Water quality detection equipment and use method

Abstract

The invention relates to the technical field of water body environment detection, in particular to water quality detection equipment and a using method, and the water quality detection equipment comprises a water body sampling module which is provided with a water quality sampling pipeline, is configured with a self-adaptive sampling frequency adjusting strategy and is used for adjusting the collection frequency of a water sample to be detected conveyed to the water quality detection module; the water sample storage module is provided with a water quality detection cavity, a detection probe and a detection cavity anti-clogging module and is used for carrying out anti-clogging treatment on a water sample flowing path in the detection cavity; the probe self-cleaning module is correspondingly connected with the detection probe, is provided with a cleaning triggering strategy and is used for automatically cleaning the surface of the detection probe; the water quality detection module is connected with the detection probe and is used for receiving a detection signal of the detection probe and analyzing to obtain water quality parameters. The method has the effect of improving the continuity and authenticity of data detection during water quality detection.

Description

Technical Field

[0001] The present application relates to the technical field of water environment detection, and in particular to a water quality detection device and a method of use. Background Art

[0002] Water quality testing is a technology that measures the physical, chemical and biological indicators in water bodies. It can judge the water quality and identify pollution risks by analyzing various parameters of the water body, providing a basis for water resource protection, pollution control and water environment management. It is a basic work to ensure water resource security. In related technologies, water quality testing mostly relies on online monitoring equipment or portable testing instruments. The core detection component is the probe that contacts the water sample. During the long-term detection process, pollutants such as algae, suspended particles, and microbial films are easily attached to the probe surface. Existing equipment generally lacks automatic probe cleaning function and requires manual disassembly and cleaning on a regular basis. If the cleaning is not timely or thorough, pollutants will cover the probe sensing area and interfere with the detection signal. This is especially true when detecting indicators such as dissolved oxygen and turbidity that rely on direct contact with the probe, which can easily cause data deviations. Regarding the above-mentioned related technologies, since the detection equipment probe cannot be cleaned automatically, it not only increases the manual maintenance cost and reduces the detection continuity, but may also cause distortion of detection data due to the attachment of pollutants, making it difficult to accurately reflect the true condition of the water body, seriously affecting the timeliness and reliability of water quality monitoring. Summary of the Invention

[0003] In order to improve the continuity and authenticity of data detection during water quality testing, the present application provides a water quality testing device.

[0004] In the first aspect, the present application provides a water quality detection device, which adopts, for example, the following technical solution: A water quality testing device, comprising: The water sampling module is provided with a water quality sampling pipeline and is equipped with an adaptive sampling frequency adjustment strategy for adjusting the frequency of collecting water samples to be tested and delivered to the water quality detection module; The water sample storage module is provided with a water quality detection chamber and a detection probe, and is equipped with a detection chamber anti-clogging module for preventing the water sample flow path in the detection chamber from being blocked; The probe self-cleaning module is connected to the detection probe and is configured with a cleaning trigger strategy for automatically cleaning the surface of the detection probe; The water quality detection module is connected to the detection probe and is used to receive the detection signal from the detection probe and analyze it to obtain water quality parameters.

[0005] By adopting the above technical solution, the water sampling module can dynamically adjust the sampling frequency, the anti-clogging module of the water sample storage module prevents blockage of the detection chamber, the probe self-cleaning module ensures the cleanliness of the probe, and the water quality detection module analyzes the detection signal to obtain water quality parameters. The equipment can realize adaptive adjustment of the sampling frequency, reduce the risk of blockage in the detection chamber, maintain the cleanliness of the probe, thereby improving the efficiency and accuracy of water quality detection and ensuring stable operation of the equipment.

[0006] Optionally, the detection cavity anti-clogging module is configured with a clogging control strategy, including: Analyze the pressure fluctuation value and suspended solids concentration value of the water flow at the water inlet of the water quality detection chamber, and calculate the siltation threshold value using the preset siltation index analysis model; When the siltation threshold is greater than the preset risk siltation threshold, the corresponding siltation cleaning mode in the preset anti-siltation database is matched according to the siltation threshold, and the pipeline is cleaned according to the preset cleaning mode; Generate a congestion clearing feedback curve based on the change of the congestion threshold value over time during the clearing process, and use a preset feedback learning model to learn the clearing parameters and establish a clearing parameter optimization database; According to the cleaning parameter optimization database, the cleaning parameters and the corresponding cleaning mode are used to build a parameter iteration relationship and iterate the cleaning parameters.

[0007] By adopting the above technical solution, the clogging risk in the detection chamber is determined by calculating the clogging threshold, and then the corresponding cleaning mode is matched for cleaning. The cleaning parameters are optimized and iterated in combination with the feedback learning model. This helps to accurately identify the clogging risk, improve the pertinence and efficiency of clogging cleaning, continuously optimize the cleaning effect, and reduce the interference of clogging on detection.

[0008] Optionally, a blockage clearing assessment sub-strategy is also configured, including: Analyze the silt cleaning mode and the detection working state of the water quality detection equipment to determine whether the silt cleaning mode meets a preset detection interference condition, wherein the detection interference condition is that the silt cleaning mode affects the detection working state; If the conditions are met, the preset cleaning delay sub-strategy is triggered to delay the clogging cleaning time of the water sample flow path; If it is not satisfied, the blockage cleaning mode with the least impact on the water sample flow path is selected, and the water sample flow path is cleaned.

[0009] By adopting the above technical solution, the sub-strategy first evaluates the interference of the cleaning mode on the detection, and then decides to delay cleaning or select the low-impact mode based on the evaluation results. This helps to avoid the blockage cleaning process interfering with the detection work, ensure the continuity and accuracy of the detection, and take into account the blockage cleaning effect.

[0010] Optionally, the preset feedback learning model is learned using, for example, the following calculation formula: ; in, It is the correction value of the flow resistance adjustment parameter for the next iteration, which is used to optimize the operating parameters of the detection cavity anti-clogging module. It is a preset iteration coefficient used to control the basic amplitude of each correction amount, which is set according to the material of the detection chamber and the impurity characteristics of the water sample. is the partial derivative of the current adjustment parameter with respect to the flow resistance, Represents the actual flow resistance of the detection cavity, The anti-clogging adjustment parameters for the current iteration, including flushing pressure and airflow intensity, Indicates the preset target flow resistance, that is, the ideal resistance threshold when the detection chamber is operating normally. The actual total flow resistance of the detection chamber is calculated by real-time monitoring of the inlet and outlet pressure difference. The exponential coefficient of the preset target resistance and actual resistance ratio is used to amplify or reduce the impact of the difference between the two on the correction amount. Indicates that the pollutants attached to the inner wall of the detection cavity occupy the equivalent flow cross section, reflecting the degree of clogging. The maximum cross-section of the pollutant allowed in the detection chamber, It is the exponential coefficient of the preset pollutant occupancy rate, which is used to adjust the effect of the siltation degree on the correction amount. It is the preset weighted coefficient of historical correction, which is used to balance the impact of historical iteration on current correction. is the number of historical iterations involved in the calculation, is the weight of the j-th historical revision, is the correction value of the anti-clogging adjustment parameter of the jth historical iteration.

[0011] By adopting the above technical solution, the model comprehensively considers factors such as current flow resistance, target resistance, degree of siltation and historical correction amount to calculate the correction amount, which is used to optimize the operating parameters of the anti-siltation module, helping to make the cleaning parameter adjustment more in line with the actual siltation situation, improve the accuracy and adaptability of parameter optimization, and enhance the operating efficiency of the anti-siltation module.

[0012] Optionally, the adaptive sampling frequency adjustment strategy includes: Analyze historical test data and real-time monitoring parameters of water quality samples, and determine the initial sampling frequency level using a preset sampling frequency grading model; When the change amplitude of the real-time monitoring parameter exceeds the preset fluctuation threshold, the corresponding sampling frequency level in the preset frequency adjustment database is matched according to the change amplitude, and the number of sampling times is adjusted according to the matched level.

[0013] By adopting the above technical solution, the strategy first determines the initial frequency based on historical and real-time data, and then dynamically adjusts it according to parameter fluctuations, which helps to achieve on-demand adjustment of the sampling frequency, reduce unnecessary sampling when the water quality is stable, and increase the sampling density when the fluctuation is large, thereby improving the rationality of sampling and the timeliness of the data.

