A cleaning system and method for an environmental monitoring five-parameter sensor

Abstract

This application discloses a cleaning system for five-parameter environmental monitoring sensors, including a sensor data monitoring unit, an intelligent decision-making unit, an adaptive cleaning execution unit, and a closed-loop verification unit. The sensor data monitoring unit collects water quality data in real time. The intelligent decision-making unit obtains corresponding coefficients, combines them with weight parameters corresponding to the water quality data, and uses historical trend correction parameters and their weight parameters to determine whether each sensor needs cleaning and the appropriate cleaning method. The adaptive cleaning execution unit cleans only the sensors that require cleaning. The closed-loop verification unit verifies the cleaning effect. This system can make cleaning decisions based on an adaptive algorithm that considers both real-time and historical pollution levels, avoiding waste of cleaning resources and cleaning delays. It can also perform more thorough cleaning on each sensor using a method that matches its pollution characteristics, avoiding over-cleaning of other sensors. This application also discloses a cleaning method for five-parameter environmental monitoring sensors.

Description

Technical Field

[0001] This invention belongs to the field of environmental monitoring technology, and in particular relates to a cleaning system and method for a five-parameter environmental monitoring sensor. Background Technology

[0002] The five-parameter sensor is a pH sensor, conductivity sensor, dissolved oxygen sensor, and turbidity sensor (including a conventional five-parameter core monitoring module) used for environmental monitoring. Since the measured values ​​of pH, conductivity, and dissolved oxygen are all affected by temperature, a temperature sensor is built into each of the pH, dissolved oxygen, and conductivity sensors. Data is collected in real time by the temperature sensor and dynamic compensation is performed to ensure monitoring accuracy. Therefore, when measuring pH in water, the temperature in the water is also measured, and a separate temperature sensor is no longer required. Using this five-parameter sensor, continuous monitoring of key physicochemical indicators of water bodies can be achieved, reflecting the basic chemical properties and physical characteristics of water bodies. This is of great significance for judging water quality and the degree of environmental pollution. Its monitoring targets cover a variety of water bodies, including drinking water sources, groundwater, surface water, and pollution sources. Among them, drinking water sources have fewer suspended solids and slower pollution, resulting in a long sensor pollution cycle. Pollution sources (such as sewage treatment plant outlets) have more suspended solids and denser microorganisms, resulting in a short pollution cycle. The pollution rate of surface water / groundwater is between the two mentioned above, and it is easily affected by rainfall and runoff, resulting in sudden increases in turbidity (e.g., turbidity > 200 NTU).

[0003] Five-parameter water quality analyzers equipped with the aforementioned five-parameter sensors can achieve continuous, real-time, and automatic monitoring of water quality data. However, since the measured objects are mostly untreated source water containing impurities such as suspended solids, algae, and microorganisms, the sensors are prone to contamination if immersed in it for a long time, leading to performance drift. Specific contamination manifestations are as follows: Inadequate cleaning of the glass electrode of the pH sensor can easily lead to the formation of white crystals such as CaCO3, colloidal biofilms, or an oily sheen on the surface; the salt bridge pores in the liquid network of the reference electrode are easily blocked, resulting in decreased measurement accuracy, performance degradation, and shortened service life; Inadequate cleaning of the conductivity sensor's electrode can easily lead to the deposition of white scale such as CaCO3 and CaSO4, as well as Fe2O3 (reddish-brown) on the surface. Metal oxides such as MnO2 (black) can easily cause bridging between electrodes due to the accumulation of suspended matter, leading to decreased measurement accuracy, failure of temperature compensation, and electromagnetic field distortion. Inadequate cleaning of dissolved oxygen sensors can cause the formation of viscous biofilms (50-200 μm thick) on the surface of the fluorescent film, resulting in a 30% to 70% decrease in fluorescence quenching efficiency. Inorganic salt crystallization can change the refractive index of the optical path, and oil stains can form a light barrier (up to 90% light path obstruction). Inadequate cleaning of turbidity sensors can cause water stains / fingerprints to remain on the optical window (changing the refractive index by ±0.05), form biofilms (transmittance decreases by 80% when the thickness is >50 μm), or CaCO3 crystals (causing a scattering angle shift of 5° to 15°).

[0004] In summary, to ensure the accuracy and reliability of monitoring data, timely and efficient cleaning of sensors is essential. However, existing cleaning technologies generally employ a "one-size-fits-all" approach with fixed cycles, failing to consider the varying pollution rates of different water bodies, such as drinking water sources (slow-pollution) and pollution sources (fast-pollution). Cleaning timing relies on timed triggers or manual judgment, lacking integration with sensor monitoring data and the ability to make intelligent decisions based on actual pollution conditions. Furthermore, they fail to provide differentiated cleaning for different sensors based on specific pollution situations (such as crystallization in pH probes or fluorescent film contamination in dissolved oxygen probes), resulting in incomplete cleaning and inaccurate monitoring data. Measurement data deviations (such as dissolved oxygen measurement deviation of ±5-15%, turbidity transmittance decrease of 80%), and the lack of verification measures for whether pollution exceeds the standard, lead to the following problems: over-cleaning, which damages the sensor coating / cap, resulting in waste of cleaning agents and water resources, and shortening the sensor lifespan; poor cleaning effect, with residual pollutants, and no supplementary cleaning or verification measures after pollution accumulation, causing data deviations; no additional cleaning is performed when there is a sudden increase in turbidity, making it impossible to accurately reflect the actual pollution risk, and the sensor becomes contaminated with pollutants, such as the turbidity optical window transmittance decreasing by 80%, causing turbidity data to become invalid. Summary of the Invention

[0005] To address the aforementioned issues, this invention provides a cleaning system and method for five-parameter environmental monitoring sensors. This system enables cleaning decisions based on an adaptive algorithm that considers both real-time and historical pollution levels, improving the accuracy of cleaning triggering, avoiding waste of cleaning resources and cleaning delays. Furthermore, it allows for more thorough cleaning of each sensor using a method tailored to its specific pollution characteristics, and verifies the cleaning results to ensure compliance with standards, preventing over-cleaning of other sensors and potential damage.

[0006] The present invention provides a cleaning system for a five-parameter environmental monitoring sensor, comprising a sensor data monitoring unit, an intelligent decision-making unit, an adaptive cleaning execution unit, and a closed-loop verification unit;

[0007] The intelligent decision-making units are all communicatively connected to the sensor data monitoring unit, the adaptive cleaning execution unit, and the closed-loop verification unit.

