-visible absorption spectrum analysis. The unit linearly adjusts the dosage of the corresponding purifying agent according to the quantification result of the spectrum water quality analysis unit. The dosage of the purifying agent is determined by the following formula:

[0025] wherein, is the purifying agent dosage, unit: g / h; is the target pollutant concentration detected by the spectrum water quality analysis unit, unit: mg / L; is the dispensing ratio coefficient, which is determined according to the purification efficiency experiment of the purifying agent. For example, the value of the dispensing ratio coefficient of PAC is 0.05, and the value of the dispensing ratio coefficient of powdered activated carbon is 0.1. ​​The reference dosage is set as 5 g / h, which is used to ensure that there is enough purifying agent to maintain the basic purification effect even when the pollutant concentration is extremely low. When the spectral water quality analysis unit detects refractory organic matter, the ultrasonic degradation module is started simultaneously, and the degradation time is adjusted according to the pollutant concentration, generally for 10-30 minutes. The closed-loop buffer unit includes a flow buffer cavity and a water quality online detector. The volume of the flow buffer cavity is set to 1 m 3 , made of stainless steel, which has good corrosion resistance. Inside the cavity, there is a liquid level sensor for monitoring the liquid level in the cavity. By adjusting the opening of the water inlet valve and the water outlet valve, the water outlet rate is adjusted to the set range (for example, 2-3 m 3 / h) when the collection flow fluctuates; the water quality online detector uses a multi-parameter water quality sensor to detect the turbidity, COD, ammonia nitrogen and other indicators of rainwater in real time, with detection accuracies of ±1 NTU, ±5 mg / L and ±0.1 mg / L, respectively. When the water quality is detected to be substandard, i.e., the turbidity exceeds 10 NTU or the COD exceeds 50 mg / L or the ammonia nitrogen exceeds 1 mg / L, the rainwater is backflowed to the pretreatment link for re-purification treatment, and at the same time, an adjustment signal is fed back to the dynamic scene decision module to adjust the filtration accuracy level and the purifying agent dosage parameters in the collection strategy.

[0026] The full-link feedback module is used for edge computing, device health monitoring and user interaction functions, converts raw monitoring data into feature parameters for dynamic scene decision module, and monitors the running state of key components of the system. The module includes an edge computing node, a device health monitoring node and a user interaction node. The edge computing node uses an embedded processor, model STM32F407, for preprocessing the raw data of the environmental adaptive monitoring unit. The preprocessing process includes data filtering (using mean filtering algorithm, window size set to 5), outlier rejection (using 3σ criterion) and feature parameter extraction. The extracted feature parameters include the average value of the water accumulation rate, the change rate of the pollutant residual data, the peak value of the drainage state data, etc. The processed data is compressed using the LZ77 compression algorithm with a compression ratio of 3:1 to reduce data transmission volume. The node locally stores the preset historical data within 72 hours, using an SD card with a capacity of 32GB and a CSV format for data storage, which is convenient for subsequent calling and analysis. The device health monitoring node is deployed with vibration, temperature and current sensors on the core working components of the self-maintenance collection control module and the spectrum linkage processing module. The vibration sensor is a piezoelectric vibration sensor with a measurement range of 0-100Hz and an accuracy of ±0.1Hz. The temperature sensor is a PT100 platinum resistance sensor with a measurement range of -20℃-100℃ and an accuracy of ±0.5℃. The current sensor is a Hall current sensor with a measurement range of 0-5A and an accuracy of ±0.01A. The node diagnoses the risk of component loosening or overload by analyzing sensor data. The vibration sensor data is used to determine the loosening state of the component, and the judgment basis is the degree of deviation from the rated working frequency band, which is determined when the following formula is satisfied: deviation from the rated working frequency band

