Quantum dot spectrum sensor-based water quality monitoring method and system

Abstract

The invention provides a water quality monitoring method and system based on a quantum dot spectrum sensor, which can obviously improve the accuracy and reliability of water quality monitoring and effectively reduce the false alarm rate. The method comprises the following steps: obtaining an initial chemical oxygen demand time sequence of a to-be-detected water body based on a quantum dot spectrum sensor, screening out candidate judgment moments, and screening out a chemical oxygen demand time sequence of a preset second duration according to a preset second time interval; obtaining at least one auxiliary hydrology and water quality parameter time sequence, and removing working condition abnormal data to obtain a screened chemical oxygen demand time sequence; extracting a target chemical oxygen demand time sequence from the screened chemical oxygen demand time sequences; when the sub evaluation parameter of each sub chemical oxygen demand time sequence and the overall evaluation parameter of the target chemical oxygen demand time sequence meet a third preset condition, determining a trend abnormal moment; and generating alarm information based on whether a trend abnormal moment exists in a preset third duration before the candidate judgment moment.

Description

Technical Field

[0001] This application relates to the field of water quality monitoring technology, and more specifically, to a water quality monitoring method and system based on a quantum dot spectral sensor. Background Technology

[0002] Water quality monitoring is a crucial aspect of environmental protection and water resource management. Existing water quality monitoring systems typically collect and store water quality data, then directly compare this data with preset thresholds to determine if anomalies exist. When anomalies are detected, an alarm is generated and sent to the user. However, this monitoring method is relatively crude in its data analysis. Considering the complexity of the aquatic environment, the causes of these anomalies can be varied, including non-aquatic pollution. Furthermore, if the anomaly data is not properly processed, there is a risk of triggering false alarms, making it difficult to meet the requirements for high-precision and high-reliability pollution alarms. Summary of the Invention

[0003] In view of this, the purpose of this application is to provide a water quality monitoring method and system based on a quantum dot spectral sensor to overcome the problems in the prior art.

[0004] In a first aspect, embodiments of this application provide a water quality monitoring method based on a quantum dot spectral sensor, the method comprising: The initial chemical oxygen demand (COD) time series of the water body to be tested is obtained based on a quantum dot spectral sensor. Monitoring times that meet the first preset condition are selected as candidate judgment times according to a preset first time interval. The first preset condition is that the COD at the monitoring time is greater than the first preset threshold. The first preset threshold is obtained based on the COD within a preset first time period before the monitoring time. A chemical oxygen demand (COD) time series of a preset second duration is selected from the initial COD time series according to a preset second time interval; Obtain at least one auxiliary hydrological and water quality parameter time series corresponding to the chemical oxygen demand time series with the preset second duration. When the first data point in the auxiliary hydrological and water quality parameter time series meets the second preset condition, the first data point is determined to be abnormal operating data. The second preset condition is that the first data point in the at least one auxiliary hydrological and water quality parameter time series is less than the corresponding second preset threshold, the preset first duration is less than the preset second duration, and the preset first time interval is less than the preset second time interval. Remove the abnormal operating data from the chemical oxygen demand time series with the preset second duration to obtain the filtered chemical oxygen demand time series; From the screened chemical oxygen demand time series, a target chemical oxygen demand time series that can reflect the long-term stable trend of data change is extracted; The target chemical oxygen demand (COD) time series is divided into multiple sub-COD time series; When the sub-evaluation parameters of each of the sub-chemical oxygen demand (COD) time series and the overall evaluation parameters of the target COD time series meet the third preset condition, the end time of the target COD time series is taken as the trend abnormality time. Based on whether the trend abnormality time exists within a preset third time period before the candidate judgment time, an alarm message is generated, wherein the preset third time period is greater than or equal to the preset second time interval, the sub-evaluation parameters are calculated from the median of the slopes of all paired data points in the sub-COD time series, and the overall evaluation parameters are calculated from the median of the slopes of all paired data points in the target COD time series.

[0005] In some technical solutions of this application, the generation of alarm information based on whether the aforementioned abnormal trend moment exists within a preset third time period before the candidate judgment moment includes: If the abnormal trend occurs within a preset third time period before the candidate judgment time, then based on the monitoring data corresponding to the time of the initial oxygen demand time series, the cause of the data meeting the third preset condition is determined, and Type I alarm information and Type II alarm information are generated for the candidate judgment time. The Type I alarm information and the Type II alarm information are pushed according to preset paths respectively; and / or, The step of extracting a target chemical oxygen demand (COD) time series that reflects the long-term stable trend of the data from the screened COD time series includes: Based on a preset filtering method, the target chemical oxygen demand (COD) time series is extracted from the filtered COD time series; and / or, The auxiliary hydrological and water quality parameter time series includes at least one of conductivity time series and liquid level time series.

[0006] In some technical solutions of this application, when at least one abnormal trend moment has been confirmed, the step of removing the abnormal operating condition data from the chemical oxygen demand (COD) time series of the preset second duration to obtain the filtered COD time series includes: From the chemical oxygen demand (COD) time series of the preset second duration, the abnormal operating condition data and the data affecting abnormal trends are removed to obtain the filtered COD time series, wherein the data affecting abnormal trends are data within a set time period before and / or after the time of the abnormal trend; and / or, The method further includes: determining the accuracy of each alarm, and adjusting the third preset condition according to the preset alarm confidence control interval, the sub-evaluation parameter values ​​corresponding to the occurrence of an accurate alarm, and the overall evaluation parameter.

[0007] In some technical solutions of this application, the aforementioned third pre-defined condition includes: Each of the sub-evaluation parameters satisfies its corresponding sub-evaluation condition, and the overall evaluation parameter satisfies the overall evaluation condition.

[0008] In some technical solutions of this application, the above-mentioned sub-evaluation conditions include sub-slope thresholds, and the overall evaluation conditions include an overall slope threshold; each of the sub-evaluation parameters satisfies its corresponding sub-evaluation condition, and the overall evaluation parameter satisfies the overall evaluation condition, specifically including: All of the sub-evaluation parameters are greater than the corresponding sub-slope thresholds, and the overall evaluation parameter is greater than the overall slope threshold.

[0009] In some technical solutions of this application, the reasons for determining the generation of data that meets the third preset condition based on monitoring data corresponding to the time of the initial chemical oxygen demand time series include: Based on any one or more of the monitoring data collected by the target sensor and the working status and location status of the monitoring device, determine whether the cause of the data that meets the third preset condition is an abnormality of the device itself. And / or, based on the data collected by the target sensor related to the temporal variation characteristics of the contact state between the monitoring device and the bottom of the water body and / or water quality parameters, determine whether the cause of the data that meets the third preset condition is an abnormal external environment.

[0010] In some technical solutions of this application, determining whether the cause of the data satisfying the third preset condition is an abnormality of the device itself, based on one or more of the data collected by the target sensor that are related to the working status and location status of the monitoring device, includes: The data currently collected by the monitoring device is compared with the corresponding data currently collected by the reference device to determine whether the sensor is abnormal. And / or, compare the current operating data of the preset component in the monitoring device with the pre-stored standard operating data to determine whether the preset component is abnormal; And / or, when the current attitude data of the monitoring device is greater than a preset attitude threshold, the monitoring device is determined to have an abnormal attitude tilt; and / or, The determination of whether the data satisfying the third preset condition is caused by an abnormal external environment, based on the data collected by the target sensor related to the contact state between the water quality monitoring device and the bottom of the water body and / or the time-series change characteristics of water quality parameters, includes: When the monitoring device is currently in a submerged state, and the second liquid level at its location, as obtained by the monitoring device, gradually decreases and the chemical oxygen demand gradually increases within a preset fourth time period, it is determined that the monitoring device has an abnormal bottom environment. And / or, if the temperature and conductivity values ​​collected by the monitoring device are both gradually decreasing during the preset fifth time period, it is determined that the monitoring device is experiencing an abnormal temperature environment; If the minimum or average dissolved oxygen content at the monitoring device is lower than the preset oxygen content threshold during a preset sixth time period, the water monitoring device is determined to be abnormally surrounded by plants.