[0014] Optionally, a sampling frequency feedback adjustment sub-strategy is also included, using, for example, the following steps: Generate a frequency adjustment feedback curve based on the data change trend after the sampling frequency adjustment, analyze the change stability of water quality samples with the preset trend prediction model, and establish the correlation between frequency level and change stability; The sampling frequency level is dynamically iterated based on the association relationship. When the change stability continues to be higher than the benchmark value, the frequency level is reduced to the preset basic level in order of level.

[0015] By adopting the above technical solution, this sub-strategy optimizes the sampling frequency based on the data change trend and reduces the frequency when the water quality is stable, which helps to further reduce invalid sampling, save equipment resources and energy consumption, and at the same time ensure that the sampling frequency can respond in time when the water quality changes, balancing detection efficiency and resource consumption.

[0016] Optionally, the water quality detection module is further configured with a data quality assessment strategy: The probe contamination index is determined by analyzing the degree of foreign matter adhesion on the surface of the detection probe, and the detection reliability of the detection probe is calculated using a preset detection evaluation model; Dynamically assigning detection weights to detection probes of different probes based on changes in detection credibility to determine detection weights for different detection parameters; The detection parameters are preprocessed according to the detection weights of the detection parameters and the preset data weighted processing model to obtain the optimized water quality parameters.

[0017] By adopting the above technical solution, the strategy takes into account the impact of probe contamination on detection and optimizes data processing by dynamically allocating weights, which helps to reduce detection errors caused by probe contamination, improve the reliability and accuracy of water quality parameters, and enhance the quality of detection data.

[0018] Optionally, the weight calculation model for dynamically allocating detection weights adopts, for example, the following formula: ; in, is the dynamic weight of the i-th probe, 、 、 is the preset impact factor coefficient, is the exit sensitivity threshold of the i-th probe, is the current probe contamination index of the i-th probe, is the variance of the data measured by the i-th probe in the most recent time window, reflecting the data stability, is a preset calculation constant.

[0019] By adopting the above technical solution, the model calculates weights based on the probe's factory sensitivity, current contamination status, and data stability, which helps to make the weight distribution more in line with the actual detection capabilities of each probe and further improve the rationality and effectiveness of detection data fusion.

[0020] Optionally, the cleaning triggering strategy includes: Periodically and automatically check the measured value of the detection probe against the preset standard value to determine the probe detection drift; According to the probe detection drift, the corresponding pollution impact index in the preset pollution reference database is matched, and the cleanliness of the detection probe is calculated with the preset pollution index trigger threshold; When the cleanliness of the probe is lower than the preset minimum cleanliness, the detection probe self-cleaning instruction is triggered to perform self-cleaning on the detection probe; If the probe cleanliness is higher than the preset minimum cleanliness, the automatic cleaning instruction will be triggered according to the preset cleaning cycle.

[0021] By adopting the above technical solution, this strategy combines probe detection drift and cleanliness to determine the contamination status, and triggers cleaning on demand or on schedule, which helps to promptly remove probe surface contamination, maintain probe detection accuracy, reduce detection deviations caused by probe contamination, and ensure detection stability.

[0022] In a second aspect, the present application provides a method for using a water quality testing device, using, for example, the following technical solution: A method for using a water quality testing device, comprising: The water sample pretreatment step is equipped with a detection cavity anti-clogging module, which is used to control the flow state and pre-treat impurities of the water sample entering the device to prevent the detection cavity from clogging; Multi-parameter synchronous detection step, synchronous data acquisition of pre-treated water samples through multiple built-in water quality probes, including pH probe, dissolved oxygen probe, turbidity probe and conductivity probe; The data fusion analysis step is equipped with a data quality assessment strategy to perform validity screening and fusion calculation on the collected parameter data to generate a comprehensive water quality evaluation index; The equipment status self-diagnosis and maintenance steps are equipped with a probe self-cleaning module, which automatically executes the cleaning procedure according to the probe contamination status and preset maintenance cycle, and diagnoses and warns the overall operating status of the equipment.

[0023] By adopting the above technical solution, water sample pretreatment prevents cavity blockage, multi-parameter synchronous detection realizes comprehensive data collection, data fusion analysis generates a comprehensive evaluation index, and equipment self-diagnosis and maintenance ensure the equipment status. This method helps to achieve full process optimization of water quality testing, improve the comprehensiveness of testing, the validity of data and the stability of equipment, and enhance the scientific nature of water quality evaluation.

[0024] In summary, this application includes at least one of the following beneficial technical effects: 1. The water sampling module can dynamically adjust the sampling frequency. The anti-clogging module of the water sample storage module prevents clogging of the detection chamber. The probe self-cleaning module ensures probe cleanliness. The water quality detection module analyzes the detection signal to obtain water quality parameters. This equipment can achieve adaptive adjustment of the sampling frequency, reduce the risk of detection chamber clogging, maintain probe cleanliness, thereby improving the efficiency and accuracy of water quality testing and ensuring stable operation of the equipment. 2. Calculate the blockage threshold to determine the blockage risk within the detection chamber, then match the corresponding cleaning mode for cleaning. Combined with the feedback learning model, optimize the cleaning parameters and iterate. This helps to accurately identify blockage risks, improve the pertinence and efficiency of blockage cleaning, continuously optimize the cleaning effect, and reduce the interference of blockage on detection. 3. Combining probe detection drift and cleanliness to determine the contamination status and triggering cleaning on demand or on schedule helps to promptly remove probe surface contamination, maintain probe detection accuracy, reduce detection deviation caused by probe contamination, and ensure detection stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a method flow chart of steps S100 to S103 in this application.

[0026] Figure 2 It is a method flow chart of steps S104 to S1042 in this application.

[0027] Figure 3 It is a method flow chart of steps S200 to S203 in this application.

[0028] Figure 4 It is a method flow chart of steps S300 to S302 in this application.

[0029] Figure 5 It is a method flow chart of steps S400 to S403 in this application. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical solutions and advantages of this application more clear, the following Figure 1-5 It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application.

[0031] The embodiments of the present invention are described in further detail below with reference to the accompanying drawings.

[0032] An embodiment of the present application discloses a water quality detection device, in which the sampling frequency can be dynamically adjusted through a water body sampling module, the anti-clogging module of the water sample storage module prevents blockage of the detection chamber, the probe self-cleaning module ensures the cleanliness of the probe, and the water quality detection module analyzes the detection signal to obtain water quality parameters. The device can realize adaptive adjustment of the sampling frequency, reduce the risk of blockage of the detection chamber, maintain the cleanliness of the probe, thereby improving the efficiency and accuracy of water quality detection and ensuring stable operation of the equipment.

[0033] Reference Figure 1 , water quality testing equipment consists of the following modules: The water sampling module is provided with a water quality sampling pipeline and is equipped with an adaptive sampling frequency adjustment strategy for adjusting the frequency of collecting water samples to be tested and delivered to the water quality detection module; In this embodiment, the water sampling line is made of corrosion-resistant materials, such as 316 stainless steel or PVC-U. The line diameter is designed based on the sample flow rate (for example, a 15mm inner diameter accommodates a sampling flow rate of 0.5-2L / min). A filter (50μm pore size) is installed at the end of the line to intercept large impurities.

[0034] The adaptive sampling frequency adjustment strategy helps the device quickly increase the sampling frequency and capture sudden changes in water quality. The specific sampling frequency adjustment strategy will be further disclosed in subsequent steps.

[0035] The water sample storage module is provided with a water quality detection chamber and a detection probe, and is equipped with a detection chamber anti-clogging module for preventing the water sample flow path in the detection chamber from being blocked; The detection chamber is made of transparent polytetrafluoroethylene and features a cylindrical interior (60 mm diameter, 120 mm height). The chamber walls are equipped with water inlets and outlets (staggered to create a swirling flow) to ensure uniform contact between the water sample and the detection probe. The detection probe (e.g., pH electrode, dissolved oxygen sensor) is inserted vertically into the center of the chamber, with the sensing end of the probe 20 mm from the chamber bottom to prevent sedimentation interference. The chamber anti-clogging module includes a micro submersible agitator (5 W power, 1500 rpm) and a timed backflush assembly. The agitator activates every 15 minutes for 30 seconds to prevent suspended particles from settling. The backflush assembly injects air into the chamber using a high-pressure air pump (0.3 MPa) every morning to clear biofilm or impurities adhering to the inner walls of the pipes. For example, for water bodies with high algae content, the backflush frequency can be increased to every six hours to prevent algae growth from clogging the flow path.

