Automatic water quality monitoring method

Abstract

According to the technical scheme, the automatic water quality monitoring method is characterized by comprising the following steps that on-site sampling is conducted, and sample information is input into a printing tool and printed on a sample; transporting and handing over the water sample, scanning and inputting all sample information into a management system, and recording the sample entering time; intelligent sample distribution and circulation; performing intelligent water sample detection by using a water quality analyzer; data analysis and transmission and intelligent sample distribution and circulation comprise a sample test distribution strategy and a quality control scheme, in the method, each sample has a unique digital identity card, the system automatically distributes tasks, optimizes a detection path, greatly improves circulation efficiency, avoids deviation caused by manual selection, can continuously run for 24 hours, and has a wide application prospect. Rapid treatment of a large number of samples is achieved, and the high-throughput requirement of modern monitoring is met.

Description

Technical Field

[0001] This invention relates to a monitoring method, and more specifically, to an automated water quality monitoring method. Background Technology

[0002] To comprehensively promote the digital transformation of the ecological and environmental monitoring system and build a modern monitoring model centered on data intelligence, technological innovation is also needed in the field of water quality monitoring. Traditional water quality laboratory analysis processes rely on manual operation, which is not only inefficient but also suffers from data delays and human errors, making it difficult to meet the new era's demands for high-frequency, high-precision, and fully traceable monitoring. Against this backdrop, intelligent unmanned surface water monitoring laboratories have emerged, becoming an important technological carrier for promoting the automation and intelligent transformation of water quality monitoring. These laboratories deeply integrate artificial intelligence, the Internet of Things, and big data analysis technologies. By integrating various intelligent analytical instruments, they enable unmanned operation of pretreatment steps based on existing equipment. They can flexibly adapt to changes in testing items and technical requirements, comprehensively access sample site information, equipment operating status, and analysis results data, and introduce intelligent review algorithms to achieve intelligent upgrades in data quality control and interpretation. This provides more timely and reliable technical support for water environment management and decision-making, laying a key technological foundation for building a digital and intelligent water ecological environment monitoring system characterized by "precise perception, intelligent judgment, and dynamic early warning." Summary of the Invention

[0003] To address the shortcomings of existing technologies, an automated water quality monitoring method is provided, which designs unmanned operation for the pretreatment steps based on existing equipment.

[0004] To achieve the above objectives, the following technical solution is provided: an automated water quality monitoring method, comprising the following steps: S1: On-site sampling, inputting sample information into a printing tool and printing it onto the sample; S2: Water sample transportation and handover, all sample information is scanned and entered into the management system, and the time of sample entry is recorded; S3: Intelligent sample distribution and transfer; S4: Use a water quality analyzer for intelligent water sample testing; S5: Data analysis and transmission.

[0005] Intelligent sample allocation and circulation includes sample testing allocation strategies and quality control schemes.

[0006] Furthermore, the sample testing allocation strategy includes the following steps: S1: Screening water samples; S2: The system scans each sample in the order of its number on the sample stage to check if all parameters have been detected. If the detection is complete, proceed to S6; otherwise, proceed to S3. S3: Confirm whether the sampling station and corresponding analysis module of the specified analysis module for the sample are idle. If not idle, execute S6; if idle, execute S4. S4: Grab the sample and place it into the designated analysis module for testing; S5: Waiting for the robotic arm to become idle; S6: Scan the next sample; S7: If a sampling station has finished sampling during the period, the robotic arm will prioritize retrieving the water sample. S8: Repeatedly scan all samples until all parameters of all samples have been detected.

[0007] Furthermore, the screening of water samples includes the following steps: S11-1: Manually assign an analysis module to each water sample on the sample stage.

[0008] Furthermore, the screening of water samples includes the following steps: S11-2: Manually select the water sample to be tested on the sample stage; S12: The robotic arm scans each sample and enters the parameters that need to be measured for each sample, as well as other sample information.

[0009] Furthermore, the screening of water samples also includes the following steps: S13: The system automatically assigns and determines the corresponding analysis module for each water sample based on the entered information.

