Smart water station sampling process information transmission method and system based on data perception

Abstract

The present invention belongs to the technical field of information collection and transmission, and discloses a method and system for transmitting information of a sampling process of a smart water station based on data perception; the method comprises: deploying smart water stations and edge devices; the smart water stations collect real-time water quality data and send it to corresponding edge devices; the smart water stations collect historical water quality data and send it to corresponding edge devices, the edge devices analyze the real-time water quality data and historical water quality data, calculate water quality change data and data thresholds; the edge devices evaluate the water quality status according to the water quality change data, data thresholds and real-time water quality data; the edge devices optimize the sampling strategy according to the water quality status, and send it to the corresponding smart water stations, and the smart water stations perform sampling according to the optimized sampling strategy; the present invention can improve the timeliness and accuracy of water quality monitoring, ensure water resource management and public health; and improve the efficiency of information transmission, thereby improving the overall reliability and adaptability of the system.

Description

Technical Field

[0001] The present invention relates to the technical field of information collection and transmission, and more specifically, to a method and system for transmitting sampling process information of a smart water station based on data perception. Background Art

[0002] As global water shortage and water pollution become increasingly serious, water quality monitoring and management have become increasingly important. Traditional water quality monitoring relies on manual sampling and laboratory analysis. Although it can provide accurate data, it has the disadvantages of long response time, limited coverage and high human resource consumption. With the development of technologies such as the Internet of Things (IoT), big data and artificial intelligence, smart water stations have emerged and become a core component of modern water quality monitoring. Smart water stations integrate sensors, GPS positioning and communication technologies to achieve real-time monitoring and data transmission of water quality parameters, and can automatically collect multiple water quality indicators such as pH value, dissolved oxygen, and heavy metal concentration. Decision makers can understand water quality conditions in a timely manner, respond quickly to sudden pollution incidents, and optimize water resource management.

[0003] However, although smart water stations have shown significant advantages in water quality monitoring, they still face challenges such as data transmission efficiency and data accuracy; how to effectively transmit sampling process information and ensure the real-time and reliability of data is a key issue in the design of smart water station systems; a Chinese patent with publication number CN110736508A discloses a smart city water environment monitoring system; it includes a monitoring station and several drones, the drone includes a chassis and a propeller mechanism that drives the chassis to move, the monitoring station is equipped with a water quality detector, the chassis is equipped with correspondingly connected wireless communication modules, positioning modules and mobile power supplies, the drone transmits information to the monitoring station through the wireless communication module, the chassis is connected to a camera, the outside of the chassis is provided with a temperature monitor, and the bottom of the chassis is connected to a real-time monitoring device and several water sampling mechanisms, which can reduce equipment consumption, can flexibly sample water bodies at different locations in the water area, facilitate detection, and monitor data more accurately.

[0004] Although the above technology can realize the mobility of monitoring stations, there are obvious deficiencies in the mobility logic. Specifically, the lack of clear basis and standards for mobile decision-making causes the smart water station to move at inappropriate times or places, which in turn affects the monitoring effect. This inflexible design makes the system unable to respond quickly to water quality changes or emergencies. At the same time, the lack of dynamic optimization of sampling indicators and sampling frequency leads to insufficient data representativeness, poor real-time data monitoring, and inefficient operation of the monitoring system, which affects the overall data quality. In addition, the transmission efficiency and reliability of sampling process information are not solved.

[0005] In view of this, the present invention proposes a smart water station sampling process information transmission method and system based on data perception to solve the above problems. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art and achieve the above-mentioned purpose, the present invention provides the following technical solution: a data-aware smart water station sampling process information transmission method, applied to edge devices, comprising:

[0007] Receive real-time water quality data collected by smart water stations;

[0008] Receive historical water quality data collected by smart water stations, analyze real-time and historical water quality data, and calculate water quality change data and data thresholds;

[0009] Assess water quality status based on water quality change data, data thresholds and real-time water quality data;

[0010] According to the water quality conditions, the sampling strategy is optimized and sent to the smart water station.