[0008] The sensor data monitoring unit is connected to the sensor for real-time water quality data collection;

[0009] The intelligent decision-making unit is used to obtain the corresponding coefficients based on the water quality data, and combine the weight parameters corresponding to the water quality data, as well as the historical trend correction parameters and their weight parameters, to obtain a conclusion on whether each sensor needs to be cleaned and the cleaning method for the sensors that need to be cleaned.

[0010] The adaptive cleaning execution unit is used to call the corresponding cleaning process according to the conclusion and the cleaning method, and only clean the sensors that need to be cleaned.

[0011] The closed-loop verification unit is used to verify the cleaning effect of the sensor after cleaning. If the verification fails, it feeds back to the intelligent decision-making unit to upgrade the cleaning level and continue cleaning.

[0012] Preferably, the cleaning system for the five-parameter environmental monitoring sensor mentioned above also includes a remote linkage unit that is communicatively connected to the intelligent decision-making unit. This unit is used to receive the cleaning process data when the closed-loop verification unit passes the verification, and then transmit the data to the intelligent decision-making unit for strategy optimization based on historical data.

[0013] Preferably, in the cleaning system for the five environmental monitoring sensors described above, the adaptive cleaning execution unit includes a sensor identification component, four sets of cleaning components, and a media supply component. The four sets of cleaning components are located inside the cleaning container and correspond to the pH sensor, conductivity sensor, dissolved oxygen sensor, and turbidity sensor, respectively. The media supply component is connected to the pure water container, cleaning agent container, and reference liquid container, respectively, through a switching component. The cleaning process includes the cleaning components, media, and cleaning procedures used for each type of sensor.

[0014] Preferably, in the cleaning system for the five-parameter environmental monitoring sensor described above, the sensor identification component is a sensor ID chip reader, used to read the chip ID on the sensor surface to identify the sensor type, and to call the cleaning components and cleaning process corresponding to the sensor type.

[0015] Preferably, in the cleaning system for the five-parameter environmental monitoring sensor described above, the cleaning component includes a soft brush, a nozzle, and an ultrasonic transducer disposed within the cleaning container, and the nozzle is connected to the media supply component, which includes a peristaltic pump for supplying the media required for the cleaning process to the corresponding cleaning component and spraying it onto the sensor from the corresponding nozzle.

[0016] Preferably, in the cleaning system of the above-mentioned environmental monitoring five-parameter sensor, the closed-loop verification unit includes a deviation calculation module and a verification container connected to the reference liquid container;

[0017] The verification container can be filled with reference liquid through the reference liquid container, and the sensor immersed in it can be immersed in the reference liquid. After the reference liquid is drained, it can be re-entered into the sensor to obtain the measured value.

[0018] The deviation calculation module is used to calculate the deviation rate between the measured value and the benchmark value, and to determine whether the cleaning has been completed based on the deviation rate.

[0019] Preferably, in the cleaning system for the five environmental monitoring sensors described above, the remote linkage unit includes an edge computing module, a cloud platform communication module, and a remote control interface.

[0020] This invention provides a cleaning method for an environmental monitoring five-parameter sensor, utilizing the environmental monitoring five-parameter sensor cleaning system as described in any of the above claims, comprising:

[0021] Real-time water quality data collection;

[0022] Based on the water quality data, the corresponding coefficients are obtained, and combined with the weight parameters corresponding to the water quality data, as well as the historical trend correction parameters and their weight parameters, a conclusion is drawn as to whether each sensor needs to be cleaned and the cleaning method for the sensors that need to be cleaned.

[0023] Based on the conclusions and the cleaning method, the corresponding cleaning process is invoked, and cleaning is performed only on the sensors that need to be cleaned.

[0024] The cleaning effect of the sensor after cleaning is verified. If the verification fails, the intelligent decision-making unit is notified to upgrade the cleaning level and continue cleaning.

[0025] Preferably, in the above-mentioned cleaning method for the five environmental monitoring sensors, the step of obtaining the corresponding coefficients based on the water quality data, and combining the weight parameters corresponding to the water quality data, as well as the historical trend correction parameters and their weight parameters, to determine whether each sensor needs cleaning and the cleaning method for the sensors that need cleaning includes:

[0026] According to the additional cleaning execution decision formula, when W1*K1+W2*K2+W3*K3≥Decision threshold T x When the time is right, the corresponding sensor is cleaned. Here, W1 is the turbidity weight, W2 is the real-time contamination weight, W3 is the historical contamination weight, x=1 corresponds to the pH cleaning decision threshold, x=2 corresponds to the conductivity cleaning decision threshold, x=3 corresponds to the dissolved oxygen cleaning decision threshold, x=4 corresponds to the turbidity cleaning decision threshold, K1 is the turbidity coefficient, K2 is the real-time contamination coefficient specific to each parameter, and K3 is the historical contamination coefficient.

[0027] Preferably, in the above-mentioned cleaning method for the five-parameter environmental monitoring sensor, the step of verifying the cleaning effect of the sensor after cleaning, and feeding back to the intelligent decision-making unit when the verification fails to upgrade the cleaning level and continue cleaning, includes:

[0028] The verification container is filled with reference liquid through the reference liquid container, and the sensor is immersed in it. After the reference liquid is drained, the reference liquid is re-entered into the reference liquid container, immersed in the sensor and stabilized for a period of time to obtain the measured value.

[0029] Calculate the deviation rate between the measured value and the benchmark value, and determine whether the cleaning has been completed based on the deviation rate. If the cleaning is complete, resume monitoring; otherwise, perform a second cleaning and re-verify. If the cleaning and verification are repeated three times and the result is still unqualified, issue a cloud platform warning to remind manual maintenance.