[0011] Secondly, embodiments of this application provide a water quality monitoring system based on a quantum dot spectral sensor, the water quality monitoring system comprising: The acquisition module is used to obtain the initial chemical oxygen demand (COD) time series of the water body to be tested based on a quantum dot spectral sensor, and to select monitoring times that meet a first preset condition as candidate judgment times according to a preset first time interval. The first preset condition is that the COD at the monitoring time is greater than a first preset threshold, and the first preset threshold is obtained based on the COD within a preset first time period before the monitoring time. The filtering module is used to filter out a chemical oxygen demand time series of a preset second duration from the initial chemical oxygen demand time series according to a preset second time interval; The determination module is used to acquire at least one auxiliary hydrological and water quality parameter time series corresponding to the chemical oxygen demand time series of the preset second duration. When the first data point in the auxiliary hydrological and water quality parameter time series meets the second preset condition, the first data point is determined to be abnormal operating data. The second preset condition is that the first data point in the at least one auxiliary hydrological and water quality parameter time series is less than the corresponding second preset threshold, the preset first duration is less than the preset second duration, and the preset first time interval is less than the preset second time interval. The removal module is used to remove the abnormal operating data from the chemical oxygen demand time series of the preset second duration to obtain the filtered chemical oxygen demand time series. The extraction module is used to extract a target chemical oxygen demand time series that reflects the long-term stable trend of the data from the screened chemical oxygen demand time series. The segmentation module is used to divide the target chemical oxygen demand time series into multiple sub-chemical oxygen demand time series; An alarm module is used to determine the end time of the target chemical oxygen demand time series as an abnormal trend time when the sub-evaluation parameters of each of the sub-chemical oxygen demand time series and the overall evaluation parameters of the target chemical oxygen demand time series meet a third preset condition; and to issue an alarm based on whether the abnormal trend time exists within a preset third time period before the candidate judgment time, wherein the preset third time period is greater than or equal to the preset second time interval, the sub-evaluation parameters are calculated from the median of the slopes of all paired data points in the sub-chemical oxygen demand time series, and the overall evaluation parameters are calculated from the median of the slopes of all paired data points in the target chemical oxygen demand time series.

[0012] Thirdly, embodiments of this application provide an electronic device, a processor, a memory, and a bus. The memory stores machine instructions executed by the processor. When the electronic device is running, the processor communicates with the memory via the bus. When the machine instructions are executed by the processor, the steps of the water quality monitoring method based on the quantum dot spectral sensor described above are performed.

[0013] Fourthly, embodiments of this application provide a computer storage medium storing a computer program, which, when run by a processor, executes the steps of the water quality monitoring method based on a quantum dot spectral sensor described above.

[0014] The technical solutions provided by the embodiments of this application may include the following beneficial effects: The method of this application includes obtaining an initial chemical oxygen demand (COD) time series of the water body to be tested; selecting monitoring times that meet a first preset condition as candidate judgment times according to a preset first time interval, wherein the first preset condition is that the COD at the monitoring time is greater than a first preset threshold, and the first preset threshold is obtained based on the COD within a preset first time period before the monitoring time; selecting a COD time series of a preset second time period from the initial COD time series according to a preset second time interval; obtaining at least one auxiliary hydrological and water quality parameter time series corresponding to the COD time series of the preset second time period; when a first data point in the auxiliary hydrological and water quality parameter time series meets a second preset condition, the first data point is determined to be abnormal operating data, wherein the second preset condition is that the first data point in the at least one auxiliary hydrological and water quality parameter time series is less than the corresponding second preset threshold, the preset first time period is less than the preset second time period, and the preset first time interval is less than the preset second time interval; from the initial COD time series, selecting a COD time series of a preset second time period; selecting a COD time series of a preset second time period from the initial COD time series ... In a preset second-duration chemical oxygen demand (COD) time series, abnormal operating data are removed to obtain a filtered COD time series. From the filtered COD time series, a target COD time series reflecting the long-term stable trend of the data is extracted. The target COD time series is divided into multiple sub-COD time series. When the sub-evaluation parameters of each sub-COD time series and the overall evaluation parameters of the target COD time series meet a third preset condition, the end time of the target COD time series is taken as the trend abnormality moment. An alarm is issued based on whether the trend abnormality moment exists within a preset third duration before the candidate judgment moment, wherein the preset third duration is greater than or equal to the preset second time interval. The sub-evaluation parameters are calculated from the median of the slopes of all paired data points in the sub-COD time series, and the overall evaluation parameters are calculated from the median of the slopes of all paired data points in the target COD time series. This application can significantly improve the accuracy and reliability of water quality monitoring and effectively reduce the false alarm rate through the above method.

[0015] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A schematic flowchart of a water quality monitoring method based on a quantum dot spectral sensor provided in an embodiment of this application is shown. Figure 2 This illustration shows a schematic diagram of a water quality monitoring system based on a quantum dot spectral sensor provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the accompanying drawings in this application are for illustrative and descriptive purposes only and are not intended to limit the scope of protection of this application. Furthermore, it should be understood that the schematic drawings are not drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of this application. It should be understood that the operations in the flowcharts may not be implemented in sequence, and steps without logical contextual relationships may be reversed or implemented simultaneously. In addition, those skilled in the art, guided by the content of this application, may add one or more other operations to the flowcharts, or remove one or more operations from the flowcharts.

[0019] Furthermore, the described embodiments are merely some, not all, of the embodiments of this application. The components of the embodiments of this application described and illustrated herein can typically be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0020] It should be noted that the term "comprising" will be used in the embodiments of this application to indicate the presence of the features declared thereafter, but does not exclude the addition of other features.

[0021] Water quality monitoring is a crucial aspect of environmental protection and water resource management. Existing water quality monitoring systems typically collect and store water quality data, then directly compare this data with preset thresholds to determine if anomalies exist. When anomalies are detected, an alarm is generated and pushed to the user. This monitoring method is relatively crude in its data analysis process. Considering the complexity of the aquatic environment, the causes of these anomalies can be varied, including non-aquatic pollution. Furthermore, if the anomaly data is not properly processed, there is a risk of triggering false alarms, making it difficult to meet the requirements for high-precision and high-reliability pollution alarms. Therefore, this application provides a water quality monitoring method and system based on a quantum dot spectral sensor, which is described below through embodiments.

[0022] Figure 1 The diagram illustrates a flow chart of a water quality monitoring method based on a quantum dot spectral sensor provided in this application, wherein the method includes steps S101-S107; specifically: S101. Obtain the initial chemical oxygen demand (COD) time series of the water body to be tested based on a quantum dot spectral sensor, and select monitoring times that meet the first preset condition as candidate judgment times according to a preset first time interval. The first preset condition is that the COD at the monitoring time is greater than the first preset threshold. The first preset threshold is obtained based on the COD within a preset first time period before the monitoring time. The above-mentioned initial COD time series of the water body to be tested based on a quantum dot spectral sensor can obtain spectral information (reflection, transmission, scattering, etc.) through a quantum dot spectral sensor, and then obtain the initial COD time series based on the spectral information and a preset model.

[0023] S102. Select a chemical oxygen demand time series of a preset second duration from the initial chemical oxygen demand time series according to a preset second time interval; S103. Obtain at least one auxiliary hydrological and water quality parameter time series corresponding to the chemical oxygen demand (COD) time series with the preset second duration. When the first data point in the auxiliary hydrological and water quality parameter time series meets the second preset condition, the first data point is determined to be abnormal operating data. The second preset condition is that the first data point in the at least one auxiliary hydrological and water quality parameter time series is less than the corresponding second preset threshold, the preset first duration is less than the preset second duration, and the preset first time interval is less than the preset second time interval. The monitoring equipment includes other sensors in addition to the quantum dot spectral sensor, and the auxiliary hydrological and water quality parameters can be measured by other sensors and / or the quantum dot spectral sensor.

[0024] S104. Remove the abnormal operating condition data from the chemical oxygen demand time series with the preset second duration to obtain the filtered chemical oxygen demand time series; the abnormal operating condition data will cause relatively prominent outliers in the chemical oxygen demand time series, affecting the trend of the entire data.