[0011] Furthermore, the sampling strategy includes sampling indicators, sampling frequency and sampling points; the sampling indicators are the data types in the real-time water quality data collected by the smart water station, the sampling frequency is the number of times the smart water station collects real-time water quality data per unit time, and the sampling point is the location of the smart water station; edge devices are deployed for each smart water station;

[0012] The method for optimizing the sampling points comprises:

[0013] Mark the smart water station corresponding to the edge device that evaluates the water quality status as abnormal as an abnormal water station; obtain the water station coordinates of all smart water stations; obtain the central coordinates of the water source and mark them as the source coordinates; calculate the Euclidean distance between each smart water station and the source coordinates according to the source coordinates and the water station coordinates, and mark them as the source distance; sort the source distances from small to large to generate a first sequence table; according to the positive order of the first sequence table, set an increasing digital number for each smart water station in turn, that is, the range of the digital number is [1, m], and m is the number of smart water stations; among them, if there are multiple smart water stations with the same source distance, the corresponding digital numbers are the same;

[0014] The smart water stations with digital numbers smaller than the corresponding digital numbers of the abnormal water stations are marked as upstream water stations; based on the water station coordinates of the abnormal water station and each upstream water station, the Euclidean distance between each upstream water station and the abnormal water station is calculated and marked as the water station distance; all water station distances are sorted from small to large to generate a second sequence table; the upstream water station corresponding to the first water station distance in the second sequence table is marked as an adjacent water station; based on the water station distance between the adjacent water station and the abnormal water station, the abnormal water station and the adjacent water stations are gradually moved to find the pollution source, and the abnormal water station is moved to the pollution source.

[0015] Furthermore, the step of finding the pollution source includes:

[0016] Step 1: Preset the initial search interval [a0, b0], where a0 = 0 and b0 is the distance between the adjacent water station and the abnormal water station; set the segmentation ratio h;

[0017] Step 2: According to the segmentation ratio h, the initial search interval [a0, b0] is divided into two new points ζ1 and ζ2, ζ1 = a0 + (1-h)(b0-a0), ζ2 = a0 + h (b0-a0);

[0018] Step 3: Move the abnormal water station in the direction of the adjacent water station by ζ1, and move the adjacent water station in the direction of the abnormal water station by ζ2; after the movement is completed, according to the water quality conditions, the abnormal water station and the adjacent water station respectively collect corresponding data in the real-time water quality data; according to the data collected by the abnormal water station and the adjacent water station, respectively calculate the corresponding change rate; mark the change rate corresponding to the abnormal water station as the third change rate, and mark the change rate corresponding to the adjacent water station as the fourth change rate;

[0019] Step 4: If the third change rate is greater than or equal to the fourth change rate, the minimum value of the updated search interval is the minimum value of the search interval before the update, the maximum value is the point corresponding to the fourth change rate, and the abnormal water station is moved in the opposite direction of the adjacent water station by ζ1; if the third change rate is less than the fourth change rate, the minimum value of the updated search interval is the point corresponding to the third change rate, the maximum value is the maximum value of the search interval before the update, and the adjacent water station is moved in the opposite direction of the abnormal water station by ζ2;

[0020] Step 5: Preset the width threshold, subtract the minimum value from the maximum value in the updated search interval to obtain the interval width; compare the interval width with the width threshold; if the interval width is greater than or equal to the width threshold, then re-divide the updated search interval according to the segmentation ratio h to obtain two new points ζ1 and ζ2, and return to step 3; if the interval width is less than the width threshold, obtain the water station coordinates corresponding to the current abnormal water station and the adjacent water station respectively, and mark them as real-time coordinates; take the mean coordinate of the two real-time coordinates as the pollution source.

[0021] Furthermore, the method for optimizing the sampling frequency includes:

[0022] The real-time water quality data and water quality change data are used as analysis data, and the analysis data is input into the trained frequency prediction model to predict the corresponding sampling frequency and mark it as the predicted frequency; the sampling frequency of the smart water station is replaced with the corresponding predicted frequency, and the real-time water quality data is sampled;

[0023] The training process of the frequency prediction model includes:

[0024] A corresponding sampling frequency is set for group A of analysis data in advance, where A is an integer greater than 1; the analysis data and the corresponding sampling frequency are converted into a corresponding set of feature vectors; each set of feature vectors is used as the input of a structural evaluation model, and the structural evaluation model uses a set of predicted sampling frequencies corresponding to each set of analysis data as output, and uses the actual sampling frequency corresponding to each set of analysis data as a prediction target, where the actual sampling frequency is the sampling frequency corresponding to the analysis data collected in advance; the prediction results are evaluated for model accuracy using the mean absolute percentage error (MAPE), and when the calculated MAPE is less than the preset MAPE, the training of the structural evaluation model is completed; a structural evaluation model that predicts the sampling frequency based on the analysis data is generated; the structural evaluation model is a deep belief network model.