[0027] wherein, is the nominal working frequency of the component, for example, the nominal working frequency of the wear self-compensating valve group is 5Hz, and the nominal working frequency of the self-cleaning filter assembly is 8Hz; is the preset frequency tolerance threshold, which is set according to the working characteristics of the component, generally 0.5Hz; and are the lower and upper limits of the rated working frequency band, respectively, set to and ​When the temperature sensor detects that the component temperature exceeds 80℃, or the current sensor detects that the working current exceeds 1.2 times of the rated current, it is determined that there is an overload risk, and the system immediately issues a pre-warning signal. The user interaction node adopts an industrial touch screen with a size of 10 inches and a resolution of 1024x600, which is used to provide a visual interface to show the system running state, including real-time data of each monitoring parameter, current waterlogging scene type, collection strategy parameters, water quality detection results, equipment running state, etc.; at the same time, it receives manual intervention instructions from the user, such as manually adjusting the upper limit of the collection rate, starting the self-cleaning function, pausing collection, etc., the instructions are input through touch operation of the touch screen, and the system executes the input instructions after verification; the node generates a periodic rainwater recycling report once a day, the report generation time is set to 2 o'clock in the morning every day, and the report content includes the rainfall, collected rainwater, water quality compliance rate, equipment running time, energy consumption, etc. on that day, the report can be exported through a U disk or transmitted to the terminal equipment of the management personnel through the network.

[0028] The edge computing node in the full-link feedback module and the dynamic scene decision module have a bidirectional data channel established therebetween, and Ethernet is adopted for data transmission, with a transmission rate of 100 Mbps, ensuring the real-time nature of data transmission. The edge computing node pushes the processed feature parameters to the dynamic scene decision module in real time, providing data support for scene judgment and collection strategy generation; the dynamic scene decision module returns the generated collection strategy matching the current waterlogging scene type and the feedback adjustment signal from the spectrum linkage processing module to the edge computing node, and the edge computing node guides the immediate response operation of the self-maintaining collection control module according to these signals, for example, when receiving a signal to adjust the filtering precision level, immediately control the self-cleaning filter assembly to switch the combination mode of the filter medium, increase or decrease the filtering precision, to ensure that the system can quickly adapt to the changes of the waterlogging scene, and optimize the collection effect and water quality.

[0029] During the operation of the system, first, the environmental self-adaptive monitoring unit comprehensively collects the water accumulation related data of the underground garage, the water accumulation rate data compensated by temperature can avoid the interference of temperature change on water level monitoring, the pollutant residue data verified by the bionic surface contact angle and the conductive film resistance can improve the accuracy of pollutant detection, and the drainage state data including flow, pressure and pipeline vibration frequency can comprehensively reflect the operation status of the drainage system. Subsequently, the dynamic scene decision module combines external real-time weather data and the built-in scene feature library to accurately determine the current water accumulation scene type through a dynamic weight iterative algorithm, generate an adaptive collection strategy, and solve the problem of mis-collection of contaminated rainwater or waste of high-quality rainwater caused by the inability to distinguish different water accumulation scenes and fixed collection strategies in the prior art. Then, the self-maintenance collection control module executes the collection operation according to the collection strategy, the wear self-compensation valve group can automatically compensate for wear and tear, prolonging the service life, the self-cleaning filter assembly can remove impurities in time to ensure the filtering efficiency, and the intelligent anti-backflow system can effectively prevent sewage backflow and protect the initial water quality of the collected rainwater. Then, the spectrum linkage processing module accurately identifies the pollutant composition and concentration through ultraviolet-visible absorption spectrum analysis, dynamically adjusts the type and dosage of the purification agent, and combines ultrasonic degradation to handle difficult-to-degrade organic matter to ensure that the treated rainwater meets the standards. The closed-loop buffer unit can cope with flow fluctuations to avoid non-compliant rainwater entering the storage link. Finally, the full-link feedback module optimizes data transmission and processing efficiency through edge computing, the equipment health monitoring node can detect equipment failure risks in time, the user interaction node facilitates real-time monitoring and operation by management personnel, and the bidirectional data channel realizes the coordinated linkage between modules to ensure that the system can dynamically adjust according to the actual situation and continuously optimize the operation effect.

[0030] Embodiment Two In order to better enable those skilled in the art to fully understand and implement the present application, the specific implementation principles of the present application are further supplemented below in conjunction with a specific application scenario.