[0025] S105. Extract the target chemical oxygen demand time series that can reflect the long-term stable trend of data from the screened chemical oxygen demand time series. S106. Divide the target chemical oxygen demand time series into multiple sub-chemical oxygen demand time series; S107. When the sub-evaluation parameters of each of the sub-chemical oxygen demand (COD) time series and the overall evaluation parameters of the target COD time series meet the third preset condition, the end time of the target COD time series is taken as the trend abnormality time. Based on whether the trend abnormality time exists within a preset third time period before the candidate judgment time, an alarm message is generated, wherein the preset third time period is greater than or equal to the preset second time interval, the sub-evaluation parameters are calculated from the median of the slopes of all paired data points in the sub-COD time series, and the overall evaluation parameters are calculated from the median of the slopes of all paired data points in the target COD time series.

[0026] This application significantly improves the accuracy and reliability of water quality monitoring and effectively reduces the false alarm rate through the above-described methods. Some embodiments of this application are described in detail below. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0027] This application provides a water quality monitoring method, which can be executed by an electronic device such as a terminal device or a server. The terminal device can be a user equipment (UE), mobile device, user terminal, terminal, cellular phone, cordless phone, personal digital assistant (PDA), handheld device, computing device, vehicle-mounted device, wearable device, etc. The method can be implemented by a processor calling computer-readable instructions stored in memory. Alternatively, the method can be executed by a server.

[0028] The water quality monitoring method is applied to a collection of water bodies, such as rivers, lakes, seas, reservoirs, and groundwater. The water body includes not only water but also solvents, suspended solids, bottom sediments, and aquatic organisms. The embodiments of this application do not limit the types of water bodies.

[0029] When monitoring water quality, water quality indicators may include Chemical Oxygen Demand (COD), turbidity, total phosphorus content, ammonia nitrogen content, permanganate index, total suspended solids, biological oxygen demand (BOD), total organic carbon (COD), sulfate content, chloride content, dissolved iron content, dissolved manganese content, dissolved copper content, dissolved zinc content, nitrate content, nitrite content, total nitrogen content, fluoride content, selenium content, total arsenic content, total mercury content, total cadmium content, chromium content, total lead content, total cyanide, volatile phenol content, coliform bacteria content, and sulfide content. In this application's embodiment, the monitoring of the water body is primarily aimed at determining whether the collected water quality data has reached a candidate judgment moment or a trend anomaly moment. Considering factors such as processing efficiency, this application has selected COD as a water quality indicator. That is, the data analyzed in the embodiments of this application is a chemical oxygen demand time series (for ease of description, the collected chemical oxygen demand time series is referred to as the initial chemical oxygen demand time series. Here, "initial" is only used to distinguish it from the subsequent chemical oxygen demand time series and is not limited).

[0030] In practical implementation, the initial chemical oxygen demand (COD) time series of water bodies can be measured using monitoring equipment, such as monitoring equipment equipped with quantum dot spectral sensors. The quantum dot spectral sensor can measure incident light (e.g., light transmitted, scattered, or fluoresced after passing through a predetermined area of ​​water sample) based on the physical and optical properties of nanocrystals to obtain the spectral information of the incident light. For example, the quantum dot spectral sensor may include a nanocrystal chip made of pre-defined nanocrystals, wherein the nanocrystal chip contains a certain arrangement of nanocrystals (e.g., a nanocrystal array), wherein the nanocrystals have light absorption or emission characteristics corresponding to the COD.

[0031] Light transmitted and / or scattered through water can be affected by substances in the water (e.g., suspended solids, pollutants, etc.), thereby obtaining spectral information on chemical oxygen demand (COD). Based on this spectral information and a pre-set model, a relevant water quality time series can be obtained. This application does not limit the working principle of the quantum dot spectral sensor.

[0032] Quantum dot spectrophotometers can be used to obtain chemical oxygen demand (COD) in real time. COD measurements obtained at multiple times from the same location can be compiled into a COD time series (initial COD time series). Compared to the method of sampling water quality and then conducting laboratory tests to measure COD, quantum dot spectrophotometers can achieve online, in-situ, high-frequency, and real-time measurement. For example, the measurement frequency can be increased from once a day to once every 3-60 minutes, preferably 5-30 minutes, particularly preferably 8-20 minutes, and most preferably 10-15 minutes, which is much higher than traditional testing methods. Therefore, COD time series can be obtained at a higher frequency.

[0033] The chemical oxygen demand (COD) of water can be continuously monitored using a quantum dot spectrometer at a preset location to obtain a COD time series (initial COD time series). In the example, the COD time series can be represented as: {(t1,x1,1),(t2,x1,2),(t3,x1,3),…,(tc,x1,c),…}, where c is any positive integer, tc represents the c-th time, and x1,c represents the chemical oxygen demand measured at the c-th time.

[0034] After obtaining the initial chemical oxygen demand (COD) time series of the water body to be tested, this embodiment monitors the water body based on the initial COD time series. During the monitoring process, to avoid false alarms, this embodiment monitors based on both candidate judgment times and trend anomaly times. Candidate judgment times represent moments when pollution or other anomalies may occur, while trend anomaly times indicate moments when trend anomalies may occur. Trend anomalies refer to the baseline of the data (or signal) (i.e., the baseline of the signal in a static or unexcited state) experiencing slow, non-periodic fluctuations over time, rather than remaining at zero or a stable level. By jointly judging the candidate judgment times and trend anomaly times, the cause of the alarm is confirmed, and corresponding alarm information is generated, improving the reliability of the monitoring results.

[0035] For candidate judgment times: monitoring times that meet the first preset condition are selected as candidate judgment times according to the preset first time interval, wherein the first preset condition is that the chemical oxygen demand at the monitoring time is greater than the first preset threshold, and the first preset threshold is obtained based on the chemical oxygen demand within the preset first time period before the monitoring time.

[0036] The preset first time interval here refers to the interval period for scanning the initial chemical oxygen demand (COD) time series to find potential anomalies. This interval is usually greater than or equal to the sampling frequency of data acquisition, aiming to reduce computational load and achieve efficient preliminary screening. For example, if the sampling frequency is 10 minutes / scan, the preset first time interval can be set to 10 minutes or 1 hour, etc. The candidate judgment time refers to the specific monitoring time when the COD measurement value meets the first preset condition during scanning according to the preset first time interval. This time indicates that an abnormal event requiring further attention may have occurred. The first preset condition refers to the criterion used to determine whether a monitoring time can become a candidate judgment time. Specifically, it is defined as: the COD measurement value at this monitoring time is greater than the first preset threshold. To improve the accuracy of candidate judgment times, the first preset threshold is a dynamically changing judgment threshold. Its value is not fixed but calculated based on COD data within a specific historical time period (i.e., the preset first duration) prior to the current monitoring time. This design allows the threshold to adapt to the baseline level and normal fluctuation range of water quality. The preset first duration refers to the length of the historical data time window used to calculate the first preset threshold. For example, it can be 24 hours, 48 ​​hours, or 72 hours. This time window traces back from the current monitoring time.

[0037] In practice, the first preset time interval is set to, for example, 3 minutes, and the first preset duration is set to, for example, 24 hours. The initial chemical oxygen demand (COD) time series is iterated every 3 minutes. For each scanned monitoring moment, all COD data points from the preceding 24 hours (first duration) are extracted. Based on all COD data points from the preceding 24 hours (first duration), a first preset threshold is obtained through statistical calculation. This can be achieved using one or a combination of the following methods: mean-standard deviation method, quantile method, median method, moving window extreme value method, etc. To avoid the influence of large fluctuations on the calculation results, points exceeding the set threshold can be removed before calculation. The measured COD value at the monitoring moment is compared with the first preset threshold. If the measured COD value is greater than the first preset threshold, the first preset condition is met, and it is marked as a candidate judgment moment, with its timestamp recorded. Otherwise, the moment is ignored, and the scan continues to the next interval.

[0038] For identifying anomaly moments: The second time-series chemical oxygen demand (COD) data is extracted, and invalid abnormal operating conditions are identified and removed using auxiliary hydrological and water quality parameters. Then, the target COD time series for the long-term trend is extracted from the purified data and divided into multiple subsequences for refined evaluation. An anomaly moment is confirmed when the evaluation parameters of all subsequences and the overall series meet the third preset condition.