[0036] The probe self-cleaning module is connected to the detection probe and is configured with a cleaning trigger strategy for automatically cleaning the surface of the detection probe; This module consists of a circular brush (nylon, 0.1mm diameter) and a drive motor. The brush wraps around the outside of the detection probe (with a spacing of 0.5mm) and is driven by the motor in a reciprocating motion (10mm stroke, 5mm / s). The cleaning trigger strategy includes two mechanisms: a timed trigger (default is every four hours) and a conditional trigger. The specific conditional trigger mechanism will be further explained later.

[0037] During the cleaning process, a brush is sprayed with a small amount of cleaning fluid (such as 0.1% dilute hydrochloric acid for scale contamination). After cleaning, the device is rinsed with pure water for 10 seconds to ensure that no residual residue interferes with the test. For example, for pH probes that are used to detect high-hardness water for a long time, a conditional trigger mechanism can be used to promptly remove surface scale and maintain detection accuracy.

[0038] The water quality detection module is connected to the detection probe and is used to receive the detection signal from the detection probe and analyze it to obtain water quality parameters.

[0039] This module features a built-in data acquisition card (sampling frequency 1kHz) and an embedded processor (such as an ARM Cortex-M4). It receives electrical signals from the probes (e.g., mV signals from pH electrodes and current signals from dissolved oxygen) and converts them into water quality parameters (e.g., pH, dissolved oxygen concentration, turbidity, etc.) using a calibration algorithm (e.g., three-point calibration). The module stores standard curves for each parameter (e.g., the current-concentration relationship for dissolved oxygen in the range of 0-20 mg / L) and supports data caching (capacity of 100,000 records) and remote transmission (via 4G / NB-IoT). For example, if a dissolved oxygen concentration is detected to be less than 5 mg / L, it is automatically flagged as "potential hypoxia," triggering an alert and simultaneously uploading the data to the monitoring platform.

[0040] The detection chamber anti-clogging module is configured with a clogging control strategy, including the following steps: Step S100: analyzing the pressure fluctuation value and the suspended solids concentration value of the water flow at the water inlet of the water quality detection chamber, and calculating the siltation threshold value using a preset siltation index analysis model; Accurate monitoring of water flow conditions and impurity levels is fundamental to assessing siltation risk. Step S100 deploys a high-frequency pressure sensor (with a sampling frequency of 100 Hz) and a laser suspended solids detector (with a detection accuracy of 0.1 mg / L) at the water inlet of the detection chamber to collect real-time water flow pressure and suspended solids concentration data. Pressure fluctuations are calculated by calculating the standard deviation of pressure data over a 5-second period, reflecting the degree of water flow disturbance. The suspended solids concentration is directly measured using the real-time concentration output by the detector (unit: mg / L).

[0041] The siltation index analysis model uses a weighted fusion algorithm to comprehensively calculate pressure fluctuation values ​​and suspended solids concentration values ​​according to preset weights (e.g., pressure fluctuation accounts for 30% and suspended solids concentration accounts for 70%). The formula is: siltation threshold = 0.3 × (pressure fluctuation value / pressure fluctuation upper limit) + 0.7 × (suspended solids concentration / suspended solids concentration upper limit), where the pressure fluctuation upper limit is set at 5 kPa (corresponding to the pressure fluctuation when the pipeline is significantly blocked) and the suspended solids concentration upper limit is set at 50 mg / L (corresponding to the critical concentration that easily causes siltation). For example, when the detected pressure fluctuation value is 3 kPa (i.e., 3 / 5 = 0.6) and the suspended solids concentration is 30 mg / L (i.e., 30 / 50 = 0.6), the siltation threshold = 0.3 × 0.6 + 0.7 × 0.6 = 0.6. This value can intuitively reflect the degree of siltation risk and provide a quantitative basis for subsequent cleanup decisions.

[0042] Step S101: When the siltation threshold is greater than a preset risk siltation threshold, a corresponding siltation cleaning mode in a preset anti-siltation database is matched according to the siltation threshold, and siltation is cleaned in the pipeline according to the preset cleaning mode; Targeted silt removal is key to ensuring efficient cleaning. Step S101 presets a risk siltation threshold (e.g., 0.5). When the siltation threshold exceeds this value, the system automatically calls an anti-siltation database (which pre-stores three cleaning modes: Mode 1 for siltation thresholds 0.5 < 0.7, Mode 2 for siltation thresholds 0.7 < 0.9, and Mode 3 for siltation thresholds > 0.9).

[0043] Specifically, Mode 1 uses low-pressure water flushing: a built-in flushing pump delivers 0.2MPa water flow in the reverse direction of the water inlet for one minute, using the impact of the water flow to remove lightly attached impurities. Mode 2 uses high-pressure flushing combined with air disturbance: a 0.6MPa high-pressure water flow is first flushed in the forward direction for 30 seconds, followed by a 0.1MPa compressed air disturbance for 10 seconds, repeated three times. This removes more stubborn clogs through the synergistic effect of water pressure and airflow. Mode 3 uses chemical cleaning combined with mechanical scraping: a 0.5% citric acid solution is first injected for five minutes to dissolve scale-like clogs, followed by a built-in micro-scraper (diameter matching the pipe) that scrapes back and forth along the inner wall twice to completely remove heavy clogs. For example, when the clog threshold is 0.8, the system switches to Mode 2, performing alternating cleaning with high-pressure water and airflow. This prevents damage to the pipe from excessive cleaning while effectively removing clogs.

[0044] Step S102: generating a congestion clearing feedback curve based on the change in the value of the congestion threshold over time during the clearing process, and performing cleaning parameter learning using a preset feedback learning model to establish a cleaning parameter optimization database; Dynamic feedback on cleaning effectiveness is key to iterative parameter optimization. Step S102 generates a feedback curve (for example, in Mode 2 cleaning, the curve might show a trend of rapid decline, gentle fluctuation, and stabilization) using the cleaning start time as the starting point (time in seconds on the horizontal axis) and the real-time collected congestion threshold as the vertical axis. By analyzing the curve's characteristics (e.g., rate of decline, threshold at stabilization), the effectiveness of the current cleaning parameters (e.g., high pressure value and airflow duration in Mode 2) is evaluated.

[0045] The specific analysis model will be further explained in subsequent steps. Based on the feedback curve, the system associates and stores cleaning parameters (such as flushing pressure, duration, airflow intensity, etc.) with corresponding cleaning effects (such as the time it takes for the clogging threshold to drop to a stable value) to form an optimization database. For example, if, during a Mode 2 cleaning, the curve is found to decrease slowly at a pressure of 0.6 MPa, the database will record the corresponding relationship between this parameter and the effect. In subsequent learning processes, the pressure can be adjusted to 0.65 MPa for optimization testing, gradually improving the matching degree between the cleaning parameters and the actual clogging situation, thereby achieving continuous improvement in cleaning efficiency.

[0046] The preset feedback learning model uses the following calculation formula for learning: ; in, It is the correction value of the flow resistance adjustment parameter for the next iteration, which is used to optimize the operating parameters of the detection cavity anti-clogging module. It is a preset iteration coefficient used to control the basic amplitude of each correction amount, which is set according to the material of the detection chamber and the impurity characteristics of the water sample. is the partial derivative of the current adjustment parameter with respect to the flow resistance, Represents the actual flow resistance of the detection cavity, The anti-clogging adjustment parameters for the current iteration, including flushing pressure and airflow intensity, Indicates the preset target flow resistance, that is, the ideal resistance threshold when the detection chamber is operating normally. The actual total flow resistance of the detection chamber is calculated by real-time monitoring of the inlet and outlet pressure difference. The exponential coefficient of the preset target resistance and actual resistance ratio is used to amplify or reduce the impact of the difference between the two on the correction amount. Indicates that the pollutants attached to the inner wall of the detection cavity occupy the equivalent flow cross section, reflecting the degree of clogging. The maximum cross-section of the pollutant allowed in the detection chamber, It is the exponential coefficient of the preset pollutant occupancy rate, which is used to adjust the effect of the siltation degree on the correction amount. It is the preset weighted coefficient of historical correction, used to balance the impact of historical iteration on current correction. is the number of historical iterations involved in the calculation, is the weight of the j-th historical revision, is the correction value of the anti-clogging adjustment parameter of the jth historical iteration.

[0047] Step S103: constructing a parameter iteration relationship based on the cleaning parameters in the cleaning parameter optimization database and the corresponding cleaning modes, and iterating the cleaning parameters.