[0030] As described above, the cleaning system for the five-parameter environmental monitoring sensors provided by this invention includes a sensor data monitoring unit, an intelligent decision-making unit, an adaptive cleaning execution unit, and a closed-loop verification unit. The intelligent decision-making unit is communicatively connected to the sensor data monitoring unit, the adaptive cleaning execution unit, and the closed-loop verification unit. The sensor data monitoring unit is communicatively connected to the sensors and is used to collect water quality data in real time. The intelligent decision-making unit is used to obtain corresponding coefficients based on the water quality data, and, in conjunction with the weight parameters corresponding to the water quality data, as well as historical trend correction parameters and their weight parameters, to determine whether each sensor needs cleaning and the cleaning method for the sensors that need cleaning. The adaptive cleaning execution unit is used to invoke the corresponding cleaning process according to the conclusion and the cleaning method, cleaning only the sensors that need cleaning. The closed-loop verification unit is used to verify the cleaning effect of the sensors after cleaning. If the verification fails, it feeds back to the intelligent decision-making unit to upgrade the cleaning level and continue cleaning. Therefore, it can make cleaning decisions based on adaptive algorithms of real-time and historical contamination, improve the accuracy of cleaning triggering, avoid the problems of wasted cleaning resources and cleaning lag, and can also use a method that matches the contamination characteristics of each sensor for more thorough cleaning. Furthermore, it verifies the cleaning effect to ensure that the cleaning meets the standards and avoids over-cleaning other sensors and damaging them. The cleaning method for the five-parameter environmental monitoring sensors provided by the present invention has the same advantages as the cleaning system for the five-parameter environmental monitoring sensors provided by the present invention. Attached Figure Description

[0031] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0032] Figure 1 A schematic diagram of an embodiment of a cleaning system for a five-parameter environmental monitoring sensor provided by the present invention;

[0033] Figure 2 A schematic diagram of a specific embodiment of a cleaning system for a five-parameter environmental monitoring sensor;

[0034] Figure 3 A schematic diagram of the architecture of a specific embodiment of a cleaning system for a five-parameter environmental monitoring sensor;

[0035] Figure 4 A front view of the adaptive cleaning execution unit;

[0036] Figure 5 This is a schematic diagram of the cleaning components;

[0037] Figure 6 This is a schematic diagram of the rear of the adaptive cleaning execution unit;

[0038] Figure 7 This is a schematic diagram of an embodiment of a cleaning method for a five-parameter environmental monitoring sensor provided by the present invention;

[0039] Figure 8 A flowchart for verifying the cleaning effect of the sensor after cleaning;

[0040] Figure 9 A schematic diagram of the process for ensuring cleanup through cyclical retry and manual intervention;

[0041] Figure 10 This is a schematic diagram of the overall process for cleaning a five-parameter sensor for environmental monitoring.

[0042] The annotations in the attached figures are explained as follows:

[0043] Sensor data monitoring unit 1, intelligent decision-making unit 2, adaptive cleaning execution unit 3, closed-loop verification unit 4, remote linkage unit 5, sensor identification component 31, four sets of cleaning components 32, media supply component 33, cleaning container 34, switch component 331, measuring cup 6, measuring cup cleaning brush 7, electrode 8, soft brush 321, nozzle 322, ultrasonic transducer 323, peristaltic pump 332, high-pressure cleaning component 333, control board 35, verification container 9. Detailed Implementation

[0044] The core of this invention is to provide a cleaning system and method for five-parameter environmental monitoring sensors. It can make cleaning decisions based on an adaptive algorithm of real-time and historical pollution, improve the accuracy of cleaning triggering, avoid waste of cleaning resources and cleaning delays, and perform more thorough cleaning on each sensor in a way that matches its pollution characteristics. Furthermore, it verifies the cleaning effect to ensure that the cleaning meets the standards and avoids over-cleaning other sensors, which could damage them.

[0045] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0046] An example implementation of the cleaning system for a five-parameter environmental monitoring sensor provided by this invention. Figure 1 As shown, Figure 1 This is a schematic diagram of an embodiment of a cleaning system for a five-parameter environmental monitoring sensor provided by the present invention. The cleaning system for the five-parameter environmental monitoring sensor may include a sensor data monitoring unit 1, an intelligent decision-making unit 2, an adaptive cleaning execution unit 3, and a closed-loop verification unit 4.

[0047] The intelligent decision-making unit 2 is communicatively connected to the sensor data monitoring unit 1, the adaptive cleaning execution unit 3, and the closed-loop verification unit 4.

[0048] Sensor data monitoring unit 1 is connected to the sensor for real-time water quality data acquisition;

[0049] The intelligent decision unit 2 is used to obtain the corresponding coefficients based on the water quality data, and combine the weight parameters corresponding to the water quality data, as well as the historical trend correction parameters and their weight parameters, to obtain the conclusion of whether each sensor needs to be cleaned and the cleaning method for the sensors that need to be cleaned.

[0050] The adaptive cleaning execution unit 3 is used to call the corresponding cleaning process based on the conclusion and cleaning method, and only clean the sensors that need to be cleaned.

[0051] The closed-loop verification unit 4 is used to verify the cleaning effect of the sensor after cleaning. If the verification fails, it will be fed back to the intelligent decision unit to upgrade the cleaning level and continue cleaning.

[0052] It should be noted that the conventional five parameters refer to temperature, pH, dissolved oxygen, conductivity, and turbidity. The measured values ​​of pH, conductivity, and dissolved oxygen are all affected by temperature; therefore, a temperature sensor is built into the pH, dissolved oxygen, and conductivity sensors. Data is collected in real time by the temperature sensor, and dynamic compensation is performed to ensure monitoring accuracy. Therefore, when measuring pH in water, the water temperature is also measured, so a separate temperature sensor is not required. The cleaning system for this environmental monitoring five-parameter sensor can make decisions and execute cleaning based on an adaptive algorithm. Specifically, this adaptive algorithm is a core algorithm that correlates water body type, sensor type, real-time pollution status (turbidity + parameter-specific pollution), and historical pollution data. It includes dual logic of "periodic cleaning trigger" and "additional cleaning decision," which differs from the "single trigger mode" of existing technologies. It can achieve precise cleaning control of "routine cleaning + water body-specific cycle + dynamic replenishment." The water body-specific cycle is based on different water bodies such as drinking water sources, groundwater, surface water, and pollution sources. To mitigate pollution risks, based on a pre-set basic cleaning cycle (e.g., 24 hours for drinking water sources, 8 hours for pollution sources), specialized cleaning agents (such as buffer solutions, citric acid, etc.) need to be added. Cleaning effectiveness verification is performed after each cleaning cycle to address the issue that existing technologies with a "uniform cycle" cannot adapt to different water pollution rates. The conventional cleaning mode refers to routine cleaning performed after each water sample measurement process, relying solely on clean water and physical cleaning (ultrasonic vibration, soft brush wiping), without the involvement of chemical reagents, thus avoiding sensor corrosion and secondary water pollution. There is no effectiveness verification. Additional cleaning decisions are made outside the basic cycle, using a multi-weighted fusion formula to determine whether additional cleaning is needed. The core logic is "turbidity-driven + real-time pollution assistance + historical pollution correction." Supplementary cleaning is performed when pollution exceeds the standard before the cycle expires, requiring the addition of specialized cleaning agents (such as buffer solutions, citric acid, etc.), and effectiveness verification is performed afterward. This adapts to sudden or severe pollution. It should also be noted that because turbidity directly reflects the suspended solids content in the water (a core indicator of pollution risk), a separate pollution quantification coefficient is set for it. During cleaning, modular cleaning components can be used. Specifically, dedicated cleaning units are designed for the contamination characteristics of sensors with different parameters. These units include ultrasonic transducers, spray modules, wiping units, and drying modules. They can be automatically matched through ID recognition. The closed-loop verification unit 4 mentioned above can verify the cleaning effect through calibration tests with a reference liquid after cleaning. The deviation rate is used to determine whether secondary cleaning is needed to ensure that the cleaning meets the standards. All units work together to form a complete process of "identification-decision-cleaning-verification-optimization". This enables unattended automatic maintenance of the five-parameter sensors, improves the accuracy and reliability of monitoring data, and reduces maintenance costs and workload.