[0039] The basis for extracting the second-length chemical oxygen demand (COD) time series is a preset second time interval. Here, the second time interval refers to the trigger period for extracting a data segment from the initial COD time series for in-depth analysis. This second time interval is typically longer than the preset first time interval, and its purpose is to extract data representing a longer duration of anomalies, making it easier to distinguish from pollution anomalies. For example, it can be set to 12 hours or 24 hours. The preset second duration refers to the length of the historical data time window required for in-depth trend analysis. This duration is significantly longer than the preset first duration, aiming to provide sufficiently long data to identify slow, continuous data trend anomalies. For example, it can be set to 5 days, 7 days, or 10 days. The second-length COD time series refers to a continuous data subset of the initial COD time series with a length equal to the preset second duration. This series is the core analytical object for subsequent data cleaning, trend extraction, and anomaly confirmation. Optionally, the preset second duration is 60 to 240 times the preset first duration, for example, it can be 60, 70, 90, 120, 200, 240, etc., and the preset second time interval is 140 to 500 times the preset first time interval, for example, it can be 140, 150, 170, 200, 300, 400, 450, 500, etc., which will not be elaborated here. Of course, the preset second duration and preset first duration, as well as the preset second time interval and preset first time interval, can also be calculated based on historical data and a trained model.

[0040] Due to the acquisition environment of quantum dot spectral sensors, abnormal operating conditions may occur in the chemical oxygen demand (COD) time series. To identify and remove these abnormal data in advance and improve the efficiency of identifying trend anomalies, data cleaning is necessary. The basis for data cleaning is the auxiliary hydrological and water quality parameter time series corresponding to the second-time-length COD time series. This correspondence is based on the synchronicity of the acquisition time. Specifically, it is a set of other physical or chemical parameters used to determine the operating status of the monitoring equipment, acquired synchronously with the second-time-length COD time series and arranged in chronological order. Its core purpose is not to directly assess water quality, but rather to diagnose the equipment's operating condition at the time of data acquisition. Typical auxiliary hydrological and water quality parameters include, but are not limited to: conductivity: a measure of the electrical conductivity of water. Pure water has extremely low conductivity, and air has nearly zero conductivity. Liquid level: the depth of the monitoring equipment probe relative to the water surface.

[0041] The chemical oxygen demand (COD) time series for a second time period is monitored based on at least one auxiliary hydrological and water quality parameter time series. When any auxiliary hydrological and water quality parameter of a first data point in the second time period COD time series meets a second preset condition, the COD of that first data point is marked as abnormal operating data. The second preset condition refers to the logical criterion used to determine whether a first data point corresponds to abnormal operating data. Specifically, it is defined as: the value of a first data point in the time series of at least one auxiliary hydrological and water quality parameter is less than the corresponding second preset threshold. The second preset threshold is a critical value set for each auxiliary hydrological and water quality parameter to determine whether the equipment is in an out-of-water state. This threshold is a fixed value pre-set based on physical characteristics and experimental data. Abnormal operating data refers to data points collected when the monitoring equipment is in an abnormal underwater operating state. The most significant abnormal operating condition is an out-of-water state, i.e., the sensor's sensitive element is not in sufficient contact with the measured water body.

[0042] In one optional implementation, the intuitive indicator for determining whether the quantum dot spectrometer is in an out-of-water state is the liquid level at which the quantum dot spectrometer is located (for distinction, this liquid level is referred to as the first liquid level, which is the liquid level at which the quantum dot spectrometer is located when collecting each initial chemical oxygen demand (COD) in the initial COD time series). The first liquid level is compared with a preset first liquid level threshold. If the first liquid level is greater than or equal to the first liquid level threshold, the quantum dot spectrometer is in a submerged state; if the first liquid level is less than the first liquid level threshold, the quantum dot spectrometer is in an out-of-water state. Here, the first liquid level threshold is the lowest liquid level that completely covers the monitoring of the quantum dot spectrometer.

[0043] On the other hand, when the quantum dot spectrometer is submerged in water, the ions in the water maintain its conductivity at a certain level. However, when the quantum dot spectrometer is removed from the water, the surrounding environment becomes air, and the conductivity of air is much lower than that of water, causing a significant drop in the conductivity value collected by the sensor. Based on this characteristic, whether the quantum dot spectrometer is in an off-water state can also be reflected by the conductivity value collected by the quantum dot spectrometer.

[0044] Furthermore, the purpose of this embodiment is to remove water-free data from the second-term chemical oxygen demand (COD) time series. It is necessary to determine whether the quantum dot spectrometer was in a water-free state when collecting data for each initial COD value in the initial COD time series. Therefore, the time for monitoring conductivity needs to be limited; that is, only the monitoring conductivity within the corresponding time period of the initial COD time series needs to be determined. Here, the corresponding time period represents the monitoring conductivity collected within the same time period as the initial COD time series. For example, if the time period corresponding to the second-term COD time series is t11~t12, then the monitoring conductivity is also the conductivity within t11~t12. After determining the monitoring conductivity, it can be determined whether the quantum dot spectrometer is in a water-free state by comparing the monitoring conductivity with a preset conductivity threshold. For example, if the monitoring conductivity is less than the conductivity threshold, the quantum dot spectrometer is in a water-free state; otherwise, the quantum dot spectrometer is in a water-immersed state. Correspondingly, the water-free period of the quantum dot spectral sensor can be determined, and the chemical oxygen demand collected during the water-free period can be removed to obtain the screened chemical oxygen demand time series (all the screened chemical oxygen demand time series constitute the screened chemical oxygen demand time series).

[0045] Because the screened chemical oxygen demand (COD) time series contains interference noise and short-term fluctuations, it can mask the long-term trend of water quality data, making it impossible to accurately identify true trend anomalies. This application embodiment requires removing the aforementioned interferences after obtaining the screened COD time series. Specifically, this application embodiment adopts a trend extraction method: extracting the target COD time series that reflects the long-term stable trend of the data from the screened COD time series.

[0046] In practical implementation, preset filtering methods can be used, such as moving average filtering, exponentially weighted moving average filtering, and Kalman filtering. Considering the loss of data details and the effectiveness of noise filtering, this embodiment adopts Savitzky-Golay filtering: by performing polynomial fitting on the data within a preset sliding time window, it can effectively suppress random noise while preserving the overall trend of the data to the greatest extent (such as the abnormal characteristics of a slow upward / downward trend), avoiding the true trend anomaly being masked by "oversmoothing". For example, the sliding window can be set to 5, 6, or 7, etc.

[0047] To avoid missed or false alarms, this embodiment of the application, after obtaining the target chemical oxygen demand (COD) time series, does not directly process the target COD time series, but instead splits it into multiple sub-COD time series. This embodiment of the application uses both the target COD time series and each sub-COD time series as the object of analysis, ensuring the accuracy of the monitoring results.

[0048] When decomposing a target chemical oxygen demand (COD) time series, the time series can be decomposed using the same time interval or different time intervals. For example, if the preset time period is 9 days, a uniform 3-day time interval can be used to decompose the target COD time series into sub-COD time series of 1-3 days, 4-6 days, and 7-9 days. Alternatively, 4-day and 5-day time intervals can be used to decompose the target COD time series into sub-COD time series of 1-4 days and 5-9 days.

[0049] When analyzing both the target chemical oxygen demand (COD) time series and its sub-COD time series, this embodiment sets corresponding evaluation conditions for each. When the sub-evaluation parameters of each sub-COD time series and the overall evaluation parameters of the target COD time series meet a third preset condition, the end time of the target COD time series is designated as a trend anomaly moment. Specifically, the third preset condition includes that each sub-evaluation parameter meets its corresponding sub-evaluation condition and the overall evaluation parameter meets the overall evaluation condition. By determining whether the overall evaluation parameter meets its corresponding evaluation condition and whether each sub-evaluation parameter meets its corresponding evaluation condition, it is determined whether the end time of the target COD time series is a trend anomaly moment.