[0010] Furthermore, the quality control plan includes the following steps: S1: Prepare at least 100 sets or more of samples; S2: Set the water sample analysis task with monitoring parameters. Each analytical instrument needs to perform 10% parallel samples / day, 10% quality control / day, and 2 blanks / day.

[0011] S3: After the analysis task is issued, record the start time of the analysis; S4: Record the number of sample groups analyzed the following day.

[0012] Furthermore, the quality control plan also includes the following steps: S5: Steps S1-S4 were tested continuously for 2 weeks, and the system recorded the number of analyses for each detection factor; Record the type, frequency, and recovery time of system malfunctions; Record the handling of used sample bottles and the collection of waste liquid; Record the specific tasks and duration of each day's manual work; Record the number of water samples, parallel samples, quality controls, and blanks analyzed daily; Record the types and quantities of faults during operation, including robot malfunctions, sample conveyor belt malfunctions, and instrument malfunctions. Record the automatic stacking of used sample bottles and the collection of waste liquid; Record the number of manual operations and the duration.

[0013] In summary, the above technical solution has the following beneficial effects: In this method, each sample has a unique "digital ID card," and its status, location, and handover time can be queried and traced in real time, ensuring the integrity of the sample chain. The system automatically allocates tasks, optimizes the detection path, and significantly improves the flow efficiency. At the same time, it avoids the bias caused by human selection. The quality control link is fully or partially automatically executed as an essential part of the process, ensuring the stability and reliability of data quality. Meanwhile, the instrument can run continuously for 24 hours, realizing the rapid processing of a large number of samples and meeting the high-throughput requirements of modern monitoring. Data is processed and transmitted immediately after it is generated, greatly shortening the cycle from sampling to obtaining results, supporting rapid decision-making. Combining sample site information, instrument status, etc., the system can perform preliminary intelligent judgment and review, mark abnormal data, and improve the accuracy and credibility of data reports. Attached Figure Description

[0014] Figure 1 This is a flowchart illustrating the first embodiment.

[0015] Figure 2 This is a flowchart illustrating the second embodiment.

[0016] Figure 3 This is a flowchart illustrating the third embodiment. Detailed Implementation

[0017] Automated water quality monitoring methods include the following steps: S1: On-site sampling, inputting sample information into a printing tool and printing it onto the sample; S2: Water sample transportation and handover, all sample information is scanned and entered into the management system, and the time of sample entry is recorded; S3: Intelligent sample distribution and transfer; S4: Use a water quality analyzer for intelligent water sample testing; S5: Data analysis and transmission.

[0018] Intelligent sample allocation and circulation includes sample testing allocation strategies and quality control schemes.

[0019] This automated method deeply integrates the Internet of Things, big data, and laboratory information management systems. At the sampling site, staff directly input key information such as sample number, location, and time into portable devices and generate scannable labels (such as QR codes) that are printed on the sample containers. Typically, the sample containers are 1000mL HDPE water sample bottles uniformly provided by the smart unmanned laboratory. The sampling volume for each type of water sample is generally 1000mL. If more parameters need to be analyzed, multiple bottles can be collected as needed.

[0020] The sample labels are printed using a portable printer during sampling and can be connected to the smart unmanned laboratory system via an APP or WeChat mini-program. The label content includes: the sampler, sampling location, sampling completion date and time (accurate to the hour), test items, unique item number, unique sample code, what preservative was added, and water surface characteristics.

[0021] After the samples are delivered to the unmanned laboratory, the sample handover should be completed in the unmanned laboratory as soon as possible. All sample information should be scanned and entered into the intelligent unmanned laboratory management system. At the same time, the time when the sample enters the laboratory should be recorded. When the sample is handed over and entered, the sample can be placed on the sample table in the laboratory and the sample number on the sample table should be recorded so that the laboratory sample circulation system can determine the testing order.