[0053] As described above, the cleaning system for the five-parameter environmental monitoring sensors provided by this invention includes a sensor data monitoring unit, an intelligent decision-making unit, an adaptive cleaning execution unit, and a closed-loop verification unit. The intelligent decision-making unit is communicatively connected to the sensor data monitoring unit, the adaptive cleaning execution unit, and the closed-loop verification unit. The sensor data monitoring unit is communicatively connected to the sensors and is used to collect water quality data in real time. The intelligent decision-making unit is used to obtain corresponding coefficients based on the water quality data, and, in conjunction with the weight parameters corresponding to the water quality data, as well as historical trend correction parameters and their weight parameters, to determine whether each sensor needs cleaning and the cleaning method for the sensors that need cleaning. The adaptive... The cleaning execution unit is used to call the corresponding cleaning process based on the conclusion and cleaning method, and only cleans the sensors that need to be cleaned. The closed-loop verification unit is used to verify the cleaning effect of the sensors after cleaning. When the verification fails, it feeds back to the intelligent decision unit to upgrade the cleaning level and continue cleaning. Therefore, it can make cleaning decisions based on adaptive algorithms of real-time contamination and historical contamination, improve the accuracy of cleaning triggering, avoid the problems of cleaning resource waste and cleaning lag, and can also use a method that matches the contamination characteristics of each sensor for more thorough cleaning, and verify the cleaning effect to ensure that the cleaning meets the standards, avoiding over-cleaning of other sensors and damaging the sensors.

[0054] refer to Figure 2 , Figure 2 This is a schematic diagram of a specific embodiment of a cleaning system for a five-parameter environmental monitoring sensor. In this embodiment, a remote linkage unit 5, communicatively connected to the intelligent decision-making unit 2, may also be included. This remote linkage unit 5 receives the cleaning process data when the closed-loop verification unit 4 passes verification, performs data aggregation and analysis, and transmits it to the intelligent decision-making unit 2 for strategy optimization based on historical data. This remote linkage unit 5 supports cloud collaboration and is adaptable to flexible adjustments in various water body scenarios.

[0055] For details, please refer to Figure 3 , Figure 3 This is a schematic diagram of the architecture of a specific embodiment of a cleaning system for a five-parameter environmental monitoring sensor. The system is data-driven, algorithm-enabled, and continuously self-optimizes through closed-loop verification. Starting with environmental perception, it makes cleaning decisions through intelligent analysis, verifies the effects immediately after precise execution, and uses the results for remote optimization and strategy adjustment, forming a complete, efficient, reliable, and sustainably evolving working closed loop, as detailed below:

[0056] (1) The core driver is the adaptive algorithm: The top layer of the system is the adaptive algorithm engine, which is responsible for processing key parameters (such as water type, weight parameters, and specific coefficients) and driving the entire system to make intelligent judgments, execute accurately and continuously optimize. It is the foundation for realizing the "intelligent" and "adaptive" capabilities.

[0057] (2) The sensor input layer is used for data acquisition: This is the system's "sense". The sensor array continuously collects real-time monitoring data such as water quality, providing accurate and real-time data input for all subsequent decisions. It is the starting point of the entire intelligent process.

[0058] (3) The intelligent decision-making layer is used for analysis and judgment, and the intelligent control unit serves as the decision-making center, where it analyzes the perceived data and runs three types of parallel judgment logic:

[0059] Routine cleaning after measurement: Triggered during daily monitoring intervals, basic cleaning is performed.

[0060] Periodic cleaning judgment: When the monitoring time reaches the "dedicated period" set for a specific water body, it is triggered to perform planned deep maintenance.

[0061] Additional cleaning judgment: By calculating multi-parameter indicators, once the preset threshold is exceeded, emergency cleaning is immediately triggered to deal with sudden pollution.

[0062] (4) Execution layer: Used for actions and verification. Decision instructions are issued to this layer, which is divided into two key action units:

[0063] Adaptive cleaning unit: Responsible for accurately executing cleaning actions, and can dynamically adjust the cleaning method and intensity according to instructions.

[0064] The closed-loop verification unit is responsible for performance evaluation. It performs quantitative verification of the cleaning effect immediately after cleaning (such as calculating the deviation rate), forming a key quality closed loop.

[0065] (5) Verify and optimize the closed loop, used to verify the results and form decision branches:

[0066] If the verification is successful, the process enters the remote linkage unit to collect and analyze data, and optimize the cleaning strategy based on historical data to achieve system self-evolution.

[0067] If the verification fails, a "feedback adjustment instruction" is generated and returned to the decision center, which may trigger an "upgrade process" (such as secondary cleaning or alarm) to ensure that the problem is resolved.

[0068] Here, "verification" refers to the baseline solution test after additional cleaning. Specifically, after cleaning, the sensor is immersed in the baseline solution to collect data, determining whether the sensor has recovered to a state where "monitoring accuracy meets the standards." Therefore, "verification" confirms whether "this cleaning achieved the expected cleaning effect." When the verification is "yes" (the effect is qualified), it means that this cleaning was successful. However, the purpose of the remote linkage unit's "strategy optimization" is to optimize subsequent cleaning strategies based on "complete data from this cleaning" (such as the current water type, pollution level, cleaning parameters used, and the effect). This includes "accumulating effective data to make future cleaning more accurate and reduce ineffective operations," and conducting "long-term strategy iteration." When the verification result is "No" (the effect is not up to standard), it means that the cleaning was not clean. "Initiating upgrade processing" here means triggering a "more powerful cleaning plan" (such as "secondary additional cleaning" in the technical plan: increasing reagent concentration, increasing ultrasonic intensity, and extending cleaning time), which is a "remedial operation for this failure to meet the standard". If it is still not up to standard after multiple "upgrade processing", then "operation and maintenance warning (manual intervention improvement)" will be triggered. It does not directly "upgrade the cleaning method itself", but "upgrade the execution intensity of this cleaning".