[0050] In an optional implementation, the sub-evaluation parameter is calculated from the median of the slopes of all paired data points in the sub-chemical oxygen demand (COD) time series, and the overall evaluation parameter is calculated from the median of the slopes of all paired data points in the target COD time series. Here, a paired data point refers to a one-to-one data pair consisting of a COD value and its corresponding acquisition time. In specific implementations, the Tylsen slope can be used. A paired data point can be represented as (tc, Vc), where tc represents the c-th time point and Vc represents the COD measured at the c-th time point. Alternatively, the sub / overall evaluation parameter can be calculated from the truncated mean, weighted median, etc., of the slopes of all paired data points in the sub / target COD time series; wherein, the truncated mean is obtained by sorting the slopes of all paired data points by size, removing a certain percentage of extreme values ​​at both ends (e.g., removing the 10% maximum slope and the 10% minimum slope), and then calculating the arithmetic mean of the remaining slopes. The weighted median assigns different weights to the slopes of all pairs of data points (e.g., weighted by the time distance of the data points: the slopes of recent data have higher weights, and the slopes of older data have lower weights), and then sorts them by weight cumulatively. The slope value that has a cumulative weight of 50% of the total weight is found and used as the final statistic.

[0051] For example, iterate through all paired data points within the sub-chemical oxygen demand (COD) time series or the target COD time series, and calculate the slope by combining each pair: (V j - V i ) / (t j - t i ), where t j >t i If i and j are any positive integers, then take the median of all these slopes to obtain the sub-Tyerson slope (sub-evaluation parameter) or the overall Tyerson slope (overall evaluation parameter).

[0052] In practical implementation, the sub-evaluation conditions for sub-evaluation parameters can be limited by sub-slope thresholds, and the overall evaluation conditions for overall evaluation parameters can be limited by overall slope thresholds. Whether the preset evaluation conditions are met can be determined by comparing the (sub- or overall) evaluation parameters with the (sub- or overall) slope thresholds. Specifically, if all sub-evaluation parameters are greater than their corresponding sub-slope thresholds and the overall evaluation parameter is greater than the overall slope threshold, it is determined whether the end time of the target chemical oxygen demand time series is an abnormal trend moment. For example, the target chemical oxygen demand (COD) time series is from October 1st to October 7th, 2023, divided into three sub-COD time series: October 1st-3rd, October 3rd-5th, and October 5th-7th; the Tylsen slopes of each sub-COD time series are 0.85, 0.92, and 0.88 (mg / L) / day, respectively; the overall Tylsen slope is 0.88 (mg / L) / day; the preset thresholds are: sub-slope threshold of 0.8 and overall slope threshold of 0.8; since all evaluation parameters are greater than the corresponding thresholds, the third preset condition is met, and the end time of the target series, 24:00 on October 7th, 2023, is determined as the time of trend anomaly.

[0053] After obtaining the candidate judgment time and the trend anomaly time, an alarm message is generated based on whether the trend anomaly time exists within a preset third time period before the candidate judgment time. It is important to note that the third time period satisfies the following condition: the preset third time period is greater than or equal to the preset second time interval. This is to ensure a sufficient time window to accommodate potential trend anomalies, matching the analysis scheduling cycle and avoiding missed judgments due to an excessively short time window. It is typically set to 1-2 times the preset second time interval; for example, if the preset second time interval is 12 hours, the preset third time period can be set to 24 hours.

[0054] Based on the candidate judgment time, a preset third time period is traced back to determine the verification time window. Within the verification time window, it is checked whether there are any recorded trend anomaly times. If at least one trend anomaly time exists within the time window, one type of alarm information is generated; otherwise, another type of alarm information is generated. These alarms are used to alert users to water pollution and equipment or environmental anomalies, improving the accuracy and effectiveness of monitoring data and avoiding false alarms about water pollution caused by equipment or environmental anomalies. In specific implementations, the alarms can be in the form of text messages, sound, or flashing lights. This application embodiment does not limit the specific generation method or form of the alarm information.

[0055] In an optional implementation, in order to improve the efficiency of anomaly handling when an alarm is triggered, this embodiment of the application further determines the alarm type of the abnormal data (device anomaly, external environment anomaly, and anomaly caused by pollution), and then generates alarm information based on the abnormal data and alarm type.

[0056] In actual monitoring work, the main causes of abnormal data trends include two situations: abnormal operating conditions of the monitoring equipment and abnormalities in the environment in which the monitoring equipment is located (generally excluding water quality abnormalities caused by pollution). In order to determine whether the abnormal data trend is caused by the monitoring equipment itself, the environment in which the monitoring equipment is located, or a combination of both (i.e., the alarm types in this application embodiment can be divided into equipment alarms, environmental alarms, mixed alarms, and pollution alarms), this application embodiment needs to detect the status of the monitoring equipment and the environment in which it is located. That is, to acquire the monitoring data corresponding to the time of the initial chemical oxygen demand time series, determine the cause of the data that meets the third preset condition, and generate Class I alarm information (indicating that the alarm is caused by the equipment itself or the external environment) and Class II alarm information (for example, the alarm is a pollution alarm caused by factors other than the equipment itself or the external environment). The Class I alarm information and the Class II alarm information are pushed according to preset paths respectively. For example, Class I alarm information is pushed to the equipment maintenance team, while Class II alarm information is pushed to the technical analysis team or users for manual judgment (such as sampling and re-analysis). Different levels of alarm information use different push frequencies and notification methods.

[0057] Specifically, the status of the monitoring equipment includes its operational status and location status. The operational status is mainly reflected in the equipment data during operation, while the location status is mainly reflected in the equipment's attitude and position data. By detecting the equipment data, it's possible to determine if there are any equipment alarm types. By detecting the attitude and position data, it's possible to determine if there are any environmental alarm types. If both equipment and environmental alarm types exist, it's classified as a mixed alarm type. When an alarm is triggered, the alarm type is identified, leading to more accurate alarm information and improved efficiency in handling anomalies. The monitoring process includes determining whether the data meeting a third preset condition is caused by an equipment malfunction, based on one or more of the monitoring data collected by the target sensor related to the equipment's operational status and location status; and / or, determining whether the data meeting the third preset condition is caused by an external environmental anomaly, based on data collected by the target sensor related to the contact status between the monitoring equipment and the bottom of the water body and / or the temporal variation characteristics of water quality parameters.

[0058] In an optional implementation, during actual monitoring, equipment alarm types and environmental alarm types can be further subdivided into more sub-alarm types. To enable users to handle anomalies more directly, this embodiment further distinguishes between equipment alarm types and environmental alarm types: equipment alarm types are divided into multiple first sub-alarms, and environmental alarm types are divided into multiple second sub-alarms. After determining the alarm type corresponding to the target chemical oxygen demand time series using the above method, it is also necessary to further determine the target sub-alarms under each alarm type. Then, alarm information is generated based on the abnormal data and the target sub-alarms, allowing the alarm to be pushed to the corresponding recipients based on the specific target sub-alarm, thus improving the timeliness of abnormal data processing.

[0059] The first sub-alarm here includes sensor malfunctions, preset component malfunctions, attitude tilt malfunctions, and bottom environment malfunctions of the monitoring equipment; the second sub-alarm includes temperature environment malfunctions and plant environment malfunctions. The process of identifying the target sub-alarm is to determine whether sensor malfunctions, preset component malfunctions, attitude tilt malfunctions, and bottom environment malfunctions exist, as well as whether temperature environment malfunctions and plant environment malfunctions exist.

[0060] Considering the occurrence mechanism, data characteristics, and interference logic of each sub-alarm, in order to ensure the accuracy of the determination result, this application embodiment sets appropriate determination methods for each first sub-alarm and second sub-alarm.

[0061] Regarding sensor malfunctions: If a sensor malfunctions, the collected data will be continuously distorted, and the fault cannot be detected solely by the data collected by the malfunctioning sensor. Therefore, to avoid this situation, this application embodiment sets up a reference device to compare the current device data of the sensor with the corresponding data collected by the reference device. When the current device data differs from the reference device data or the difference is significant (greater than a preset first reference threshold), the sensor malfunction is determined. The reference device here can be the same sensor, a device that detects the working status of the sensor, upstream or downstream devices, or even the monitoring device itself, utilizing data from a time other than the current moment, such as a time before a set duration, under normal conditions.

[0062] Regarding preset component anomalies: Preset components here refer to devices other than sensors. When a preset component (excluding sensors) malfunctions, its normal operating characteristics can generally be quantified through data. A more efficient approach is to compare the current device data with the data of a normal device under normal operating conditions. If the operating data of the preset component on the current device differs significantly from the operating data of the same component on a normal device (greater than a preset second reference threshold), the preset component is determined to be malfunctioning.