[0031] This embodiment selects an underground garage in a medium-grade residential area in a city as an application scenario. The underground garage has a building area of about 6500 square meters and is a two-story structure with 150 designed parking spaces. The surrounding area of the residential area has a concentrated green area that requires a certain amount of water resources for daily maintenance. At the same time, the underground garage is close to the entrance of the residential area, and rainwater backflow and ground pollutants accumulating with rainwater often occur during rainfall. Therefore, the demand for real-time water level monitoring and reasonable rainwater recovery is typical, and the system is deployed and operated in this scenario for actual application and verification.

[0032] In the system installation phase, the deployment of the environmental self-adaptive monitoring unit needs to be combined with the garage terrain and the distribution characteristics of pollutants: 6 water accumulation rate monitoring devices with temperature compensation are deployed, respectively installed at the bottom of the garage entrance ramp, near the underground two-layer elevator entrance, and in the low-lying areas of the four corners of the garage, with a spacing of about 8 meters between each monitoring device, ensuring that the key areas prone to water accumulation in the garage can be covered; 3 bionic pollutant sensing devices are deployed near the garbage bin storage point and the drainage grate inside the garage entrance, which are frequently visited by personnel and are prone to attach living impurities and a small amount of oil stains; 2 double-parameter drainage state monitoring devices are installed on the main drainage trunk and the rainwater collection branch pipe of the garage, respectively, of which the monitoring device on the drainage trunk is used to capture the operation state of the municipal drainage pipe network, and the monitoring device on the collection branch pipe is used to track the fluid parameters in the rainwater collection process. The hardware devices of the dynamic scene decision module are integrated in the control cabinet of the garage duty room, connected with the meteorological data interface unit through wired network, and access to the real-time meteorological service data provided by the city meteorological bureau, including rainfall intensity, cloud movement direction and rainfall duration prediction information. The parameter threshold setting of the scene feature library unit refers to the rainfall records of the community in the past three years (such as the average annual rainfall days, the maximum rainfall duration of a single time), the common types of pollutants on the garage floor (mainly the sand, fallen leaves and a small amount of domestic sewage residues brought in by residents) and the design index of the drainage system (the diameter of the drainage trunk is 250mm, the designed drainage capacity is 4m 3 / h), and finally determines the parameter range of the three types of rainfall scenes: the water accumulation rate threshold of the initial rain pollution period is 4.5mm / min, the pollutant residue data threshold is the contact angle ≤125° and the resistance change rate ≥4.5%, and the drainage flow threshold is 2.2m 3 / h; the water accumulation rate threshold of the stable rainfall period is 2.8mm / min, the pollutant residue data threshold is the contact angle ≥138° and the resistance change rate ≤2.2%, and the drainage flow threshold is 3.2m 3 / h; the water accumulation rate threshold of the end of rainfall period is 0.9mm / min, and the pollutant residue and drainage flow parameters follow the stable rainfall period standard.

[0033] The core components of the self-maintenance collection control module need to be installed to adapt to the pipe layout of the community garage: the wear self-compensating valve group is installed at the end of the three rainwater collection branch pipes, the valve core uses Ni-Ti shape memory alloy, and the wear-resistant lining of the inner wall is made of polytetrafluoroethylene material, and the replacement cycle of the lining is preset to be once every 8 months according to the rainfall frequency of the community; the self-cleaning filter assembly is installed on the main collection pipeline after the collection branch pipes are collected, the filter medium uses a composite structure of quartz sand and activated carbon, and according to the characteristics that the ground pollutants in the community are small, the quartz sand particle size is adjusted to 0.4-0.9mm, the activated carbon particle size is 0.3-0.5mm, and the gas source of the high-pressure airflow channel is provided by a small air compressor beside the garage duty room; the intelligent anti-backflow system is installed at the connection node of the collection main pipeline and the municipal drainage main pipe, the measuring probe of the liquid level difference sensor is flush with the inner wall of the pipeline, which ensures that the liquid level change on both sides can be accurately captured, and the opening pressure of the pipeline pressure relief valve is set according to the daily water pressure of the municipal pipe network. The units of the spectrum linkage processing module are concentrated in the special equipment room on the second floor of the garage underground, the area of the equipment room is about 15 square meters, the light source of the spectrum water quality analysis unit uses xenon lamp, and the cuvette is made of quartz material, and the regular cleaning cycle is set to be once every two weeks; the storage tanks of the self-adaptive purification unit store polyaluminum chloride for treating suspended solids and powdered activated carbon for adsorbing organic matter respectively, the liquid level sensor of the storage tank is linked with the control cabinet of the duty room, and the low liquid level automatically prompts to add medicine; the flow buffer cavity of the closed-loop buffer unit is connected with the rainwater storage pool underground, the volume of the storage pool is designed according to the greening irrigation demand of the community, and the liquid level sensor in the buffer cavity is used to adjust the water inflow and outflow rate to ensure the stability of the water flow.