[0069] refer to Figure 4 , Figure 5 and Figure 6 , Figure 4 This is a front view of the adaptive cleaning execution unit. Figure 5 This is a schematic diagram of the cleaning components. Figure 6 This is a rear view of the adaptive cleaning execution unit. In another specific embodiment of the cleaning system for the aforementioned five-parameter environmental monitoring sensors, the adaptive cleaning execution unit 3 may include a sensor identification component 31, four sets of cleaning components 32, and a media supply component 33. The four sets of cleaning components 32 are located within the cleaning container 34 and correspond to the pH sensor, conductivity sensor, dissolved oxygen sensor, and turbidity sensor, respectively. The media supply component 33 is connected to a pure water container, a cleaning agent container, and a reference liquid container via a switching component 331, which can be a multi-channel switching valve. The cleaning process includes the cleaning components, media, and cleaning procedures used for each sensor. Furthermore... Figure 4 The measuring cup 6, measuring cup cleaning brush 7, and electrode 8 were also displayed.

[0070] In another specific embodiment of the cleaning system for the five-parameter environmental monitoring sensor described above, the sensor identification component is a sensor ID chip reader, which is used to read the chip ID on the sensor surface to identify the sensor type and to call the cleaning components and cleaning process corresponding to the sensor type.

[0071] Continue to refer to Figure 5 In a preferred embodiment of the cleaning system for the five-parameter environmental monitoring sensor described above, the cleaning component 32 may include a soft brush 321, a nozzle 322, and an ultrasonic transducer 323 disposed within a cleaning container. The nozzle 322 is connected to the media supply component 33, which may include a peristaltic pump 332. The peristaltic pump 332 supplies the media required for the cleaning process to the corresponding cleaning component 32 and sprays it onto the sensor from the corresponding nozzle 322. Specifically, the media supply component may have a built-in independent sealed chamber for storing pure water, neutral buffer solution, and 0.1 mol / L citric acid solution, respectively. The supply volume is precisely controlled by a micro peristaltic pump (the volume used for a single cleaning is ≤50 mL). Additionally, it may also include a high-pressure cleaning component 333 and a control board 35.

[0072] Specifically, the cleaning mode process for each sensor is shown in Table 1. Table 1 is a cleaning mode process table for each sensor of the cleaning component.

[0073] Table 1. Cleaning mode flowchart for each sensor of the cleaning component.

[0074]

[0075] Continue to refer to Figure 4 In another preferred embodiment of the cleaning system for the five-parameter environmental monitoring sensor described above, the closed-loop verification unit may include a deviation calculation module and a verification container 9 connected to the reference liquid container.

[0076] The verification container 9 can be filled with reference liquid through the reference liquid container and the sensor immersed in it can be drained and then re-entered into the sensor to obtain the measured value.

[0077] The deviation calculation module is used to calculate the deviation rate between the measured value and the benchmark value, and to determine whether the cleaning has been completed based on the deviation rate.

[0078] In another preferred embodiment of the cleaning system for the aforementioned five-parameter environmental monitoring sensors, the remote linkage unit 5 may specifically include an edge computing module, a cloud platform communication module, and a remote control interface. Specifically, the functions of the remote linkage unit 5 include:

[0079] Data Upload: Upload cleaning status (executed / not executed), cleaning parameters (K1 / K2 / K3), verification deviation rate, hardware status and other data to the cloud platform in real time;

[0080] Remote control: Supports remote triggering of cleaning and adjustment of algorithm parameters (such as W1 / W2W3 / T) via cloud platform. x ), Update the reference solution standard value;

[0081] Collaborative optimization: Adopting a collaborative architecture of "edge computing + cloud analysis", edge devices process 80% of real-time data (such as K calculation and cleaning decision), while the cloud iteratively optimizes algorithm parameters based on multi-region water quality data and synchronizes them to the edge devices.

[0082] An example implementation of the cleaning method for a five-parameter environmental monitoring sensor provided by this invention. Figure 7 As shown, Figure 7 This is a schematic diagram of an embodiment of a cleaning method for a five-parameter environmental monitoring sensor provided by the present invention. The method utilizes a cleaning system for a five-parameter environmental monitoring sensor as described above and may include the following steps:

[0083] S1: Real-time water quality data collection;

[0084] S2: Based on the water quality data, obtain the corresponding coefficients, and combine them with the weight parameters corresponding to the water quality data, as well as the historical trend correction parameters and their weight parameters, to obtain the conclusion on whether each sensor needs to be cleaned and the cleaning method for the sensors that need to be cleaned.

[0085] S3: Based on the conclusions and cleaning methods, call the corresponding cleaning process and clean only the sensors that need cleaning;

[0086] S4: Verify the cleaning effect of the sensor after cleaning. If the verification fails, feedback is given to the intelligent decision-making unit to upgrade the cleaning level and continue cleaning.

[0087] In a specific embodiment of the cleaning method for the five environmental monitoring sensors described above, the corresponding coefficients are obtained based on the water quality data. Combined with the weight parameters corresponding to the water quality data, as well as the historical trend correction parameters and their weight parameters, the conclusion regarding whether each sensor needs cleaning and the cleaning method for the sensors that require cleaning can include the following steps:

[0088] According to the additional cleaning execution decision formula, when W1*K1+W2*K2+W3*K3≥Decision threshold T x When the time is right, the corresponding sensor is cleaned. Here, W1 is the turbidity weight, W2 is the real-time contamination weight, W3 is the historical contamination weight, x=1 corresponds to the pH cleaning decision threshold, x=2 corresponds to the conductivity cleaning decision threshold, x=3 corresponds to the dissolved oxygen cleaning decision threshold, x=4 corresponds to the turbidity cleaning decision threshold, K1 is the turbidity coefficient, K2 is the real-time contamination coefficient specific to each parameter, and K3 is the historical contamination coefficient.