[0063] For example, the brushes in monitoring equipment (usually used to clean sensor probes to prevent data distortion caused by impurities) can be quantified by collecting current operating data (such as brush speed, operating current, cleaning frequency, and electrical signals corresponding to operating noise): Under normal operating conditions, the brush cleaning speed is stable at 500 rpm, the operating current is maintained at 0.3A±0.05A, and the cleaning time is fixed at 10 seconds; these data ranges accumulated through long-term stable operation can be used as a second reference threshold for judgment (such as a speed below 450 rpm or above 550 rpm, or a current exceeding the range of 0.25-0.35A, which is considered abnormal).

[0064] Regarding abnormal posture tilt: The core of abnormal posture tilt is that the physical position of the device deviates from the preset normal posture. However, this deviation is not only an abnormal value of the tilt angle, but may also be accompanied by information such as the tilt direction (e.g., tilting left / right / forward). To ensure the integrity of the information, this application embodiment collects the posture data of the device through image recognition or by acquiring the posture data of the device through position sensors in different directions. The posture data of the monitored device is compared with a preset posture threshold: when the posture data of the monitored device is not within the posture threshold range, an abnormal posture tilt is determined. For example, if the tilt threshold (posture threshold) is 15°, and the detected current posture data is 20°, then the monitored device has an abnormal posture tilt.

[0065] Regarding bottom environment anomalies: Here, bottom environment anomalies indicate that the bottom of the monitoring equipment (or sensor probe or sampling component) is in direct contact with the riverbed / pond bottom / sediment (bottom mud) of the monitored water body, deviating from the normal monitoring position. For example, organic matter in the bottom mud can adhere to the sensor probe, causing the detected chemical oxygen demand (COD) value to be significantly higher. Detection of this type of anomaly requires the connection between liquid level and COD data. Specifically: when the monitoring equipment is currently submerged in water, and the second liquid level acquired by the monitoring equipment at its location gradually decreases over a preset fourth time period while the COD gradually increases, the monitoring equipment is determined to have a bottom environment anomaly. All of the above conditions are indispensable. For example, a decrease in liquid level alone could be a normal hydrological change (such as during the dry season). If the probe is still at the normal monitoring depth, the COD data will not be abnormal, and there is no need to determine whether it is a bottom environment anomaly. An increase in COD alone could be due to true pollution caused by industrial wastewater discharge, or the probe being covered by impurities (not bottom mud), which is unrelated to bottom environment anomalies.

[0066] For abnormal temperature environments: For example, in low-temperature environments, ice forms on the surface of monitoring equipment (sensor probes, sampling ports) or in the surrounding water, blocking normal contact between the equipment and the water (e.g., ice covering the probe prevents light signal transmission and electrodes from exchanging ions with the water), leading to abnormal data distortion. The judgment process is as follows: if the temperature and conductivity values ​​collected by the monitoring equipment are gradually decreasing after a preset fifth time period, it is determined that the monitoring equipment is experiencing an abnormal temperature environment. Ice coverage requires a low-temperature environment (temperature gradually dropping to 0℃ or below, meeting the freezing conditions), and ion confinement: as the ice layer thickens, the liquid water channels available for free ion movement in the water become fewer and narrower. The migration path of ions in the ice-water mixture becomes abnormally tortuous and difficult, resulting in a significant and continuous decrease in overall conductivity. The synchronous and gradual decrease of both characteristics perfectly matches the mechanism of ice coverage triggering low temperature, physical blockage, and inhibition of biological activity, ruling out other anomalies (such as pollution only causing a decrease in DO, pH possibly increasing or decreasing, but temperature not necessarily decreasing), thus accurately determining it as an abnormal temperature environment.

[0067] Regarding abnormal plant environment conditions: This refers to an abnormal state caused by excessive growth of aquatic plants (such as aquatic grasses and algae) around the monitoring point, which entangles the sensor surface, causing the monitoring equipment to tilt, cover the detection elements, etc., resulting in distortion of data such as chemical oxygen demand (COD). The judgment process is as follows: if the minimum or average dissolved oxygen content at the monitoring equipment location is lower than a preset oxygen content threshold at a preset sixth time interval, the monitoring equipment is determined to have an abnormal plant environment. Aquatic plants continuously consume oxygen in the water through respiration, and when they entangle the device probe, they hinder the normal contact between DO and the probe, inhibiting oxygen exchange in the water, resulting in continuous loss of DO that cannot be effectively replenished. The preset oxygen content threshold is set based on the reasonable range of DO in normal water (without entanglement). When DO is lower than this threshold, it indicates that the oxygen consumption in the water is far beyond the natural level, and this change directly matches the mechanism of continuous oxygen consumption and physical obstruction caused by aquatic plant entanglement. At the same time, it can exclude other anomalies (such as pollution, which may cause a decrease in DO, but is often accompanied by an increase in chemical oxygen demand, etc., while plant entanglement only specifically affects DO), thus accurately determining it as an abnormal plant environment.

[0068] In an optional implementation, when cleaning the chemical oxygen demand (COD) time series for the second time period, to ensure the data accurately reflects the overall trend, after removing abnormal operating condition data from the COD time series for the second time period, this embodiment also removes data affecting trend anomalies. These trend anomaly-affected data are data within a set time period before and / or after the trend anomaly. This time period is set based on the following considerations: Forward influence period: typically set to T1 hour (e.g., 12-24 hours) before the trend anomaly; Backward influence period: typically set to T2 hour (e.g., 6-12 hours) after the trend anomaly. The forward period covers the stage where the abnormal trend begins to form, and the backward period includes the stage of anomaly recovery or continued impact. The specific duration can be optimized and adjusted based on historical anomaly data analysis.

[0069] The specific data identification operations are as follows: Identify all abnormal operating condition data points and all abnormal trend moments. For each abnormal trend moment, determine its forward and backward influence time periods, and mark all data points within that time period as data affected by the abnormal trend. From the preset second-length chemical oxygen demand (COD) time series: remove all data points marked as abnormal operating conditions; remove all data points marked as affected by abnormal trend trends; retain the remaining data points to form the filtered COD time series. It is important to emphasize that when abnormal operating condition data and data affected by abnormal trend trends overlap in time, a deduplication operation is performed to ensure that each abnormal data point is removed only once.

[0070] In an optional implementation, in order to ensure the adaptability of alarms to different environments, the third preset condition in this application embodiment is not fixed, but is in an updated state: the accuracy of each alarm is determined, and the third preset condition is adjusted according to the preset alarm confidence control interval, the sub-evaluation parameter values ​​corresponding to the occurrence of an accurate alarm, and the overall evaluation parameters.

[0071] Accuracy is determined for each generated alarm, i.e., whether it is truly due to a device malfunction and / or an external environmental malfunction. The calculation is based on the verified actual situation and the alarm information generated based on the above judgment process. Alarms matching the actual situation are recorded as 1, and those not matching are recorded as 0. The first preset condition is adjusted according to a preset alarm confidence control interval. For example, if there are 5 alarms caused by device malfunctions and / or external environmental malfunctions, and the above method determines that 4 alarms within the same time period are caused by device malfunctions and / or external environmental malfunctions, then the accuracy rate is 80%. If the alarm confidence control interval is 70%~79% accuracy, then no adjustment to the third preset condition is needed. If the alarm confidence control interval is 82%~90% accuracy, then adjustment is needed. The adjustment data is obtained based on the sub-evaluation parameter values ​​and the overall evaluation parameter value corresponding to the 4 accurately determined alarms.

[0072] Record the key parameters when an accurate alarm occurs: the sub-assessment parameter values ​​of each sub-chemical oxygen demand (COD) time series, the overall assessment parameter value of the target COD time series, and calculate the average or median of the above parameters as the adjustment benchmark. Adjust the relevant thresholds in the third preset condition based on the statistical results: sub-slope threshold adjustment: based on the average or median of each sub-assessment parameter value; overall slope threshold adjustment: based on the average or median of the overall assessment parameter value.

[0073] This application significantly reduces false alarms and missed alarms, ensuring the reliability of water quality alarms; it also clearly identifies alarm types, enabling users to handle them precisely and reducing unnecessary manpower investment. It improves the accuracy and effectiveness of monitoring data, promotes technological advancements and integrated innovation in spectroscopic online water quality monitoring, and facilitates the widespread application of online water quality monitoring methods in pipeline network inspection and intelligent operation.