[0034] The edge computing node of the full-link feedback module shares the control cabinet with the dynamic scene decision module, an embedded processor is used to pre-process the raw data of the environment adaptive monitoring unit, including filtering, outlier rejection and feature parameter extraction, the extracted feature parameters include the change trend of the water accumulation rate of each monitoring point, the fluctuation range of the pollutant residual data and the mean value of the drainage pipeline vibration frequency, and the pre-processed data is transmitted to the dynamic scene decision module through the internal bus; the sensors of the equipment health monitoring node are deployed on the key moving parts: the vibration sensors are installed on the valve core drive mechanism of the wear self-compensating valve group, the scraper motor of the self-cleaning filter assembly and the transducer of the ultrasonic degradation module, the temperature sensors are attached to the shells of the motors and the valve body of the valve group, and the current sensors are connected in series in the power supply circuit of each power unit; the user interaction node uses a 10-inch industrial touch screen, which is installed on the operation table in the duty room, the interface layout is divided into real-time monitoring area, scene decision area, equipment status area and historical data area, which is convenient for management personnel to intuitively view system operation information.

[0035] After the system was put into daily operation, during a rainfall process in autumn, the environmental self-adaptive monitoring unit first started data collection: the water accumulation rate monitoring unit with temperature compensation collected the water accumulation rate at the bottom of the entrance ramp of the garage, which was 4.2 mm / min, and the water accumulation rate at other monitoring points was between 3.8-4.3 mm / min; the bionic pollutant sensing unit detected that the surface contact angle was 120°, and the resistance change rate of the conductive film was 4.8%; the double-parameter drainage state monitoring unit showed that the flow rate of the drainage main pipe was 2.0 m 3 / h, the flow rate of the collection branch pipe was 1.8 m 3 / h, and the pipeline vibration frequency was 3.8 Hz. After the edge computing unit preprocessed these raw data, the feature parameters were pushed to the dynamic scene decision module, and the meteorological data interface unit obtained the current rainfall intensity of 15 mm / h, and the meteorological department predicted that the duration of this rainfall was about 1.5 hours. The dynamic scene decision module called the scene feature library, combined with the dynamic weight iterative algorithm (the forgetting factor a was preset to 0.65), and assigned the initial weights (0.3, 0.5, and 0.2, respectively) to the water accumulation rate, pollutant residue, and drainage state three types of data in the initial rain pollution period. Through weight calculation and scene threshold comparison, it was determined that the current was in the initial rain pollution period, and then the corresponding collection strategy was generated: the upper limit of the collection rate was 1.2 m 3 / h, the filter precision level was set to high level (the double-layer filter medium of quartz sand and activated carbon was enabled), the purification agent delivery parameters were the delivery proportion coefficient β of polyaluminum chloride = 0.045, the reference delivery amount γ = 4.5 g / h, and the delivery proportion coefficient β of powdered activated carbon = 0.09, the reference delivery amount γ = 6 g / h.