[0048] Reference Figure 2 , and also configured with siltation clearing assessment sub-strategies, including: Step S104: analyzing the silt clearing mode and the detection working state of the water quality detection equipment to determine whether the silt clearing mode satisfies a preset detection interference condition, wherein the detection interference condition is that the silt clearing mode affects the detection working state; The coordination of cleaning and testing is key to ensuring the reliability of test data. Step S104 first uses the device status monitoring module to obtain real-time test status, including whether the device is currently in the sampling cycle (e.g., sampling every 5 minutes, lasting 1 minute), whether the detection probe is in the data collection phase (e.g., a pH probe records data every 30 seconds), and whether the device is in the stable detection window for key parameters (e.g., dissolved oxygen, turbidity) (requiring stable water flow for 2 minutes).

[0049] At the same time, quantitative evaluation indicators for detection interference conditions are preset: First, the water flow disturbance intensity is analyzed: whether the water flow velocity fluctuation caused by the cleaning mode exceeds ±0.2m / s (exceeding this will affect the probe detection stability); Secondly, the fluctuation threshold of the detection signal is analyzed to determine whether the cleaning process causes the real-time detection value of the probe to deviate from the stable value by more than 5%. For example, if the stable value of the turbidity probe is 10 NTU, a deviation of more than 0.5 NTU indicates interference. A third party will test the integrity of data records: whether cleaning will cause loss or interruption of sampling data, such as cleaning during the sampling phase will result in incomplete water sample collection.

[0050] During analysis, the characteristic parameters of the silt removal mode, such as flushing pressure, airflow intensity, and duration, are matched with the detection operating state parameters. For example, during the detection phase (when the probe is recording data), a high-pressure flushing mode (0.6MPa, 30 seconds) will cause water flow disturbances of up to ±0.3m / s, exceeding the threshold and thus meeting the detection interference condition. However, a low-pressure flushing mode (0.2MPa, 10 seconds) will only cause fluctuations of ±0.1m / s during the detection phase, which does not exceed the threshold and is therefore not met.

[0051] Step S1041: If the conditions are met, trigger the preset cleaning delay sub-strategy to delay the clogging cleaning time of the water sample flow path; The core purpose of cleaning delay is to avoid critical testing periods, helping to ensure data validity. Step S1041 presets cleaning delay rules: First, prioritize delays to the interval between testing sessions (e.g., the four-minute window between the end of sampling and the next sampling session). Second, if the monitoring phase is continuous (e.g., data is collected every 10 seconds during emergency monitoring), a minimum delay is calculated (to ensure that cleaning after the delay does not affect the next round of testing, e.g., delaying to 2 minutes allows for a single data recording). Furthermore, a maximum delay threshold is set (e.g., a maximum delay of no more than 10 minutes to prevent further congestion).

[0052] For example, when the equipment is in the stable detection window period of the dissolved oxygen probe (water flow must be stable for 2 minutes), and the matching cleaning mode is high-pressure flushing (interference conditions are met), the system automatically queries the next detection interval (ending 1.5 minutes after the start of the current detection, and the interval is 3 minutes), and delays the cleaning until immediately after the end of the detection, avoiding interference with the current detection and preventing siltation accumulation.

[0053] Step S1042: If the conditions are not met, the blockage cleaning mode with the least impact on the water sample flow path is selected, and the water sample flow path is cleaned.

[0054] The principle of minimal impact is the core of balancing cleaning effectiveness and detection stability. Step S1042 pre-sets an impact rating assessment system, quantifying the impact of each cleaning mode on the flow path from three dimensions: the first dimension is the degree of path disturbance; the second dimension is the rate of change in flow resistance (a change in pipe resistance before and after cleaning of less than 5% is low impact, scored 2 points; 5%-10% is medium impact, scored 5 points; >10% is high impact, scored 8 points); and the third dimension is recovery time (the time it takes for the water flow to stabilize after cleaning is less than 30 seconds, scored 2 points; 30 seconds-1 minute is medium impact, scored 5 points; >1 minute is high impact, scored 8 points).

[0055] The system calculates the comprehensive impact score of each mode through weighted calculation and selects the mode with the lowest score.

[0056] Reference Figure 3 , the adaptive sampling frequency adjustment strategy includes: Step S200: Analyze historical test data and real-time monitoring parameters of water quality samples, and determine the initial sampling frequency level using a preset sampling frequency grading model; Historical test data refers to the cumulative water quality test results of the equipment over the past period of time (such as the last 7 days), covering key parameters such as pH value, dissolved oxygen concentration, turbidity, and conductivity. After being recorded by the storage module, this data can reflect the long-term changes in water quality (such as whether it is stable or whether there are periodic fluctuations).

[0057] Real-time monitoring parameters refer to the water quality data continuously collected by the equipment within the current short period (such as the last 5 minutes), which is transmitted to the analysis unit in real time by the water sampling module to reflect the immediate status of the water quality (such as whether there is a sudden change).

[0058] The initial sampling frequency level is the initial sampling interval level determined based on historical and real-time data. It serves as the benchmark for subsequent frequency adjustments. A higher level (such as L5) indicates more frequent sampling (shorter intervals), while a lower level (such as L1) indicates longer sampling intervals.

[0059] First, the historical detection data is processed: the fluctuation variance of each parameter is calculated through a sliding window algorithm (the window size is set to 24 hours) (the smaller the variance, the more stable the historical data), and the "historical fluctuation coefficient" is determined according to the variance range (for example, a variance <0.1 corresponds to a coefficient of 1.0, 0.1-0.3 corresponds to 2.0, and >0.3 corresponds to 3.0).

[0060] Secondly, process the real-time monitoring parameters: take the average monitoring value in the last 5 minutes (collected every 10 seconds, and the average is taken to reduce accidental errors), calculate the deviation ratio between the average value and the industry standard value (for example, the dissolved oxygen standard value is 5 mg / L, the current average value is 5.3 mg / L, and the deviation ratio is 6%), and determine the "real-time status coefficient" based on the deviation ratio (for example, deviation <5% corresponds to 1.0, 5%-10% corresponds to 2.0, and >10% corresponds to 3.0).

[0061] Finally, the initial grade is calculated using a preset sampling frequency grading model: the model adds a weighted sum of the historical volatility coefficient (weight 60%) and the real-time status coefficient (weight 40%) to obtain an initial grade score, and then matches the corresponding frequency grade based on the score (e.g., 5 grades L1-L5, L1 is the basic grade with a sampling interval of 60 minutes; L5 is the highest grade with a sampling interval of 1 minute).

[0062] By integrating historical patterns with real-time status, the initial sampling frequency will neither ignore current changes due to complete reliance on history, nor fall into blind high-frequency sampling due to only looking at real-time conditions. This provides a reasonable starting point for subsequent adjustments and balances detection accuracy and resource consumption.

[0063] For example, if the fluctuation variance of pH data over the past seven days is 0.03 (relatively stable, with a historical fluctuation coefficient of 1.0), and the real-time average pH value within five minutes is 7.1 (a deviation of 1.4% from the standard value of 7.0, with a real-time state coefficient of 1.0), then the initial grade score = 0.6 × 1.0 + 0.4 × 1.0 = 1.0, matching the L1 grade, and the initial sampling frequency is set to 60 minutes / time.

[0064] Step S201: When the variation range of the real-time monitoring parameter exceeds a preset fluctuation threshold, the corresponding sampling frequency level in the preset frequency adjustment database is matched according to the variation range, and the number of sampling times is adjusted according to the matched level.

[0065] The amplitude of change refers to the ratio of the difference between the current real-time monitoring parameter and the parameter at the previous moment to the parameter value at the previous moment (expressed as a percentage), which is used to quantify the instantaneous change degree of water quality parameters (for example, the dissolved oxygen at the previous moment was 5.0 mg / L, and it is currently 6.0 mg / L, with a change amplitude of 20%).

[0066] The fluctuation threshold is a preset critical value for judging whether the water quality has changed significantly. The thresholds for different parameters are different (for example, the fluctuation threshold for pH is ±0.2, and for dissolved oxygen is ±0.5 mg / L). Exceeding this value indicates that the water quality may enter an unstable state.

[0067] The preset frequency adjustment database is a pre-stored correspondence between the "variation range" and the "sampling frequency level", such as a variation range of 5%-15% corresponds to the L2 level, 15%-30% corresponds to the L3 level, etc., which is used to quickly match the adjustment strategy.