[0074] Figure 2 This invention illustrates a schematic diagram of a water quality monitoring system based on a quantum dot spectral sensor, according to an embodiment of this application. The system includes: The acquisition module is used to obtain the initial chemical oxygen demand (COD) time series of the water body to be tested based on a quantum dot spectral sensor, and to select monitoring times that meet a first preset condition as candidate judgment times according to a preset first time interval. The first preset condition is that the COD at the monitoring time is greater than a first preset threshold, and the first preset threshold is obtained based on the COD within a preset first time period before the monitoring time. The filtering module is used to filter out a chemical oxygen demand time series of a preset second duration from the initial chemical oxygen demand time series according to a preset second time interval; The determination module is used to acquire at least one auxiliary hydrological and water quality parameter time series corresponding to the chemical oxygen demand time series of the preset second duration. When the first data point in the auxiliary hydrological and water quality parameter time series meets the second preset condition, the first data point is determined to be abnormal operating data. The second preset condition is that the first data point in the at least one auxiliary hydrological and water quality parameter time series is less than the corresponding second preset threshold, the preset first duration is less than the preset second duration, and the preset first time interval is less than the preset second time interval. The removal module is used to remove the abnormal operating data from the chemical oxygen demand time series of the preset second duration to obtain the filtered chemical oxygen demand time series. The extraction module is used to extract a target chemical oxygen demand time series that reflects the long-term stable trend of the data from the screened chemical oxygen demand time series. The segmentation module is used to divide the target chemical oxygen demand time series into multiple sub-chemical oxygen demand time series; An alarm module is used to determine the end time of the target chemical oxygen demand time series as an abnormal trend time when the sub-evaluation parameters of each of the sub-chemical oxygen demand time series and the overall evaluation parameters of the target chemical oxygen demand time series meet a third preset condition; and to generate alarm information based on whether the abnormal trend time exists within a preset third time period before the candidate judgment time, wherein the preset third time period is greater than or equal to the preset second time interval, the sub-evaluation parameters are calculated from the median of the slopes of all paired data points in the sub-chemical oxygen demand time series, and the overall evaluation parameters are calculated from the median of the slopes of all paired data points in the target chemical oxygen demand time series.

[0075] The alarm module is used to generate alarm information based on whether the trend anomaly occurs within a preset third time period before the candidate judgment time, including: If the abnormal trend occurs within a preset third time period before the candidate judgment time, the cause of the data that meets the third preset condition is determined based on the monitoring data corresponding to the time of the initial oxygen demand time series. Type I alarm information and Type II alarm information are generated for the candidate judgment time, and the Type I alarm information and the Type II alarm information are pushed according to preset paths respectively.

[0076] The extraction module is used to extract a target chemical oxygen demand (COD) time series that reflects the long-term stable trend of the data from the screened COD time series, including: Based on a preset filtering method, the target chemical oxygen demand (COD) time series is extracted from the filtered COD time series; and / or, The auxiliary hydrological and water quality parameter time series includes at least one of conductivity time series and liquid level time series.

[0077] Remove module, for When at least one abnormal trend moment has been identified, the step of removing the abnormal operating condition data from the preset second-duration chemical oxygen demand (COD) time series to obtain the filtered COD time series includes: From the chemical oxygen demand time series with the preset second duration, the abnormal operating condition data and the data affecting the abnormal trend are removed to obtain the filtered chemical oxygen demand time series, wherein the data affecting the abnormal trend is the data within a set time period before and / or after the time of the abnormal trend.

[0078] The device also includes an adjustment module for determining the accuracy of each alarm and adjusting the third preset condition based on the preset alarm confidence control range, the sub-evaluation parameter values ​​corresponding to the occurrence of an accurate alarm, and the overall evaluation parameters.

[0079] The alarm module has a third preset condition for alarms, which includes: each of the sub-evaluation parameters meets the corresponding sub-evaluation condition and the overall evaluation parameter meets the overall evaluation condition.

[0080] The alarm module includes sub-slope thresholds as the sub-evaluation conditions during an alarm, and an overall slope threshold as the overall evaluation condition. Specifically, the sub-evaluation parameters must satisfy their respective sub-evaluation conditions, and the overall evaluation parameter must satisfy the overall evaluation condition. This includes situations where each sub-evaluation parameter is greater than its corresponding sub-slope threshold, and the overall evaluation parameter is greater than the overall slope threshold.

[0081] The alarm module, when issuing an alarm, determines the reasons for the data that meets the third preset condition based on the monitoring data corresponding to the time of the initial chemical oxygen demand time series, including: Based on any one or more of the monitoring data collected by the target sensor and the working status and location status of the monitoring device, determine whether the cause of the data that meets the third preset condition is an abnormality of the device itself. And / or, based on the data collected by the target sensor related to the contact state between the monitoring device and the bottom of the water body and / or the time-series change characteristics of water quality parameters, determine whether the cause of the data that meets the third preset condition is an abnormal external environment.

[0082] The alarm module, when issuing an alarm, determines whether the cause of the data meeting the third preset condition is an abnormality of the device itself, based on one or more of the data collected by the target sensor related to the working status and location status of the monitoring device. The data currently collected by the monitoring device is compared with the corresponding data collected by the reference device to determine whether the sensor is abnormal. And / or, compare the current operating data of the preset component in the monitoring device with the pre-stored standard operating data to determine whether the preset component is abnormal; And / or, when the current attitude data of the monitoring device is greater than a preset attitude threshold, the monitoring device is determined to have an abnormal attitude tilt; and / or, The determination of whether the data satisfying the third preset condition is caused by an abnormal external environment, based on the data collected by the target sensor related to the contact state between the water quality monitoring device and the bottom of the water body and / or the time-series change characteristics of water quality parameters, includes: When the monitoring device is currently in a submerged state, and the second liquid level at its location, as obtained by the monitoring device, gradually decreases and the chemical oxygen demand gradually increases within a preset fourth time period, it is determined that the monitoring device has an abnormal bottom environment. And / or, if the temperature and conductivity values ​​collected by the monitoring device are both gradually decreasing during the preset fifth time period, it is determined that the monitoring device is experiencing an abnormal temperature environment; If the minimum or average dissolved oxygen content at the monitoring device is lower than the preset oxygen content threshold during a preset sixth time period, the water monitoring device is determined to be in an abnormal plant environment.

[0083] like Figure 3 As shown, this application provides an electronic device for executing the water quality monitoring method based on a quantum dot spectral sensor as described in this application. The device includes a memory, a processor, a bus, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the water quality monitoring method based on a quantum dot spectral sensor.

[0084] Specifically, the aforementioned memory and processor can be general-purpose memory and processor, without any specific limitations. When the processor runs the computer program stored in the memory, it can execute the aforementioned water quality monitoring method based on quantum dot spectral sensors.

[0085] Corresponding to the water quality monitoring method based on quantum dot spectral sensors in this application, this application embodiment also provides a computer storage medium storing a computer program, which is executed by a processor to perform the steps of the above-described water quality monitoring method based on quantum dot spectral sensors.

[0086] Specifically, the storage medium can be a general-purpose storage medium, such as a portable disk or hard disk. When the computer program on the storage medium is run, it can execute the aforementioned water quality monitoring method based on a quantum dot spectral sensor.

[0087] In the embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. The system embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and there may be other division methods in actual implementation. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some communication interface. The indirect coupling or communication connection between systems or units may be electrical, mechanical, or other forms.

[0088] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0089] In addition, the functional units in the embodiments provided in this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0090] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0091] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In addition, the terms "first", "second", "third", etc. are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0092] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The protection scope of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the technical scope disclosed in this application; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application. All should be covered within the protection scope of this application. Therefore, the protection scope of this application should be determined by the protection scope of the claims.