[0089] It should be noted that W1 is a core indicator because turbidity directly reflects the content of suspended solids and has the highest pollution risk. It can be set to 0.6, or adjusted as needed. W2 is an auxiliary parameter used to supplement parameters not covered by turbidity, including sudden pH changes, increased conductivity, and decreased dissolved oxygen. It can be set to 0.3, or adjusted as needed. W3 is used for trend correction to avoid ineffective cleaning caused by instantaneous fluctuations. It can be set to 0.1, or adjusted as needed. When x=1, T1 can be set to 0.8; when x=2, T2... The value of T3 can be 0.6. When x=3, T3 can be 0.6. When x=4, T4 can be 0.8. The value of K1 is 1.0 when the turbidity is not less than 200 NTU, 0.8 when the turbidity is between 100 NTU and 200 NTU, and 0.5 when the turbidity is less than 100 NTU. The value of K3 can be the average of the last 3 times (K1*0.6+K2*0.3). When this average is greater than 0.6, the value of K3 is 1.0. Otherwise, the value is 0.2. The values ​​of K2 are shown in Table 2. Table 2 is a table of real-time pollution coefficient values ​​for each parameter.

[0090] Table 2. Values ​​of Real-Time Pollution Coefficient for Each Parameter

[0091]

[0092] In a specific example, a surface water scenario (cycle 4h, already running for 2h) → sudden turbidity of 220 NTU (K1=1.0), pH=3.8 (pH<4 or pH>11, K2=1.2), average of the last 3 times = 0.5 (K3=0.2) → calculation: 0.6×1.0+0.3×1.2+0.1×0.2=0.98≥0.8 → triggers additional cleaning.

[0093] In another specific embodiment of the cleaning method for the five-parameter environmental monitoring sensors described above, verifying the cleaning effect of the sensors after cleaning, and feeding back to the intelligent decision-making unit to upgrade the cleaning level and continue cleaning when the verification fails, may include the following steps:

[0094] The verification container is filled with reference liquid through the reference liquid container, and the sensor is immersed in it. After the reference liquid is drained, the reference liquid is re-entered into the reference liquid container, immersed in the sensor and stabilized for a period of time to obtain the measured value.

[0095] Calculate the deviation rate between the measured value and the benchmark value. Determine whether the cleaning has been completed based on the deviation rate. If the cleaning is complete, resume monitoring. Otherwise, perform a second cleaning and re-verify. If the cleaning and verification are repeated three times and the result is still unqualified, issue a cloud platform warning to remind manual maintenance.

[0096] It should be noted that this embodiment ensures that both "periodic cleaning" and "additional cleaning" achieve the required cleaning effect, avoiding insufficient cleaning. The micro-reference solution reservoir used can be a standard solution with pH=6.86, conductivity=1413μS / cm, and dissolved oxygen zero, etc., for reference. Figure 8 , Figure 8 The flowchart for verifying the cleaning effect of the sensor after cleaning follows the logic of "preparation-rinsing-measurement-calculation". Through standardized operation steps and objective mathematical calculations, the subjective "whether it is clean or not" is transformed into the quantifiable "whether the deviation rate meets the standard". It is the core link connecting "cleaning execution" and "quality judgment" and ensures the data quality of the entire monitoring system. It can be summarized into the following key stages: (1) Process start-up: The verification process can be triggered by two situations: the completion of daily cleaning will not trigger the verification process and will end the process directly. The completion of periodic cleaning and additional cleaning can trigger the verification process. (2) Preparation and rinsing: First, the media supply system releases the reference liquid of the corresponding parameters into the verification tank to provide a standard environment for verification. Then, the sensor is immersed and rinsed for 60 seconds. This step is crucial. The purpose is to remove any contaminants or liquid from the previous test that may remain on the sensor surface to ensure the accuracy of subsequent measurements. After rinsing, the verification tank is emptied. (3) Formal Measurement: The system releases the reference liquid again, and then the sensor is immersed for a second time and stabilized for 120 seconds. This stabilization time is longer to allow the sensor to fully balance with the standard environment before collecting the measured value. This step (secondary liquid addition and longer stabilization time) is designed to obtain a stable and reliable measurement reading. (4) Effect Evaluation: At the end of the process, the deviation rate is calculated according to the formula: Deviation Rate = (Measured Value - Reference Value) / Reference Value × 100%. The deviation rate is a key quantitative indicator that will be directly used to determine whether the cleaning is qualified, thereby determining whether the sensor can resume monitoring or whether subsequent operations such as secondary cleaning need to be initiated.

[0097] Further reference Figure 9 , Figure 9The flowchart of the cleaning cycle retry and manual intervention guarantee mechanism is shown. When judging, if the deviation rate meets the technical requirements, monitoring is restored; if the deviation rate does not meet the technical requirements, a second cleaning is performed and the cleaning effect is verified again. The process is repeated up to 3 times. If it still does not meet the standard, an early warning is issued on the cloud platform to remind manual maintenance. The core of the control logic of automated cleaning effect verification and upgrade processing is to determine the cleaning quality by quantifying the "deviation rate" and design a cycle retry and manual intervention guarantee mechanism. The whole process can be summarized into the following key stages: (1) Verification and judgment. The process starts with "cleaning effect verification" and obtains an objective quality indicator through "deviation rate calculation". The system then compares the deviation rate with the preset "technical requirements" to form a key decision point. If the requirements are met, the process directly moves to "cleaning qualified" and then "restores monitoring". The work is completed. If the requirements are not met, the process enters a "processing cycle" that includes a retry mechanism. (2) Loop retry mechanism: When cleaning fails, the system does not immediately alarm, but starts an automatic retry process with a limited number of attempts: First, "record cleaning count: +1". Then, determine whether the cumulative cleaning count is "≤3 times". If it does not exceed 3 times: the system automatically "executes cleaning" and returns to the starting point of the process to re-verify the effect and calculate the deviation rate, forming a closed loop of "cleaning → verification → judgment" until it is qualified or the retry limit is reached. If it has reached 3 times: it means that automatic cleaning can no longer solve the problem, and the process is upgraded. (3) Upgrade processing: When automatic loop cleaning still fails after 3 times, the system will jump out of the loop, trigger "cloud platform alarm" and "notify manual maintenance". This marks the switch of the process from automatic processing mode to manual intervention mode, where maintenance personnel will investigate and solve the fundamental problem. The technical requirements for deviation rate can be as follows: pH value is ±0.1, dissolved oxygen is ±0.3mg / L, conductivity is ±5us / cm for reference solution concentration ≤100us / cm and ±5% for reference solution concentration >100us / cm, and turbidity is ±15% for reference solution concentration ≤100NTU and ±10% for reference solution concentration >100NTU.