[0068] When executing this step, the device calculates the change amplitude of each parameter in real time. If the change amplitude of a parameter (such as dissolved oxygen) exceeds its fluctuation threshold, the frequency adjustment is triggered: the sampling frequency level corresponding to the change amplitude is searched from the frequency adjustment database (the larger the amplitude, the higher the matching level), and the sampling frequency is adjusted according to the new level (for example, the original L1 level is 1 time / hour, and after adjustment to L3 level, it becomes 4 times / hour).

[0069] When water quality fluctuates significantly, the sampling frequency is increased to capture the details of the changes in a timely manner, avoiding missing key data due to long sampling intervals and ensuring accurate tracking of water quality dynamics.

[0070] For example: The real-time turbidity value monitored at the previous moment was 10NTU, and the current value is 13NTU. The change range = (13-10) / 10×100%=30%, which exceeds the turbidity fluctuation threshold of ±2NTU (corresponding to a change range of 20%). The L4 level corresponding to "30%-50%" in the matching frequency adjustment database is adjusted, and the number of sampling times is adjusted from the original L1 1 time / hour to 12 times / hour (5 minutes / time).

[0071] It also includes a sampling frequency feedback regulation sub-strategy, which adopts the following steps: Step S202: Generate a frequency adjustment feedback curve based on the data change trend after the sampling frequency adjustment, analyze the change stability of the water quality sample using a preset trend prediction model, and establish a correlation between the frequency level and the change stability; Frequency adjustment feedback curve: a curve with time as the horizontal axis and the fluctuation value of the water quality parameter after the sampling frequency is adjusted as the vertical axis. The fluctuation value = |current parameter value - the average value of the parameter during the period|, which reflects the changing trend of water quality after the frequency adjustment (if the curve is flat, it means the change is stable, and if it is steep, it means there are still sharp fluctuations).

[0072] Stability of changes: quantified by the "stability index", the index range is 0-1 (1 is completely stable), and the calculation method is "1-(fluctuation standard deviation of the last 10 data points / historical maximum fluctuation standard deviation)". The smaller the standard deviation and the higher the index, the more stable the water quality.

[0073] Trend prediction model: An algorithm used to analyze feedback curves (such as exponential smoothing algorithms) establishes a correlation between "sampling frequency level" and "change stability" by calculating the stability index (for example, a high-frequency level L5 usually corresponds to a low stability index, and a low-frequency level L1 corresponds to a high stability index).

[0074] When executing this step, after the sampling frequency is adjusted, the system continuously records parameter changes and generates a feedback curve; the stability index of the curve is calculated through the trend prediction model, and the difference in stability at different frequency levels is analyzed (for example, the stability index at level L4 is 0.6, and at level L3 it is 0.7), and the corresponding relationship between the level and the index is stored to form an associated database.

[0075] By evaluating whether the current sampling frequency matches the water quality stability (for example, if the stability has improved under high-frequency sampling, it means that the frequency can be appropriately reduced), data support is provided for subsequent frequency iterations to avoid oversampling or undersampling.

[0076] For example: After the frequency is adjusted to L4, the standard deviation of the dissolved oxygen data fluctuation for two consecutive hours drops from 0.8 mg / L to 0.3 mg / L, and the historical maximum standard deviation of fluctuation is 1.0 mg / L. The stability index = 1-(0.3 / 1.0) = 0.7. At this time, the L4 level is stored in association with the stability index of 0.7.

[0077] Step S203: dynamically iterating the sampling frequency level based on the association relationship, and when the change stability is continuously higher than the reference value, lowering the frequency level to a preset basic level in order of level.

[0078] Baseline value: A preset critical stability index (such as 0.7) for determining whether water quality is stable. A value higher than this value indicates that water quality changes slowly and does not require high-frequency sampling.

[0079] Dynamic iteration: refers to gradually adjusting the frequency level (such as transitioning from L4 to L3, L2, and L1) based on the continuous status of the stability index, rather than a one-time adjustment, to ensure a smooth transition.

[0080] Basic level: The preset minimum sampling frequency level (such as L1, 60 minutes / time), which is the default level when water quality is stable for a long time.

[0081] When executing this step, the system monitors the stability index in real time. If the index is higher than the benchmark value of 0.7 for three consecutive sampling periods (such as 15 minutes at level L4), the current level will be downgraded by one level; monitoring will continue at the new level. If the stability continues to meet the standard, the system will repeat the downgrade until it returns to the basic level L1.

[0082] By gradually reducing the sampling frequency after the water quality stabilizes, the energy consumed by invalid sampling and equipment loss can be reduced, and a dynamic balance of "high-frequency monitoring during fluctuations and low-frequency maintenance during stability" can be achieved, thereby improving equipment operating efficiency.

[0083] For example: If the current level is L4 (5 minutes / time), and the stability indexes for three consecutive cycles (15 minutes) are 0.75, 0.8, and 0.82 (all > 0.7), the system will automatically be downgraded to L3 (15 minutes / time). At the L3 level, if the stability index remains > 0.7 for the next three cycles, the system will be downgraded to L2 and finally return to L1, and the sampling interval will be restored to 60 minutes / time.

[0084] Reference Figure 4 , the water quality detection module is also configured with a data quality assessment strategy: Step S300: Analyzing the degree of foreign matter adhesion on the surface of the detection probe to determine the probe contamination index, and calculating the detection reliability of the detection probe using a preset detection evaluation model; In step S300, the detection probe is a sensor component that comes into direct contact with the water sample, such as a pH probe, dissolved oxygen probe, or turbidity probe. The cleanliness of its surface directly affects detection accuracy. Surface foreign matter adhesion describes the degree of foreign matter, such as dirt, microorganisms, and sediment, adhering to the probe surface. It is usually measured physically or optically, such as through transmittance attenuation or surface resistance change. The probe contamination index quantifies this adhesion, with a value between 0 and 1, where 0 indicates complete cleanliness and 1 indicates severe contamination. Higher values ​​indicate more severe contamination.

[0085] The detection assessment model is an algorithm used to convert the contamination index into a test reliability score. It takes into account factors such as probe type (for example, turbidity probes are more susceptible to contamination) and the location of contamination (for example, contamination in sensitive detection areas has a greater impact). The test reliability score, on the other hand, reflects the reliability of the probe's current test results. It also ranges from 0 to 1, with 1 indicating complete reliability. Generally speaking, a higher contamination index indicates a lower reliability score.

[0086] During specific implementation, the probe surface is first scanned using built-in auxiliary detection devices, such as micro cameras and optical sensors. The optical sensor emits light of a specific wavelength, such as 650nm red light, to detect the intensity of the reflected light. The reflectivity is high when clean and decreases when foreign matter is attached. This allows the degree of foreign matter adhesion on the surface to be calculated. The probe contamination index is then determined based on the adhesion and the preset pollution level standard, for example, a 10% decrease in reflectivity corresponds to a pollution index of 0.1, and a 50% decrease corresponds to 0.5. The pollution index is then input into the detection and evaluation model. The model will correct the pollution index based on factors such as the probe's historical pollution patterns (for example, a pH probe's pollution index increases by 0.2 per week) and the current detection environment (for example, accelerated pollution in high-turbidity water samples), and ultimately output the detection credibility.

[0087] This step accurately identifies probes affected by contamination, quantifies the reliability of their test results, and provides a "credibility scale" for subsequent data fusion, mitigating detection errors caused by contamination at the source. For example, if sediment accumulates on the surface of a turbidity probe, the optical sensor detects a 35% decrease in reflectivity compared to a clean state. Based on the standard, the contamination index is determined to be 0.35. The detection evaluation model takes into account that the probe has been used in turbid water for two days (historical data shows that the impact of contamination is slightly greater in this environment). After correcting the contamination index, the output is a detection confidence level of 0.78 (the confidence level for a clean probe is 0.95).

[0088] Step S301: dynamically assigning detection weights to detection probes of different probes according to changes in detection reliability to determine detection weights of different detection parameters; In step S301, dynamic weighting refers to the process of flexibly adjusting the contribution of each probe's measured parameter to the final comprehensive result based on real-time changes in probe detection reliability—higher reliability results in higher weights, lower reliability results in lower weights—rather than using fixed weights. The detection parameters are specific water quality indicators output by each probe, such as pH, dissolved oxygen concentration, turbidity, and conductivity. Each parameter corresponds to the detection results of one or more probes.