Claims

1. A water quality monitoring method based on a quantum dot spectral sensor, characterized in that, The method includes: The initial chemical oxygen demand (COD) time series of the water body to be tested is obtained based on a quantum dot spectral sensor. Monitoring times that meet the first preset condition are selected as candidate judgment times according to a preset first time interval. The first preset condition is that the COD at the monitoring time is greater than the first preset threshold. The first preset threshold is obtained based on the COD within a preset first time period before the monitoring time. A chemical oxygen demand (COD) time series of a preset second duration is selected from the initial COD time series according to a preset second time interval; Obtain at least one auxiliary hydrological and water quality parameter time series corresponding to the chemical oxygen demand time series with the preset second duration. When the first data point in the auxiliary hydrological and water quality parameter time series meets the second preset condition, the first data point is determined to be abnormal operating data. The second preset condition is that the first data point in the at least one auxiliary hydrological and water quality parameter time series is less than the corresponding second preset threshold, the preset first duration is less than the preset second duration, and the preset first time interval is less than the preset second time interval. Remove the abnormal operating data from the chemical oxygen demand time series with the preset second duration to obtain the filtered chemical oxygen demand time series; From the screened chemical oxygen demand time series, a target chemical oxygen demand time series that can reflect the long-term stable trend of data change is extracted; The target chemical oxygen demand (COD) time series is divided into multiple sub-COD time series; When the sub-evaluation parameters of each of the sub-chemical oxygen demand (COD) time series and the overall evaluation parameters of the target COD time series meet the third preset condition, the end time of the target COD time series is taken as the trend abnormality time. Based on whether the trend abnormality time exists within a preset third duration before the candidate judgment time, an alarm message is generated, wherein the preset third duration is greater than or equal to the preset second time interval. The sub-evaluation parameters are calculated from the median of the slopes of all paired data points in the sub-COD time series, and the overall evaluation parameters are calculated from the median of the slopes of all paired data points in the target COD time series.

2. The method according to claim 1, characterized in that, The step of generating an alarm message based on whether there is an abnormal trend within a preset third time period before the candidate judgment time includes: If the abnormal trend occurs within a preset third time period before the candidate judgment time, then based on the monitoring data corresponding to the time of the initial oxygen demand time series, the cause of the data meeting the third preset condition is determined, and Type I alarm information and Type II alarm information are generated for the candidate judgment time. The Type I alarm information and the Type II alarm information are pushed according to preset paths respectively; and / or, The step of extracting a target chemical oxygen demand (COD) time series that reflects the long-term stable trend of the data from the screened COD time series includes: Based on a preset filtering method, the target chemical oxygen demand (COD) time series is extracted from the filtered COD time series; and / or, The auxiliary hydrological and water quality parameter time series includes at least one of conductivity time series and liquid level time series.

3. The method according to claim 1, characterized in that, When at least one abnormal trend moment has been identified, the step of removing the abnormal operating condition data from the preset second-duration chemical oxygen demand (COD) time series to obtain the filtered COD time series includes: From the chemical oxygen demand (COD) time series of the preset second duration, the abnormal operating condition data and the data affecting abnormal trends are removed to obtain the filtered COD time series, wherein the data affecting abnormal trends are data within a set time period before and / or after the time of the abnormal trend; and / or, The method further includes: determining the accuracy of each alarm, and adjusting the third preset condition according to the preset alarm confidence control interval, the sub-evaluation parameter values ​​corresponding to the occurrence of an accurate alarm, and the overall evaluation parameter.

4. The method according to any one of claims 1 to 3, characterized in that, The third preset condition includes: Each of the sub-evaluation parameters satisfies its corresponding sub-evaluation condition, and the overall evaluation parameter satisfies the overall evaluation condition.

5. The method according to claim 4, characterized in that, The sub-evaluation conditions include sub-slope thresholds, and the overall evaluation conditions include an overall slope threshold; each of the sub-evaluation parameters satisfies its corresponding sub-evaluation condition, and the overall evaluation parameter satisfies the overall evaluation condition, specifically including: All of the sub-evaluation parameters are greater than the corresponding sub-slope thresholds, and the overall evaluation parameter is greater than the overall slope threshold.

6. The method according to claim 2, characterized in that, The reasons for determining the generation of data that meets the third preset condition based on monitoring data corresponding to the time of the initial oxygen demand time series include: Based on any one or more of the monitoring data collected by the target sensor and the working status and location status of the monitoring device, determine whether the cause of the data that meets the third preset condition is an abnormality of the device itself. And / or, based on the data collected by the target sensor related to the temporal variation characteristics of the contact state between the monitoring device and the bottom of the water body and / or water quality parameters, determine whether the cause of the data that meets the third preset condition is an abnormal external environment.

7. The method according to claim 6, characterized in that, Determining whether the cause of the data satisfying the third preset condition is a device malfunction, based on data collected by the target sensor that is related to any one or more of the working status and location status of the monitoring device, includes: The data currently collected by the monitoring device is compared with the corresponding data collected by the reference device to determine whether the sensor is abnormal. And / or, compare the current operating data of the preset component in the monitoring device with the pre-stored standard operating data to determine whether the preset component is abnormal; And / or, when the current attitude data of the monitoring device is greater than a preset attitude threshold, the monitoring device is determined to have an abnormal attitude tilt; and / or, The determination of whether the data satisfying the third preset condition is caused by an abnormal external environment, based on the data collected by the target sensor related to the contact state between the water quality monitoring device and the bottom of the water body and / or the time-series change characteristics of water quality parameters, includes: When the monitoring device is currently in a submerged state, and the second liquid level at its location, as obtained by the monitoring device, gradually decreases and the chemical oxygen demand gradually increases within a preset fourth time period, it is determined that the monitoring device has an abnormal bottom environment. And / or, if the temperature and conductivity values ​​collected by the monitoring device are both gradually decreasing during the preset fifth time period, it is determined that the monitoring device is experiencing an abnormal temperature environment; If the minimum or average dissolved oxygen content at the monitoring device is lower than the preset oxygen content threshold during a preset sixth time period, the water monitoring device is determined to be in an abnormal plant environment.

8. A water quality monitoring system based on a quantum dot spectral sensor, characterized in that, The water quality monitoring system includes: The acquisition module is used to obtain the initial chemical oxygen demand (COD) time series of the water body to be tested based on a quantum dot spectral sensor, and to select monitoring times that meet a first preset condition as candidate judgment times according to a preset first time interval. The first preset condition is that the COD at the monitoring time is greater than a first preset threshold, and the first preset threshold is obtained based on the COD within a preset first time period before the monitoring time. The filtering module is used to filter out a chemical oxygen demand time series of a preset second duration from the initial chemical oxygen demand time series according to a preset second time interval; The determination module is used to acquire at least one auxiliary hydrological and water quality parameter time series corresponding to the chemical oxygen demand time series of the preset second duration. When the first data point in the auxiliary hydrological and water quality parameter time series meets the second preset condition, the first data point is determined to be abnormal operating data. The second preset condition is that the first data point in the at least one auxiliary hydrological and water quality parameter time series is less than the corresponding second preset threshold, the preset first duration is less than the preset second duration, and the preset first time interval is less than the preset second time interval. The removal module is used to remove the abnormal operating data from the chemical oxygen demand time series of the preset second duration to obtain the filtered chemical oxygen demand time series. The extraction module is used to extract a target chemical oxygen demand time series that reflects the long-term stable trend of the data from the screened chemical oxygen demand time series. The segmentation module is used to divide the target chemical oxygen demand time series into multiple sub-chemical oxygen demand time series; An alarm module is used to determine the end time of the target chemical oxygen demand time series as an abnormal trend time when the sub-evaluation parameters of each of the sub-chemical oxygen demand time series and the overall evaluation parameters of the target chemical oxygen demand time series meet a third preset condition; and to generate alarm information based on whether the abnormal trend time exists within a preset third time period before the candidate judgment time, wherein the preset third time period is greater than or equal to the preset second time interval, the sub-evaluation parameters are calculated from the median of the slopes of all paired data points in the sub-chemical oxygen demand time series, and the overall evaluation parameters are calculated from the median of the slopes of all paired data points in the target chemical oxygen demand time series.

9. An electronic device, characterized in that, include: The device includes a processor, a memory, and a bus. The memory stores machine instructions that the processor executes. When the electronic device is running, the processor communicates with the memory via the bus. When the machine instructions are executed by the processor, they perform the steps of the water quality monitoring method based on a quantum dot spectral sensor as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that, The computer storage medium stores a computer program that, when executed by a processor, performs the steps of the water quality monitoring method based on a quantum dot spectral sensor as described in any one of claims 1 to 7.

Images