[0036] After the self-maintenance type collection control module received the collection strategy, it controlled the wear self-compensation valve group at the end of the three collection branch pipes to adjust the opening degree at an opening speed of 2.8 mm / s, to ensure that the total collection rate did not exceed 1.2 m 3 / h; the self-cleaning filter assembly switched to the double-layer filter mode, and set to start the self-cleaning process every 1.2 hours, the pressure in the high-pressure airflow channel was maintained at 0.55 MPa during self-cleaning, and the movement speed of the elastic scraper was 1.8 mm / s; the liquid level difference sensor of the intelligent anti-backflow system monitored the liquid level difference between the collection main pipe and the municipal drainage main pipe in real time, at this time the monitored liquid level difference was 0.25 m, which was within the preset normal range, and the elastic sealing valve remained open, and the pipeline pressure relief valve was in the closed standby state.

[0037] The collected rainwater is transported to the spectrum linkage processing module through the pipeline, and the spectrum water quality analysis unit starts the ultraviolet-visible absorption spectrum detection. The absorbance signal is detected at a wavelength of 254 nm, and the absorbance signal corresponding to the suspended solids is detected at a wavelength of 450 nm. The concentration of petroleum organic matter and suspended solids in the current rainwater is calculated by the Lambert-Beer law. The adaptive purification unit calculates the real-time dosing amount of polyaluminum chloride and powdered activated carbon according to the concentration value according to the dosing formula D = β·C p + γ, and accurately injects the rainwater treatment flow channel through the metering pump. At the same time, due to the detection of a small amount of refractory organic matter, the ultrasonic degradation module (ultrasonic frequency 20 kHz, power 480 W) is started. After the rainwater is purified, it enters the closed-loop buffer unit, and the liquid level sensor in the buffer cavity stabilizes the water outlet rate between 2.2-2.5 m 3 / h. The water quality online detector detects the turbidity, COD and ammonia nitrogen indexes of the rainwater in real time. When it is detected that each index meets the community rainwater reuse standard, the rainwater is transported to the underground storage tank. If the turbidity is detected to be out of the standard range at a certain moment, the rainwater is sent back to the pretreatment link through the reflux pipeline, and an adjustment signal is fed back to the dynamic scene decision module. The dynamic scene decision module adjusts the filtration accuracy level according to the signal.

[0038] During the entire rainfall process, the device health monitoring node of the whole link feedback module continuously collects sensor data: the vibration sensor detects that the deviation of the vibration frequency of the valve driving mechanism from the nominal working frequency is less than the preset frequency tolerance threshold δ (δ = 0.45 Hz), and it is determined that there is no loosening risk; the temperature sensor monitors that the motor shell temperature is highest at 62℃, which is lower than the preset overload temperature threshold; the current sensor detects that the working current of each power unit is within the rated current range, and no overload phenomenon occurs. The real-time monitoring area of the user interaction node synchronously displays the running parameters of each module, and the management personnel views the data through the touch screen without manual intervention operation. The system automatically operates according to the preset logic throughout the process.

[0039] After this rainfall ends, the user interaction node generates a rainwater recycling report according to the preset period. The report content includes the start and end time of this rainfall, the rainfall intensity of each period, the scene type determined by the system and the corresponding collection strategy adjustment record, the start number of the self-cleaning filter assembly, the operation time of the ultrasonic degradation module, and the working time and energy consumption data of each core device; The collected rainwater will be used for irrigation of greenery in the community, daily cleaning of the garage floor and water replenishment of the landscape pool. Through this actual application, it is verified that the deployment feasibility and operation stability of the invention in the underground garage scene of the residential community can complete the whole process operation of water level monitoring, scene judgment, rainwater collection and treatment according to the preset logic, and adapt to the actual needs of such scenes.

[0040] It is to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting; it is not intended to exclude myriad other embodiments of the present application that other inventors can develop based on the same general inventive concepts embodied by the described embodiments. That is, although the present application is described in terms of particular embodiments and implementations, it is to be understood that the terminology used is for the purpose of descriptive clarity and that it should be taken in its broadest possible sense. For example, the terms "a", "an", and "the" include both singular and plural referents unless the context clearly dictates otherwise. The terms "comprises", "comprising", "includes", "including" and the like can be used in conjunction with the term "consisting of to include the elements or steps listed after such conjunctive language, but not to the exclusion of other elements or steps. The singular forms "a", "an" and "the" include plural referents unless the context clearly dictates otherwise.