[0098] The following example illustrates the overall workflow of the above method. (Refer to...) Figure 10 , Figure 10 This is a schematic diagram of the overall process for cleaning five-parameter sensors for environmental monitoring. The intelligent monitoring and adaptive cleaning control system forms a complete "monitoring-judgment-cleaning-verification" closed loop. Its core feature is the combination of fixed-cycle maintenance and dynamic condition triggering mechanisms. The dual-trigger mechanism balances the planning and flexibility of maintenance, and the mandatory closed-loop verification ensures the effectiveness of any cleaning action, ultimately guaranteeing the accuracy and reliability of the entire monitoring system data.

[0099] The following are the key steps in the process:

[0100] (1) Initialization and parameter preset:

[0101] The process begins with "reading preset water body type parameters". This step means that the system can load different cleaning strategies and threshold parameters according to different application environments (such as different water quality and operating conditions), providing a basis for subsequent intelligent judgment.

[0102] (2) Dual-channel triggering mechanism:

[0103] After "starting continuous monitoring", the system runs two sets of judgment logic in parallel to ensure no omissions:

[0104] Path A: Event-driven routine cleaning.

[0105] After each measurement operation, perform a routine post-measurement cleaning. This is a lightweight, high-frequency preventative maintenance designed to remove surface deposits promptly and prevent contaminant buildup.

[0106] Path B: Intelligent triggering based on time and conditions.

[0107] Periodic Trigger: The system is equipped with a "measurement cycle timer". When the "continuous monitoring duration" reaches the "preset cycle", cleaning will be triggered regardless of the current data. This is a planned, in-depth maintenance backup.

[0108] Conditional Trigger: When the system calculates in real time that the multivariate parameter index T1≥0.8, or T2≥0.8, or T3≥0.8, or T4≥0.8, it immediately triggers the cleaning execution unit of the respective sensor to perform cleaning. This can sensitively respond to sudden contamination or performance drift, reflecting the system's "intelligent judgment" capability, and is a proactive and adaptive maintenance strategy.

[0109] (3) Cleaning execution and strategy selection.

[0110] When any of the above conditions are met, the process converges to the "trigger cleaning execution unit". Based on the "circular cleaning strategy" and the "overflow cleaning strategy", the system can adaptively select different physical or chemical cleaning methods according to the degree or type of contamination to optimize resource use and cleaning effect.

[0111] (4) Mandatory verification and feedback loop.

[0112] After cleaning, a "closed-loop verification process" must be followed. This is crucial for quality control. The system will verify whether the standards are met:

[0113] If the target is met: "Resume Monitoring" and "Start Measurement Cycle Timer" to begin a new monitoring cycle. If the target is not met: Proceed to the "Execute Follow-up Processing" branch, which includes activating backup sensors, repeating cleaning, adjusting parameters, or issuing a manual maintenance alert to ensure the problem is escalated.

[0114] In summary, the above-mentioned system and method provided by the present invention have the following advantages: (1) They are adaptable to monitoring multiple water body scenarios, have strong versatility, and reduce resource waste; (2) They respond promptly to additional cleaning in case of sudden pollution, avoid pollution accumulation, and improve data efficiency; (3) They are highly targeted and improve monitoring accuracy: Modular cleaning components are matched with parameter-specific modes to effectively remove pollutants specific to each sensor (such as pH probe crystals, dissolved oxygen fluorescent membrane biofilms), and combined with closed-loop verification, the accuracy of monitoring data is improved by 20%, avoiding measurement deviations caused by existing general cleaning; (4) They make accurate decisions and reduce invalid operations: Adaptive algorithms combined with dual-dimensional pollution coefficients and multi-weight formulas combined with parameter-specific standards avoid invalid cleaning caused by instantaneous fluctuations; (5) They achieve unattended maintenance, reduce probe damage rate, reduce on-site inspection frequency, and reduce operation and maintenance costs; and extend sensor lifespan by 30%: Differentiated cleaning components (such as dissolved oxygen soft brushes and pH low-intensity ultrasound) avoid damage, extending the lifespan from 1 year to 1.3 years.

[0115] The core advantage of the above solution is the targeted cleaning of "parameter-specific contamination quantification + modular components", which solves the core problem of incomplete cleaning or sensor damage caused by the generalized cleaning method of the existing technology. The core defect of the existing technology is that "regardless of pH crystallization, dissolved oxygen fluorescent membrane contamination, or conductivity scale, the same cleaning method (such as high-pressure water flushing) is used": either it is not cleaned well, resulting in monitoring deviation; or it damages the sensor and shortens its lifespan. This solution upgrades "general cleaning" to "parameter-customized cleaning," solving both the precision problem of "incomplete cleaning of specific contaminants" and the lifespan problem of "damage to sensitive components." Its second advantage is the "decision logic based on multi-parameter weight fusion," which addresses the issue of "misjudgment by a single parameter in existing sudden contamination cleaning methods." This solution improves upon this by using "turbidity-driven (W1=0.6) + other parameter-assisted (W2=0.3) + historical trend correction (W3=0.1)." This is not "triggered by a single parameter" but rather "cross-validated by multiple parameters," reducing the misjudgment rate. This avoids both "ineffective cleaning due to occasional fluctuations" and "contamination accumulation due to missed judgments by a single parameter." Another advantage is its "routine + periodic + extra" three-logic + closed-loop verification, which solves the problems of fragmented logic and lack of effectiveness guarantee in existing cleaning methods. Existing cleaning methods are "isolated logic" (either routine cleaning, periodic cleaning, or emergency contamination cleaning, which are unrelated to each other) and lack cleaning effect verification. If the cleaning is not clean after routine cleaning, it can only be corrected during periodic cleaning, and the data deviation may last for several hours. Moreover, the effect of emergency cleaning may be wasted because the effect is unknown. The improvement of this solution is "logic layering + effect closed loop": routine cleaning (light cleaning after measurement) performs "immediate maintenance", periodic cleaning performs "minimum maintenance", and extra cleaning performs "contamination repair". The three-layer logic can cover all scenarios; after extra cleaning, "baseline liquid verification" is used to ensure the effect and avoid the ineffective operation of "cleaning but not cleaning". This solution also offers the following advantages: "Extended sensor lifespan + reduced maintenance costs," addressing the high loss and maintenance issues of existing cleaning methods. Differentiated cleaning components prevent damage to dissolved oxygen fluorescent membranes and pH glass electrodes, extending their lifespan by 30%. Precise decision-making reduces the number of ineffective cleanings, and unattended operation reduces inspection frequency, thus lowering maintenance costs. This solution also achieves "multi-water body cycle adaptation," resolving the waste and lag issues caused by uniform cycles in existing cycle cleaning methods. This is an optimization of "conventional cycle cleaning," extending the drinking water source cycle from 24 hours to 144 hours, thereby reducing over-cleaning, and shortening the contamination source cycle to within 24 hours, avoiding cleaning lag. However, without "parameter-customized cleaning," even with cycle adaptation, the problem of "incomplete cleaning / damage" cannot be solved. Therefore, the greatest advantage of this solution is its ability to accurately identify the contamination level of parameters and use a cleaning method adapted to those parameters, further ensuring the correct timing of cleaning through multi-parameter logic. This is an improvement over the "generalization and uniformity" issues of existing cleaning methods.