[0022] The system automatically allocates samples to corresponding testing lines or workstations based on preset sample testing allocation strategies and quality control plans, and schedules robots or conveyor belts to complete the transfer. Quality control plans include blind sample testing, automatic spiked recovery testing, blank verification, standard sample verification, parallel sample testing, and dynamic standard sample verification. The sample is automatically fed into the designated water quality analyzer. The instrument automatically completes the sample introduction, reagent addition, reaction measurement, and data acquisition without human intervention. The raw data generated by the analyzer is automatically captured, and the system performs calibration, calculation, and validity verification based on the algorithm. Finally, the reviewed data is uploaded to the central data platform in real time. The sample analysis data is transmitted immediately after measurement, and the process information is recorded and uploaded to the platform simultaneously. It also has the function of displaying historical data curves, timely detection of cross-sectional data changes, and abnormal warnings.

[0023] The sample transfer device mainly consists of three parts: a sample stage, a robotic arm, and a sampling table. After the collected water samples are sent to the unmanned laboratory, the operator places the samples to be tested on the sample stage. The sample stage can hold up to 200 water samples (5*5*8=200). After the unmanned laboratory management system sets the testing tasks, the robotic arm grabs the water sample bottle from the sample stage according to the sample testing allocation logic, scans the water sample information, and places it on the sampling table. After the analyzer finishes sampling, the robotic arm is responsible for returning the water sample bottle to its original position. After the robotic arm places the water sample bottle on the sampling table, the sampling table grabs and fixes the water sample bottle. First, a rotating arm unscrews the cap of the water sample bottle, then the stirring rod and sampling tube are inserted into the bottle and stirring is started. After the analyzer finishes sampling, the stirring rod and sampling tube are removed from the water sample bottle and inserted into the cleaning cup on the sampling table for cleaning. Finally, the rotating arm screws the cap back on, and the robotic arm grabs the water sample bottle back to the sample stage.

[0024] In this method, each sample has a unique "digital ID card," and its status, location, and handover time can be queried and traced in real time, ensuring the integrity of the sample chain. The system automatically allocates tasks, optimizes the detection path, and significantly improves throughput efficiency. At the same time, it avoids the bias caused by human selection. The quality control links are fully or partially automated as an essential part of the process, ensuring the stability and reliability of data quality. Meanwhile, the instrument can run continuously for 24 hours, enabling rapid processing of a large number of samples and meeting the high-throughput requirements of modern monitoring. Data is processed and transmitted immediately after it is generated, greatly shortening the cycle from sampling to obtaining results and supporting rapid decision-making. Combining on-site sample information and instrument status, the system can perform preliminary intelligent judgment and review, mark abnormal data, and improve the accuracy and credibility of data reports.

[0025] The sample testing allocation strategy includes the following steps: S1: Screening water samples; S2: The system scans each sample in the order of its number on the sample stage to check if all parameters have been detected. If the detection is complete, proceed to S6; otherwise, proceed to S3. S3: Confirm whether the sampling station and corresponding analysis module of the specified analysis module for the sample are idle. If not idle, execute S6; if idle, execute S4. S4: Grab the sample and place it into the designated analysis module for testing; S5: Waiting for the robotic arm to become idle; S6: Scan the next sample; S7: If a sampling station has finished sampling during the period, the robotic arm will prioritize retrieving the water sample. S8: Repeatedly scan all samples until all parameters of all samples have been detected.

[0026] This strategy serves as the scheduling hub of the unmanned laboratory. Through a set of efficient cyclic scanning and decision-making algorithms, it directs the robotic arm to work collaboratively, ensuring that the detection tasks are carried out smoothly and orderly. The system selects the sample queue to be tested from the received sample pool. The system scans each sample in sequence according to the sample stage number. The core judgment is "Have all parameters of this sample been detected?". If they have been detected, S6 is executed to skip the sample and scan the next one. If they have not been detected, S3 is executed. For unfinished samples, the system checks whether the sampling stage and analysis module required for the next step are available. If they are not available, it executes S6 to continue scanning the next sample. If they are available, it executes S4. The system instructs the robotic arm to grab the current sample and place it in an idle designated analysis module for testing. A waiting instruction is inserted into the task queue until the robotic arm becomes idle, so that the next task can be executed. At any time during the cyclic scanning, as long as an analysis module completes the test and empties the sampling station, the system will prioritize instructing the robotic arm to perform the action of "retrieving the sample that has been tested", making room for the next round of testing. The system continuously repeats the scanning and judgment process from S2 to S7 until all parameters of all samples have been tested.