[0089] During execution, the system will monitor the changes in the detection credibility of each probe in real time: when the credibility of a probe increases, for example, from 0.7 to 0.9, the weight of its corresponding detection parameter will be increased; when the credibility decreases, for example, from 0.9 to 0.6, the weight will be reduced. In the allocation process, the inherent importance of the parameters will be taken into account. For example, in drinking water testing, the basic weight values ​​of pH value and microbial indicators are higher than turbidity, but the core basis is still detection credibility. For example, the basic weight is set to pH 0.3, dissolved oxygen 0.25, turbidity 0.2, and conductivity 0.25. If the credibility of the dissolved oxygen probe decreases, its actual weight will be reduced proportionally based on 0.25. The specific model will be further disclosed later.

[0090] This step ensures that reliable test data dominates the final results, mitigating the impact of unreliable data. This ensures that the contribution of each parameter matches its reliability, thereby enhancing the rationality of data fusion. For example, the initial base weights are pH (0.3), dissolved oxygen (0.25), turbidity (0.2), and conductivity (0.25). If, during testing, the dissolved oxygen probe is found to be contaminated, its reliability drops from 0.9 to 0.6 (a 33% decrease). Its actual weight is adjusted to 0.25 × (0.6 / 0.9) ≈ 0.17; the pH probe's reliability remains at 0.95, and its weight remains at 0.3. The final total weights are adjusted to pH 0.3, dissolved oxygen 0.17, turbidity 0.25, and conductivity 0.28 (the total weights sum to 1).

[0091] Step S302: Preprocessing the detection parameters according to the detection weights of the detection parameters and a preset data weighted processing model to obtain optimized water quality parameters.

[0092] In step S302, the data weighting model is an algorithm used to integrate the weighted values ​​of each test parameter to calculate the final comprehensive water quality parameters. This model may include processing logic such as weighted averaging, outlier removal, and trend smoothing. The optimized water quality parameters, after weight adjustment and preprocessing, are comprehensive test results that more closely reflect the actual state of the water sample, preserving the valid information of each parameter while reducing error interference.

[0093] During execution, the raw data for each test parameter is first preprocessed: significant outliers, such as those exceeding the probe's range, are removed; environmental interference, such as the effect of temperature on conductivity, is corrected; and high-frequency noise is smoothed, for example, by using a sliding average to eliminate transient fluctuations. Subsequently, each preprocessed parameter value is multiplied by the corresponding test weight to obtain a weighted value. These weighted values ​​are then integrated using a data weighting model, such as calculating a weighted average or applying weights based on parameter importance. Ultimately, the optimized water quality parameters are output.

[0094] This step transforms multi-source, varying degrees of credibility into a unified, reliable composite result through weighted integration and preprocessing. This improves the accuracy and stability of water quality parameters and provides a high-quality data foundation for subsequent water quality assessment. For example, the preprocessed data includes a pH of 7.2 (weight 0.3), dissolved oxygen of 5.8 mg / L (weight 0.17), turbidity of 6.0 NTU (weight 0.25), and conductivity of 320 μS / cm (weight 0.28). Using a weighted data processing model (e.g., weighted averaging), the optimized water quality parameters reflect "medium cleanliness, slightly alkaline pH, and normal dissolved oxygen," with the higher weights of pH and conductivity being more significant.

[0095] The weight calculation model for dynamically allocating detection weights uses the following formula: ; in, is the dynamic weight of the i-th probe, 、 、 is the preset impact factor coefficient, is the exit sensitivity threshold of the i-th probe, is the current probe contamination index of the i-th probe, is the variance of the data measured by the i-th probe in the most recent time window, reflecting the data stability, is a preset calculation constant.

[0096] Reference Figure 5 , cleaning trigger strategies include: Step S400: Periodically and automatically verifying the measured value of the detection probe with a preset standard value to determine the probe detection drift; In step S400, the probe's measurement value refers to the real-time data output by the probe during the actual detection process, such as the water sample's pH value measured by a pH probe or the dissolved oxygen concentration measured by a dissolved oxygen probe. The preset standard value is a calibrated, known, and accurate value, typically derived from the device's built-in standard solution module (e.g., pH 7.0 buffer, saturated dissolved oxygen solution, etc.), which serves as a calibration benchmark. Periodic automatic calibration is a comparison process automatically initiated by the device at regular intervals (e.g., every two hours). The comparison is performed by comparing the probe's real-time measurement value with the standard value to determine if the probe has detected any deviation.

[0097] Probe drift is a measure of this deviation, typically expressed as the difference between the measured value and the reference value (it can be divided into absolute drift and relative drift). For example, if the reference value of a pH probe is 7.0 and the measured value during calibration is 7.3, the absolute drift is 0.3. If the reference value is 5.0 and the measured value is 5.5, the relative drift is (5.5-5.0) / 5.0×100%=10%.

[0098] During calibration, the device automatically switches to a standard solution environment at a preset calibration time (e.g., on the hour or half-hour), places the probe in contact with the standard solution, and reads the measured value. The difference between the measured value and the standard value is then calculated to determine the probe drift for that period. This step promptly detects probe detection deviations caused by surface contamination, aging, and other factors, providing a direct basis for determining whether cleaning is necessary. A greater drift indicates more severe probe contamination.

[0099] For example, a turbidity probe with a nominal value of 10 NTU and automatic calibration every two hours may measure 10.2 NTU during the first calibration, with a drift of 0.2 NTU. Three hours later, during the next calibration, the measured value is 11.5 NTU, with a drift of 1.5 NTU. This indicates that the probe may be accumulating contaminants and the drift has increased significantly.

[0100] Step S401: matching the probe detection drift with the corresponding pollution impact index in the preset pollution reference database, and calculating the detection probe cleanliness with the preset pollution index trigger threshold; In step S401, the preset pollution reference database is a table pre-stored in the device that maps "probe detection drift" to a "pollution impact index." The pollution impact index quantifies the contribution of pollution to probe detection drift, ranging from 0 to 1. Higher values ​​indicate more significant drift caused by pollution (for example, a drift of 1.0 NTU might correspond to a pollution impact index of 0.5, while a drift of 2.0 NTU might correspond to a pollution impact index of 0.8). A preset pollution index trigger threshold determines whether the pollution impact has reached a critical value (e.g., 0.6) requiring attention. Exceeding this threshold indicates significant pollution interference with detection.

[0101] Probe cleanliness is an indicator of probe surface cleanliness, derived by combining drift and contamination effects. It ranges from 0 to 1 (1 being completely clean) and is typically calculated as "1 - (Contamination Impact Index / Contamination Index Trigger Threshold)." For example, if the Contamination Impact Index is 0.5 and the trigger threshold is 0.8, the cleanliness is 1 - 0.5 / 0.8 = 0.375.

[0102] During execution, the device matches the probe drift measured in step S400 with a corresponding contamination impact index in the contamination reference database. This index is then compared with the contamination index trigger threshold to determine the current probe cleanliness. This step transforms the abstract drift into a more intuitive cleanliness indicator, allowing the device to more accurately determine the probe's contamination status and provide a quantitative basis for cleaning decisions.

[0103] For example, the detection drift of a dissolved oxygen probe is 0.8 mg / L. The corresponding pollution impact index matched in the pollution reference database is 0.6. The preset pollution index trigger threshold of the equipment is 0.7. Then the cleanliness of the probe = 1-0.6 / 0.7≈0.14, indicating that the surface contamination of the probe is serious.

[0104] Step S402: When the cleanliness of the probe is lower than a preset minimum cleanliness, triggering a self-cleaning instruction of the detection probe to perform self-cleaning on the detection probe; In step S402, the preset minimum cleanliness level (e.g., 0.3) is the threshold for determining whether the probe requires immediate cleaning. If the probe cleanliness level falls below this level, surface contamination has significantly impacted detection accuracy and prompt cleaning is necessary. The probe self-cleaning command is a signal sent by the device to the probe self-cleaning module. Once triggered, the module cleans the probe surface through methods such as physical wiping, water flushing, and ultrasonic cleaning (depending on the probe type).

[0105] The execution logic is straightforward: the device compares the cleanliness level obtained in step S401 with the minimum cleanliness level. If the cleanliness level is lower, a self-cleaning instruction is immediately generated, initiating the cleaning process. This step ensures that severely contaminated probes are cleaned promptly, preventing distortion of test data due to persistent contamination and ensuring the reliability of test results.

[0106] For example, the device has a preset minimum cleanliness level of 0.3, and the cleanliness level of a pH probe is calculated to be 0.25, which is lower than the critical value. At this time, the device will immediately trigger the self-cleaning instruction and control the cleaning module to perform ultrasonic cleaning on the pH probe to remove dirt attached to the surface.

[0107] Step S403: If the cleanliness of the probe is higher than the preset minimum cleanliness, a self-cleaning instruction is issued according to a preset cleaning cycle.