[0041] While the embodiments of the application have been shown and described herein, it is to be understood that the application is not limited to these embodiments. Rather, many modifications, changes and substitutions are intended to fall within the scope of the present application, which is limited only by the scope of the claims hereinafter appended.

Claims

1. A smart water level monitoring and rainwater harvesting and treatment system for underground parking garages, characterized in that, include: The environmental adaptive monitoring unit is used to collect temperature-compensated water accumulation rate data, pollutant residue data verified by both biomimetic surface contact angle and conductive film resistance, and drainage status data including flow rate, pressure and pipe vibration frequency. The dynamic scene decision module is used to receive data from the environmental adaptive monitoring unit and access external real-time meteorological data. It combines the built-in scene feature library containing parameter thresholds for three types of rainfall scenarios: the initial rain pollution period, the stable rainfall period, and the end of the rainfall period. The module uses a dynamic weight iterative algorithm to determine the current water accumulation scenario type and generate a collection strategy that matches the current water accumulation scenario type. The collection strategy that matches the current water accumulation scenario type includes the upper limit of the collection rate, the filtration accuracy level, and the purification agent dosing parameters. The self-maintaining collection control module is used to control the opening degree of the collection valve and the working mode of the filter components according to the collection strategy that matches the current water accumulation scenario type, and to perform anti-backflow operation. The spectral linkage processing module is used to perform ultraviolet-visible absorption spectral analysis on the collected rainwater to identify the components and concentrations of pollutants, and dynamically adjust the type and dosage of purification agent based on the analysis results. The treated rainwater is stored after it meets the standards, and if it does not meet the standards, it is returned and the adjustment signal is fed back to the dynamic scene decision module. The end-to-end feedback module integrates edge computing, device health monitoring, and user interaction functions. It transforms raw monitoring data into feature parameters for use by the dynamic scene decision module and monitors the operating status of key system components.

2. The intelligent water level monitoring and rainwater harvesting and treatment system for underground parking garages according to claim 1, characterized in that, The environmental adaptive monitoring unit includes: A temperature-compensated water accumulation rate monitoring device includes a pair of corrosion-resistant ceramic probes. The probe surfaces are covered with a hydrophobic coating and integrated with a temperature sensing chip for real-time correction of temperature interference with the water level sensing signal. A biomimetic pollutant sensor has a surface with a lotus leaf-like hydrophobic structure composed of micron-level protrusions and a nano-level coating. The biomimetic pollutant sensor integrates an optical sensor and a conductive film. The optical sensor is used to capture changes in surface contact angle caused by pollutant adhesion, and the conductive film is used to detect changes in resistance value. A dual-parameter drainage status monitoring device includes a flow meter, a pressure sensor, and a pipeline vibration sensing module. The pipeline vibration sensing module is used to determine the precursors of blockage by analyzing the pipeline vibration frequency.

3. The intelligent water level monitoring and rainwater harvesting and treatment system for underground parking garages according to claim 1, characterized in that, The dynamic scenario decision-making module includes: The meteorological data interface unit is used to access real-time rainfall intensity and rainfall duration forecast data for the area where the underground parking garage is located. The scene feature library unit is used to store the parameter threshold sets corresponding to the three types of scenes: "initial rain pollution period", "stable rainfall period" and "end of rainfall period". The dynamic weight calculation unit is used to assign initial weights based on scenario type differences to the water accumulation rate data, pollutant residue data, and drainage status data according to the predicted scenario type, and to iteratively update the weight coefficients of each data through a self-learning mechanism.

4. The intelligent water level monitoring and rainwater harvesting and treatment system for underground parking garages according to claim 3, characterized in that, In the dynamic weight calculation unit executing the dynamic weight iteration algorithm, the first... The weights of each monitoring parameter in the next decision are determined by the following formula: ； in, Indicates the first The second decision-making time The weight of each monitoring parameter, Indicates the first The second decision-making time The weight of each monitoring parameter, Forgetting factor and , For the first Error evaluation factors fed back by the system after each decision. This represents the total number of monitored parameters.