[0027] By continuously exploring executable tasks through a cyclic scanning mechanism, the system maximizes time utilization and avoids resource idleness. Real-time status monitoring and scheduling of key resources such as robotic arms and analysis modules ensure optimal matching of tasks and resources. Even with manual operation steps, the efficiency of the laboratory is significantly improved. The action of "retrieving completed samples" has a higher priority, ensuring that the analysis module can be released as soon as possible to receive the next sample. This is a key mechanism to ensure high-throughput detection.

[0028] As a first embodiment, the water sample screening includes the following steps: S11-1: Manually assign an analysis module to each water sample on the sample stage.

[0029] After the samples are placed on the sample stage, the operators manually select or assign the required analytical modules in the management system according to the specific test items of each water sample. This method can cope with the complex and ever-changing test requirements of various water samples. When there are unconventional test items or special water samples, manual judgment can flexibly and accurately match the most suitable analytical equipment. It is highly adaptable, the workflow is intuitive and clear, the test targets and test devices are determined at the source, the task flow of the entire system is clear, and it is easy for personnel to understand and intervene. Based on a global understanding of the sample volume and the status of each module, the operators can make overall allocations to avoid the "uneven workload" that may occur in the automatic allocation of the system, thereby achieving a high equipment utilization rate at the manual scheduling level. However, each sample requires manual operation. When the number of samples is large and the testing frequency is high, this becomes a heavy and repetitive task, which is not only inefficient but also increases the risk of errors due to human negligence.

[0030] As a second embodiment, the water sample screening includes the following steps: S11-2: Manually select the water sample to be tested on the sample stage; S12: The robotic arm scans each sample and enters the parameters that need to be measured for each sample, as well as other sample information.

[0031] The operator pre-selects the subset of samples that need to enter the testing process in this round on the sample stage. After the system is started, the robotic arm automatically scans each selected sample during its movement and automatically enters all the parameters that need to be measured and other detailed information into the system. Compared with manually assigning analysis modules to each sample, this solution greatly reduces the workload of manual data entry. The operator only needs to make batch selections, and the specific parameter information is automatically collected by the system, which improves efficiency and reduces the risk of human error. At the same time, special samples can be separately allocated. However, when the sample volume is huge and the test parameters of a single sample are numerous, the detection task queue generated after the robotic arm scans will be extremely complex. If the system's dynamic scheduling algorithm is not powerful enough, it is very easy for the detection tasks of different samples to conflict in terms of time and resources, which may cause test chaos, such as robotic arm path conflicts, excessive waiting time of analysis modules, etc., and ultimately reduce equipment utilization.

[0032] This solution can effectively improve efficiency in small- to medium-scale scenarios or scenarios with controllable task complexity. However, its performance is highly dependent on the intelligence of the background task scheduling algorithm. When facing extreme challenges of high throughput and multiple parameters, if the scheduling logic is poor, its inherent limitations will become apparent, thereby affecting the smoothness and efficiency of the entire system.

[0033] As a third embodiment, based on the second embodiment, the water sample screening also includes the following steps: S13: The system automatically assigns and determines the corresponding analysis module for each water sample based on the entered information.

[0034] After acquiring all parameter information of the sample through scanning with a robotic arm, the system automatically assigns a suitable analysis module to each water sample based on built-in algorithms and rules, without any human intervention. This solution completely eliminates the need for manual designation or preliminary selection in the early stages, handing over the task allocation decision-making power to the system, thereby greatly reducing manual operation and improving process efficiency and automation. To achieve this function, the system needs to add and rely on an intelligent task allocation engine. This engine needs to have powerful logic processing capabilities and be able to make optimal decisions by comprehensively considering the type of detection project, the busy / idle status of each analysis module, the priority of the detection task, and the throughput of the entire system.