[0108] In step S403, the preset cleaning cycle is the regular maintenance interval set by the device for the probe (for example, every 24 hours). Even if the probe cleanliness level is above the minimum cleanliness level (i.e., contamination is not severe enough to affect detection), cleaning is automatically triggered when the cycle expires. This is a preventative maintenance strategy designed to prevent the gradual accumulation of contaminants through regular cleaning, thus avoiding sudden, severe contamination of the probe between calibrations.

[0109] During execution, the device monitors both cleanliness and time. If the cleanliness level exceeds the minimum cleanliness level, cleaning will not start immediately. Instead, the current time will be recorded. When the accumulated operating time reaches the preset cleaning interval (e.g., 24 hours have passed since the last cleaning), a self-cleaning command will be automatically issued. This step balances immediate cleaning with regular maintenance, reducing unnecessary frequent cleaning (to protect the life of the probe) while preventing contamination accumulation through periodic maintenance, maintaining long-term stable probe operation.

[0110] For example, if the cleanliness level of a turbidity probe is 0.4 (higher than the minimum cleanliness level of 0.3), the device has a preset cleaning cycle of 24 hours, and the last cleaning time was 8:00 yesterday, then at 8:00 today, even if the probe cleanliness level still meets the standard, the device will trigger the self-cleaning instruction according to the cycle and flush the probe with low-pressure water to remove lightly attached impurities.

[0111] Based on the same inventive concept, an embodiment of the present invention provides a method for using a water quality testing device, comprising: The water sample pretreatment step is equipped with a detection cavity anti-clogging module, which is used to control the flow state and pre-treat impurities of the water sample entering the device to prevent the detection cavity from clogging; This step is equipped with a detection cavity anti-clogging module, whose core function is to regulate the flow state and pre-treat impurities of water samples entering the equipment, thereby reducing the risk of detection cavity blockage from the source.

[0112] Flow state control is achieved through the device's built-in flow control valve and diversion structure: the flow control valve is adjusted to a preset stable flow rate (such as 0.8L / min) based on the initial flow rate of the water sample (such as the real-time monitoring of the water inlet rate of 1.2L / min) to avoid impurity deposition due to turbulent water flow; the diversion structure adopts a spiral diversion design to make the water sample form a vortex before entering the detection chamber, and use centrifugal force to preliminarily separate large particles of impurities (such as mud and debris with a diameter of >0.5mm).

[0113] Impurity pretreatment combines filter filtration and sedimentation pretreatment: a 50-mesh filter is set at the water inlet to intercept large suspended particles; at the same time, the flow rate of the water sample is slowed down through a buffer sedimentation chamber (volume 500mL), allowing impurities with higher density (such as heavy metal precipitates) to settle naturally under the action of gravity (sedimentation time is about 30 seconds), and the supernatant then enters the detection chamber.

[0114] During this process, the detection chamber anti-clogging module simultaneously runs the clogging control strategy: real-time monitoring of the water inlet pressure fluctuations and suspended matter concentration, calculation of the clogging threshold, if close to the risk threshold (such as the clogging threshold reaches 0.45, the preset risk threshold is 0.5), then start low-pressure flushing (0.2MPa) in advance to prevent impurities from adhering to the pipe wall.

[0115] The purpose of this step is to provide stable and clean water samples for subsequent testing, reduce testing errors and equipment blockages caused by unstable water flow or excessive impurities, and ensure the continuity of the testing process.

[0116] Multi-parameter synchronous detection step, synchronous data acquisition of pre-treated water samples through multiple built-in water quality probes, including pH probe, dissolved oxygen probe, turbidity probe and conductivity probe; In this step, multiple water quality probes built into the equipment are used to synchronously collect data on the pretreated water samples. The probes used include pH probes, dissolved oxygen probes, turbidity probes, and conductivity probes to achieve comprehensive monitoring of key parameters of the water samples.

[0117] Synchronous data acquisition is achieved through a timeline calibration mechanism: the device's internal clock synchronizes each probe to initiate detection at the same time (e.g., every 10-second interval), ensuring data corresponds to the same water sample state. The probes perform the following functions: the pH probe detects the pH of the water sample by changes in electrode potential (measuring range 0-14 pH, accuracy ±0.01 pH); the dissolved oxygen probe uses fluorescence to measure dissolved oxygen concentration in the water (measuring range 0-20 mg / L, accuracy ±0.05 mg / L); the turbidity probe measures turbidity by 90° scattered light intensity (measuring range 0-1000 NTU, accuracy ±0.1 NTU); and the conductivity probe measures the conductivity of the water sample by changes in inter-electrode resistance (measuring range 0-20,000 μS / cm, accuracy ±1%).

[0118] During the collection process, the device simultaneously runs an adaptive sampling frequency adjustment strategy: if the real-time change amplitude of a parameter (such as turbidity) exceeds the fluctuation threshold (such as ±2NTU), the sampling frequency of the parameter will be automatically increased (such as from 30 minutes / time to 5 minutes / time) to ensure that dynamic changes in water quality are captured.

[0119] The purpose of this step is to obtain comprehensive and timely water sample information through multi-parameter synchronous acquisition and dynamic sampling, providing rich and accurate raw data for subsequent data analysis.

[0120] The data fusion analysis step is equipped with a data quality assessment strategy to perform validity screening and fusion calculation on the collected parameter data to generate a comprehensive water quality evaluation index; This step is equipped with a data quality assessment strategy to identify the effectiveness and perform fusion calculations on the collected parameter data, and ultimately generate a comprehensive water quality evaluation index to improve the reliability and comprehensiveness of the data.

[0121] The effectiveness screening is mainly based on the credibility of the probe detection: the pollution index is determined by analyzing the degree of foreign matter adhesion on the surface of each probe, and the credibility is calculated in combination with the detection evaluation model (for example, the turbidity probe pollution index is 0.3, corresponding to a credibility of 0.78); data with a credibility lower than the threshold (such as 0.6) is eliminated (for example, the dissolved oxygen probe credibility is 0.52 at a certain moment, and its data is marked as invalid).

[0122] The fusion calculation is based on dynamically assigned detection weights: weights are adjusted based on the reliability of each probe (e.g., a pH probe with a reliability of 0.95 has a weight of 0.3; a dissolved oxygen probe with a reliability of 0.7 has a weight of 0.2). Valid data is then integrated using a data weighting model (e.g., weighted averaging combined with parameter importance stratification). For example, if the preprocessed data is pH 7.2 (weight 0.3), dissolved oxygen 5.8 mg / L (weight 0.2), turbidity 6.0 NTU (weight 0.25), and conductivity 320 μS / cm (weight 0.25), the fusion calculation yields a comprehensive water quality evaluation index (e.g., 85 out of 100) and generates an evaluation conclusion (e.g., "water quality is good, turbidity is slightly elevated").

[0123] The purpose of this step is to eliminate the interference of unreliable data through data screening and fusion, transform multi-parameter information into intuitive comprehensive evaluation results, and provide a scientific basis for water quality judgment.

[0124] The equipment status self-diagnosis and maintenance steps are equipped with a probe self-cleaning module, which automatically executes the cleaning procedure according to the probe contamination status and preset maintenance cycle, and diagnoses and warns the overall operating status of the equipment.

[0125] This step is equipped with a probe self-cleaning module, which automatically executes the cleaning procedure according to the probe contamination status and the preset maintenance cycle, and diagnoses and warns the overall operating status of the equipment to ensure long-term stable operation of the equipment.

[0126] Probe self-cleaning is based on a cleaning trigger strategy: the probe detection drift (such as pH probe drift 0.3pH) is determined through periodic calibration (such as once every 2 hours), and the cleanliness level (such as cleanliness 0.25) is calculated after matching the pollution impact index. If the cleanliness level is lower than the minimum cleanliness level (such as 0.3), self-cleaning is immediately triggered (such as ultrasonic cleaning for 10 seconds). If the cleanliness level meets the standard, preventive cleaning (such as low-pressure water flushing) is performed according to the preset period (such as 24 hours).

[0127] Equipment status self-diagnosis covers the operating parameters of each module: real-time monitoring of the flow of the sampling module, the pressure of the detection chamber, the energy consumption of the cleaning module, etc. If a parameter exceeds the normal range (such as a sudden increase of 10% in the detection chamber pressure), an early warning will be automatically issued (such as an audible and visual alarm + background notification) and abnormal information will be recorded (such as "2025-07-10 14:30 The detection chamber pressure is abnormal, possibly blocked").