5. The intelligent water level monitoring and rainwater harvesting and treatment system for underground parking garages according to claim 1, characterized in that, The self-maintaining collection control module includes: The wear-compensating valve assembly has a valve core made of shape memory alloy and a replaceable wear-resistant bushing on the inner wall. The opening speed of the wear-compensating valve assembly is linked to the upper limit parameter of the received collection rate. The self-cleaning filter assembly has a high-pressure airflow channel and elastic scraper between multiple layers of filter media. When a preset fixed operating time threshold is reached or when an increase in filter resistance is detected, the reverse airflow is activated to purge and scrape off impurities simultaneously. The intelligent anti-backflow system includes a resilient sealing valve, a liquid level difference sensor, and a pipeline pressure relief valve. The liquid level difference sensor is used to monitor the liquid level difference between the drainage network and the collection pipeline, and triggers the valve to close and relieve pressure when the liquid level difference is abnormal.

6. The intelligent water level monitoring and rainwater harvesting and treatment system for underground parking garages according to claim 1, characterized in that, The spectral linkage processing module includes: The spectral water quality analysis unit is used to emit ultraviolet-visible light and receive transmission spectra. It identifies and quantifies the concentration of petroleum organic matter and suspended solids by analyzing the absorption peaks at characteristic wavelengths in the 200-800nm ​​ultraviolet-visible band. An adaptive purification unit includes at least two purification agent dosing devices for petroleum-based organic matter and suspended solids, and an ultrasonic degradation module. It is used to linearly adjust the dosage of the corresponding purification agent according to the quantitative results of the spectral water quality analysis unit, and to activate the ultrasonic degradation module to process the recalcitrant organic matter identified by ultraviolet-visible absorption spectroscopy. The closed-loop buffer unit includes a flow buffer chamber and an online water quality detector, which is used to adjust the outflow rate to a set range when collecting flow fluctuations, and to return rainwater to the pretreatment stage when the water quality does not meet the standards.

7. The intelligent water level monitoring and rainwater harvesting and treatment system for underground parking garages according to claim 6, characterized in that, The amount of purifying agent added in the adaptive purification unit Determined by the following formula: ； in, This refers to the dosage of the purification agent. The concentration of the target pollutant detected by the spectral water quality analysis unit. This is the distribution ratio coefficient. This is the baseline deployment volume.

8. The intelligent water level monitoring and rainwater harvesting and treatment system for underground parking garages according to claim 1, characterized in that, The end-to-end feedback module includes: Edge computing nodes are used to preprocess the raw data of the environmental adaptive monitoring unit, extract feature parameters and compress the data volume, and at the same time store historical data within a preset 72 hours locally. The equipment health monitoring node has vibration, temperature and current sensors deployed on the core working components of the self-maintenance collection and control module and the spectral linkage processing module, which are used to diagnose the risk of component loosening or overload by analyzing sensor data. The user interaction node provides a visual interface to display the system's operating status, receives manual intervention commands from users, and generates a daily periodic rainwater harvesting report.

9. The intelligent water level monitoring and rainwater harvesting and treatment system for underground parking garages according to claim 8, characterized in that, The equipment health monitoring node determines the looseness of components by analyzing vibration sensor data, based on the vibration frequency. Deviation from rated operating frequency band The degree to which a structure is considered at risk of loosening is determined when the following formula is met: ； in, This refers to the nominal operating frequency of the component. This is the preset frequency tolerance threshold.

10. The intelligent water level monitoring and rainwater harvesting and treatment system for underground parking garages according to claim 1, characterized in that, A bidirectional data channel is established between the edge computing node in the end-to-end feedback module and the dynamic scene decision module. The edge computing node pushes feature parameters to the dynamic scene decision module, and the dynamic scene decision module sends back the generated collection strategy that matches the current water accumulation scene type and the feedback adjustment signal from the spectral linkage processing module to the edge computing node to guide the real-time response operation of the self-maintaining collection control module.

Images