[0035] The quality control plan includes the following steps: S1: Prepare at least 100 sets or more of samples; S2: Set the water sample analysis task with monitoring parameters. Each analytical instrument needs to perform 10% parallel samples / day, 10% quality control / day, and 2 blanks / day.

[0036] S3: After the analysis task is issued, record the start time of the analysis; S4: Record the number of sample groups analyzed the following day.

[0037] The quality control plan specifically includes the following: Standard sample verification / dynamic standard sample verification: Each analyzer shall complete one verification per day; Spiked recovery test: Each water sample from each sampling point should be tested once a month for each parameter, and only one analyzer is needed; Blank check: Each analyzer shall be checked once per day; Parallel sample testing: Each analyzer completes one test for every 10 samples, at least once a day; Multi-point linearity verification: Each analyzer is checked once a month; Water sample comparison: Each water sample from each sampling point is tested once a month for each parameter, and only one analyzer is needed; By setting up multi-dimensional and multi-frequency quality control links, a "full-process, multi-level, and intelligent" quality control system has been built to ensure data accuracy and instrument stability. Through daily "standard sample verification" and "blank verification", as well as monthly "multi-point linearity verification", the accuracy, sensitivity and baseline stability of each analyzer can be continuously monitored to ensure that the instruments are always in the best working condition and to ensure the accuracy of individual data points from the source. Spike recovery testing is specifically designed to assess whether there is matrix interference or operational loss throughout the entire process from sample pretreatment to analysis and detection. Parallel sample testing is used to determine the precision of analytical results and detect random errors in a timely manner. The combination of the two can accurately pinpoint whether data deviations originate from the sample itself, the pretreatment process, or the analytical instrument. The plan employs high-frequency daily checks for key indicators (such as instrument status) and periodic monthly checks for comprehensive indicators (such as overall process accuracy). At the same time, the parallel sample rule of "once for every 10 samples" achieves a dynamic balance between workload and quality control intensity, ensuring both monitoring strength and avoiding unnecessary resource consumption. Water sample comparison requires cross-validation of the results of the same sample on different instruments or at different times, which effectively ensures the long-term stability and horizontal comparability of the output data of the entire laboratory, and is crucial for environmental trend analysis. The types, frequencies, and standards of all quality control tasks have been programmed and can be automatically triggered, executed, and judged by the system. This perfectly matches the unattended nature of the smart laboratory, realizes the intelligent operation of the quality control process, and ensures the objectivity and impartiality of quality control data.

[0038] The quality control plan also includes the following steps: S5: Steps S1-S4 were tested continuously for 2 weeks, and the system recorded the number of analyses for each detection factor; Record the type, frequency, and recovery time of system malfunctions; Record the handling of used sample bottles and the collection of waste liquid; Record the specific tasks and duration of each day's manual work; Record the number of water samples, parallel samples, quality controls, and blanks analyzed daily; Record the types and quantities of faults during operation, including robot malfunctions, sample conveyor belt malfunctions, and instrument malfunctions. Record the automatic stacking of used sample bottles and the collection of waste liquid; Record the number of manual operations and the duration.

[0039] System continuity testing mainly examines whether the system's full-load throughput and system stability meet the requirements.

[0040] System full-load flux: The measurement is divided into two phases: continuous 12-hour operation and 24-hour operation. The flux calculation methods are as follows: 1. 12-hour continuous operation: System throughput can be calculated by running continuously for about 12 hours. The conversion formula is: System full-load throughput (units / hour) = Number of instruments corresponding to a certain parameter × Sample throughput of a single instrument (units / hour) × 80%; 2. 24-hour continuous operation: The system throughput can be calculated by running continuously for about 24 hours. During the full-load throughput test, the reagents used in each laboratory are also pure water.

[0041] When the system is running at full load, observe the operation of the system and analysis equipment, observe the overall system stability, and record the types of failures such as robot failure, sample conveyor belt failure, and instrument failure during operation; record the work done manually during operation, and evaluate the level of automation of the system; record the automatic stacking of used sample bottles and the collection of waste liquid.