[0116] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A cleaning system for a five-parameter environmental monitoring sensor, characterized in that, It includes a sensor data monitoring unit, an intelligent decision-making unit, an adaptive cleaning execution unit, and a closed-loop verification unit; The intelligent decision-making units are all communicatively connected to the sensor data monitoring unit, the adaptive cleaning execution unit, and the closed-loop verification unit. The sensor data monitoring unit is connected to the sensor for real-time water quality data collection; The intelligent decision-making unit is used to obtain the corresponding coefficients based on the water quality data, and combine the weight parameters corresponding to the water quality data, as well as the historical trend correction parameters and their weight parameters, to obtain a conclusion on whether each sensor needs to be cleaned and the cleaning method for the sensors that need to be cleaned. The adaptive cleaning execution unit is used to call the corresponding cleaning process according to the conclusion and the cleaning method, and only clean the sensors that need to be cleaned. The closed-loop verification unit is used to verify the cleaning effect of the sensor after cleaning. If the verification fails, it feeds back to the intelligent decision-making unit to upgrade the cleaning level and continue cleaning.

2. The cleaning system for the five-parameter environmental monitoring sensor according to claim 1, characterized in that, It also includes a remote linkage unit that is communicatively connected to the intelligent decision-making unit, which is used to receive the cleaning process when the closed-loop verification unit passes the verification, to perform data aggregation and analysis, and to transmit it to the intelligent decision-making unit for strategy optimization based on historical data.

3. The cleaning system for the five-parameter environmental monitoring sensor according to claim 2, characterized in that, The adaptive cleaning execution unit includes a sensor identification component, four sets of cleaning components, and a media supply component. The four sets of cleaning components are located inside the cleaning container and correspond to the pH sensor, conductivity sensor, dissolved oxygen sensor, and turbidity sensor, respectively. The media supply component is connected to the pure water container, cleaning agent container, and reference liquid container through a switch component. The cleaning process includes the cleaning components, media, and cleaning procedures used for each type of sensor.

4. The cleaning system for the five-parameter environmental monitoring sensor according to claim 3, characterized in that, The sensor identification component is a sensor ID chip reader, which is used to read the chip ID on the sensor surface to identify the sensor type, and to call the cleaning component and cleaning process corresponding to the sensor type.

5. The cleaning system for the five-parameter environmental monitoring sensor according to claim 4, characterized in that, The cleaning assembly includes a soft brush, a nozzle, and an ultrasonic transducer disposed within the cleaning container. The nozzle is connected to the media supply assembly, which includes a peristaltic pump for supplying the media required for the cleaning process to the corresponding cleaning assembly and spraying it from the corresponding nozzle onto the sensor.

6. The cleaning system for the five-parameter environmental monitoring sensor according to claim 5, characterized in that, The closed-loop verification unit includes a deviation calculation module and a verification container connected to the reference liquid container; The verification container can be filled with reference liquid through the reference liquid container, and the sensor immersed in it can be immersed in the reference liquid. After the reference liquid is drained, it can be re-entered into the sensor to obtain the measured value. The deviation calculation module is used to calculate the deviation rate between the measured value and the benchmark value, and to determine whether the cleaning has been completed based on the deviation rate.

7. The cleaning system for the five-parameter environmental monitoring sensor according to claim 6, characterized in that, The remote linkage unit includes an edge computing module, a cloud platform communication module, and a remote control interface.

8. A cleaning method for a five-parameter environmental monitoring sensor, characterized in that, The cleaning system using the five-parameter environmental monitoring sensor as described in any one of claims 1-7 includes: Real-time water quality data collection; Based on the water quality data, the corresponding coefficients are obtained, and combined with the weight parameters corresponding to the water quality data, as well as the historical trend correction parameters and their weight parameters, a conclusion is drawn as to whether each sensor needs to be cleaned and the cleaning method for the sensors that need to be cleaned. Based on the conclusions and the cleaning method, the corresponding cleaning process is invoked, and cleaning is performed only on the sensors that need to be cleaned. The cleaning effect of the sensor after cleaning is verified. If the verification fails, the intelligent decision-making unit is notified to upgrade the cleaning level and continue cleaning.

9. The cleaning method for the five-parameter environmental monitoring sensor according to claim 8, characterized in that, The process of obtaining corresponding coefficients based on the water quality data, and combining them with the weight parameters corresponding to the water quality data, as well as the historical trend correction parameters and their weight parameters, to determine whether each sensor needs cleaning and the cleaning methods for sensors that need cleaning includes: According to the additional cleaning execution decision formula, when W1*K1+W2*K2+W3*K3≥Decision threshold T x When the time is right, the corresponding sensor is cleaned. Here, W1 is the turbidity weight, W2 is the real-time contamination weight, W3 is the historical contamination weight, x=1 corresponds to the pH cleaning decision threshold, x=2 corresponds to the conductivity cleaning decision threshold, x=3 corresponds to the dissolved oxygen cleaning decision threshold, x=4 corresponds to the turbidity cleaning decision threshold, K1 is the turbidity coefficient, K2 is the real-time contamination coefficient specific to each parameter, and K3 is the historical contamination coefficient.

10. The cleaning method for the five-parameter environmental monitoring sensor according to claim 9, characterized in that, The process of verifying the cleaning effect of the sensor after cleaning, and feeding back to the intelligent decision-making unit when the verification fails to upgrade the cleaning level and continue cleaning, includes: The verification container is filled with reference liquid through the reference liquid container, and the sensor is immersed in it. After the reference liquid is drained, the reference liquid is re-entered into the reference liquid container, immersed in the sensor and stabilized for a period of time to obtain the measured value. Calculate the deviation rate between the measured value and the benchmark value, and determine whether the cleaning has been completed based on the deviation rate. If the cleaning is complete, resume monitoring; otherwise, perform a second cleaning and re-verify. If the cleaning and verification are repeated three times and the result is still unqualified, issue a cloud platform warning to remind manual maintenance.

Images