[0128] The purpose of this step is to maintain the probe detection accuracy through active cleaning and status diagnosis, timely detect and warn of equipment failures, reduce maintenance costs, and extend the service life of the equipment.

[0129] The above steps are closely linked, forming a closed loop from water sample pretreatment to equipment maintenance, which not only ensures the accuracy and comprehensiveness of water quality testing, but also realizes the intelligent operation and self-maintenance of the equipment.

[0130] Those skilled in the art will clearly understand that for the sake of convenience and brevity, the division of the above-mentioned functional modules is only used as an example for illustration. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working processes of the above-mentioned systems, devices, and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0131] An embodiment of the present invention provides a computer-readable storage medium storing a computer program that can be loaded by a processor and executed to implement a method for using a water quality detection device.

[0132] Computer storage media include, for example, various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0133] Based on the same inventive concept, an embodiment of the present invention provides an intelligent terminal, including a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and execute a method for using a water quality detection device.

[0134] Those skilled in the art will clearly understand that for the sake of convenience and brevity, the division of the above-mentioned functional modules is only used as an example for illustration. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working processes of the above-mentioned systems, devices, and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0135] The above are all preferred embodiments of the present application and are not intended to limit the scope of protection of this application. Unless otherwise specified, any feature disclosed in this specification (including the abstract and drawings) may be replaced by other equivalent or similar features. In other words, unless otherwise specified, each feature is merely an example of a series of equivalent or similar features.

Claims

1. A water quality testing device, characterized in that: include: The water sampling module is provided with a water quality sampling pipeline and is equipped with an adaptive sampling frequency adjustment strategy for adjusting the frequency of collecting water samples to be tested and delivered to the water quality detection module; The water sample storage module is provided with a water quality detection chamber and a detection probe, and is equipped with a detection chamber anti-clogging module for preventing the water sample flow path in the detection chamber from being blocked; The probe self-cleaning module is connected to the detection probe and is configured with a cleaning trigger strategy for automatically cleaning the surface of the detection probe; The water quality detection module is connected to the detection probe and is used to receive the detection signal from the detection probe and analyze it to obtain water quality parameters.

2. A water quality testing device according to claim 1, characterized in that: The detection chamber anti-clogging module is equipped with a clogging control strategy, including: Analyze the pressure fluctuation value and suspended solids concentration value of the water flow at the water inlet of the water quality detection chamber, and calculate the siltation threshold value using the preset siltation index analysis model; When the siltation threshold is greater than the preset risk siltation threshold, the corresponding siltation cleaning mode in the preset anti-siltation database is matched according to the siltation threshold, and the pipeline is cleaned according to the preset cleaning mode; Generate a congestion clearing feedback curve based on the change of the congestion threshold value over time during the clearing process, and use a preset feedback learning model to learn the clearing parameters and establish a clearing parameter optimization database; According to the cleaning parameter optimization database, the cleaning parameters and the corresponding cleaning mode are used to build a parameter iteration relationship and iterate the cleaning parameters.

3. A water quality testing device according to claim 2, characterized in that, It also configures a blockage clearing assessment sub-strategy, including: Analyze the silt cleaning mode and the detection working state of the water quality detection equipment to determine whether the silt cleaning mode meets a preset detection interference condition, wherein the detection interference condition is that the silt cleaning mode affects the detection working state; If the conditions are met, the preset cleaning delay sub-strategy is triggered to delay the clogging cleaning time of the water sample flow path; If it is not satisfied, the blockage cleaning mode with the least impact on the water sample flow path is selected, and the water sample flow path is cleaned.

4. A water quality testing device according to claim 2, characterized in that, The preset feedback learning model uses the following calculation formula for learning: ； in, It is the correction value of the flow resistance adjustment parameter for the next iteration, which is used to optimize the operating parameters of the detection cavity anti-clogging module. It is a preset iteration coefficient used to control the basic amplitude of each correction amount, which is set according to the material of the detection chamber and the impurity characteristics of the water sample. is the partial derivative of the current adjustment parameter with respect to the flow resistance, Represents the actual flow resistance of the detection cavity, The anti-clogging adjustment parameters for the current iteration, including flushing pressure and airflow intensity, Indicates the preset target flow resistance, that is, the ideal resistance threshold when the detection chamber is operating normally. The actual total flow resistance of the detection chamber is calculated by real-time monitoring of the inlet and outlet pressure difference. The exponential coefficient of the preset target resistance and actual resistance ratio is used to amplify or reduce the impact of the difference between the two on the correction amount. Indicates that the pollutants attached to the inner wall of the detection cavity occupy the equivalent flow cross section, reflecting the degree of clogging. The maximum cross-section of the pollutant allowed in the detection chamber, It is the exponential coefficient of the preset pollutant occupancy rate, which is used to adjust the effect of the siltation degree on the correction amount. It is the preset weighted coefficient of historical correction, used to balance the impact of historical iteration on current correction. is the number of historical iterations involved in the calculation, is the weight of the j-th historical revision, is the correction value of the anti-clogging adjustment parameter of the jth historical iteration.

5. A water quality testing device according to claim 1, characterized in that: The adaptive sampling frequency adjustment strategy includes: Analyze historical test data and real-time monitoring parameters of water quality samples, and determine the initial sampling frequency level using a preset sampling frequency grading model; When the change amplitude of the real-time monitoring parameter exceeds the preset fluctuation threshold, the corresponding sampling frequency level in the preset frequency adjustment database is matched according to the change amplitude, and the number of sampling times is adjusted according to the matched level.

6. A water quality testing device according to claim 5, characterized in that: It also includes a sampling frequency feedback regulation sub-strategy, which adopts the following steps: Generate a frequency adjustment feedback curve based on the data change trend after the sampling frequency adjustment, analyze the change stability of water quality samples with the preset trend prediction model, and establish the correlation between frequency level and change stability; The sampling frequency level is dynamically iterated based on the association relationship. When the change stability continues to be higher than the benchmark value, the frequency level is reduced to the preset basic level in order of level.

7. A water quality testing device according to claim 1, characterized in that: The water quality detection module is also equipped with a data quality assessment strategy: The probe contamination index is determined by analyzing the degree of foreign matter adhesion on the surface of the detection probe, and the detection reliability of the detection probe is calculated using a preset detection evaluation model; Dynamically assigning detection weights to detection probes of different probes based on changes in detection credibility to determine detection weights for different detection parameters; The detection parameters are preprocessed according to the detection weights of the detection parameters and the preset data weighted processing model to obtain the optimized water quality parameters.

8. A water quality testing device according to claim 7, characterized in that: The weight calculation model of the dynamic allocation detection weight adopts the following formula, for example: ； in, is the dynamic weight of the i-th probe, 、 、 is the preset impact factor coefficient, is the exit sensitivity threshold of the i-th probe, is the current probe contamination index of the i-th probe, is the variance of the data measured by the i-th probe in the most recent time window, reflecting the data stability, is a preset calculation constant.

9. A water quality testing device according to claim 8, characterized in that: The cleaning triggering strategy includes: Periodically and automatically check the measured value of the detection probe against the preset standard value to determine the probe detection drift; According to the probe detection drift, the corresponding pollution impact index in the preset pollution reference database is matched, and the cleanliness of the detection probe is calculated with the preset pollution index trigger threshold; When the cleanliness of the probe is lower than the preset minimum cleanliness, the detection probe self-cleaning instruction is triggered to perform self-cleaning on the detection probe; If the probe cleanliness is higher than the preset minimum cleanliness, the automatic cleaning instruction will be triggered according to the preset cleaning cycle.

10. A method for using a water quality testing device, using the water quality testing device according to any one of claims 1 to 9, characterized in that: include: The water sample pretreatment step is equipped with a detection cavity anti-clogging module, which is used to control the flow state and pre-treat impurities of the water sample entering the device to prevent the detection cavity from clogging; Multi-parameter synchronous detection step, synchronous data acquisition of pre-treated water samples through multiple built-in water quality probes, including pH probe, dissolved oxygen probe, turbidity probe and conductivity probe; The data fusion analysis step is equipped with a data quality assessment strategy to perform validity screening and fusion calculation on the collected parameter data to generate a comprehensive water quality evaluation index; The equipment status self-diagnosis and maintenance steps are equipped with a probe self-cleaning module, which automatically executes the cleaning procedure according to the probe contamination status and preset maintenance cycle, and diagnoses and warns the overall operating status of the equipment.

Images