[0042] The specific testing method is as follows: Prepare 100-200 or more sets of samples (tap water, pure water), and set monitoring parameters for water sample analysis tasks; each analyzer (1 detector corresponds to 1 analyzer) needs to perform 10% parallel samples / day, 10% quality control / day, and 2 blanks / day; After the analysis task is issued, record the start time of the analysis; record the number of sample groups analyzed the following day; (run continuously for 12 hours or 24 hours). The above process was tested continuously for 2 weeks, and the system recorded the number of analyses for each detection factor (24H / 12H). Record the type, frequency, and recovery time of system malfunctions; Record the handling of used sample bottles and the collection of waste liquid; Record the specific work and duration of each person each day. Total duration = number of people * individual duration. The system acquires and analyzes ≥200 samples per day (based on statistics of 24-hour continuous operation or conversion of 12-hour continuous operation). Record the number of water samples, parallel samples, quality controls, and blanks analyzed daily; Record the types and quantities of faults during operation, including robot malfunctions, sample conveyor belt malfunctions, and instrument malfunctions. Record the automatic stacking of used sample bottles and the collection of waste liquid; Record the number of manual operations and the duration.

Claims

1. An automated water quality monitoring method, characterized in that, It includes the following steps: S1: On-site sampling, inputting sample information into a printing tool and printing it onto the sample; S2: Water sample transportation and handover, all sample information is scanned and entered into the management system, and the time of sample entry is recorded; S3: Intelligent sample distribution and transfer; S4: Use a water quality analyzer for intelligent water sample testing; S5: Data analysis and transmission.

2. The intelligent sample allocation and circulation includes sample testing allocation strategies and quality control schemes.

3. The automated water quality monitoring method according to claim 1, characterized in that, The sample testing and allocation strategy includes the following steps: S1: Screening water samples; S2: The system scans each sample in the order of its number on the sample stage to check if all parameters have been detected. If the detection is complete, proceed to S6; otherwise, proceed to S3. S3: Confirm whether the sampling station and corresponding analysis module of the specified analysis module for the sample are idle. If not idle, execute S6; if idle, execute S4. S4: Grab the sample and place it into the designated analysis module for testing; S5: Waiting for the robotic arm to become idle; S6: Scan the next sample; S7: If a sampling station has finished sampling during the period, the robotic arm will prioritize retrieving the water sample. S8: Repeatedly scan all samples until all parameters of all samples have been detected.

4. The automated water quality monitoring method according to claim 2, characterized in that, The water sample screening process includes the following steps: S11-1: Manually assign an analysis module to each water sample on the sample stage.

5. The automated water quality monitoring method according to claim 2, characterized in that, The water sample screening process includes the following steps: S11-2: Manually select the water sample to be tested on the sample stage; S12: The robotic arm scans each sample and enters the parameters that need to be measured for each sample, as well as other sample information.

6. The automated water quality monitoring method according to claim 4, characterized in that, The water sample screening also includes the following steps: S13: The system automatically assigns and determines the corresponding analysis module for each water sample based on the entered information.

7. The automated water quality monitoring method according to claim 1, characterized in that, The quality control plan includes the following steps: S1: Prepare at least 100 sets or more of samples; S2: Set the water sample analysis task with monitoring parameters. Each analytical instrument needs to perform 10% parallel samples / day, 10% quality control / day, and 2 blanks / day. 8.S3: After the analysis task is issued, record the start time of the analysis; S4: Record the number of sample groups analyzed the following day.

9. The automated water quality monitoring method according to claim 6, characterized in that, The quality control plan also includes the following steps: S5: Steps S1-S4 were tested continuously for 2 weeks, and the system recorded the number of analyses for each detection factor; Record the type, frequency, and recovery time of system malfunctions; Record the handling of used sample bottles and the collection of waste liquid; Record the specific tasks and duration of each day's manual work; Record the number of water samples, parallel samples, quality controls, and blanks analyzed daily; Record the types and quantities of faults during operation, including robot malfunctions, sample conveyor belt malfunctions, and instrument malfunctions. Record the automatic stacking of used sample bottles and the collection of waste liquid; Record the number of manual operations and the duration.

Images