[0025] Furthermore, the real-time water quality data is water quality data collected in real time; the water quality data includes physical data and chemical data; the physical data includes pH value, turbidity and conductivity; the chemical data includes dissolved oxygen concentration and chemical oxygen demand;

[0026] The method for calculating water quality change data comprises:

[0027] The water quality data in the historical water quality data that are adjacent to the collection time of the real-time water quality data are marked as adjacent water quality data; the corresponding data in the adjacent water quality data are subtracted from each data in the real-time water quality data to obtain the change amount of each data in the water quality data; the change amount of each data is multiplied by the preset sampling frequency to obtain the change rate of each data in the water quality data, and used as the water quality change data.

[0028] Furthermore, the method for calculating the data threshold includes:

[0029] The data threshold includes a numerical range value and a variation range value; wherein the numerical range value and the variation range value correspond one-to-one to each data in the water quality data;

[0030] The calculation method of numerical range value includes:

[0031] Count the number of groups of water quality data in the historical water quality data and add 1 as the number of data; take the real-time water quality data and the historical water quality data as the total water quality data; add the data of the same type in the total water quality data in turn, and divide it by the number of data to obtain the data mean corresponding to each data in the water quality data; subtract the corresponding data mean from each data in the total water quality data and then square it to obtain the data difference corresponding to each data in the total water quality data; add the data difference of each data in the total water quality data in turn, and then divide it by the number of data to obtain the data variance corresponding to each data in the water quality data; take the square root of each data variance to obtain the data standard deviation corresponding to each data in the water quality data; add three times the corresponding data standard deviation to the data mean corresponding to each data in the water quality data as the maximum value of the corresponding numerical range, and subtract three times the corresponding data standard deviation from the numerical mean corresponding to each data in the water quality data as the minimum value of the corresponding numerical range.

[0032] Furthermore, the calculation method of the variation range value includes:

[0033] Subtract 1 from the number of data as the number of changes; collect all calculated water quality change data and mark them as total change data; add the same type of change rates in the total change data to an analysis set respectively, and the analysis set corresponds to the change rate in the water quality change data one by one; sort the change rates in each analysis set from small to large to generate a corresponding sorting table; multiply the number of changes by one quarter to obtain the first number; multiply the number of changes by three quarters to obtain the second number; wherein, if the first number or the second number is not an integer, round up; in each sorting table, mark the change rate in the first number as the first change rate, and mark the change rate in the second number as the second change rate; subtract the first change rate from the second change rate corresponding to each analysis set to obtain the change rate interval; multiply the change rate interval of each analysis set by 1.5 to obtain the data interval; subtract the data interval from the first change rate of each analysis set as the minimum value of the corresponding change range value; add the data interval to the second change rate of each analysis set as the maximum value of the corresponding change range value.

[0034] Furthermore, the water quality condition includes normal water quality and abnormal water quality; the abnormal water quality includes abnormal pH value, excessive dissolved oxygen concentration, excessive turbidity, excessive conductivity and excessive chemical oxygen demand;

[0035] The method for optimizing the sampling index comprises:

[0036] If the water quality condition is an abnormal pH value, the redox potential, total nitrogen content, total phosphorus content and heavy metal content are collected, that is, the real-time water quality data collected by the smart water station also includes redox potential, total nitrogen content, total phosphorus content and heavy metal content; the redox potential is the ratio of oxides and reductants in the water body, the total nitrogen content is the content of all nitrogen in the water body, the total phosphorus content is the total amount of all phosphorus in the water, and the heavy metal content is the concentration of heavy metals in the water body;

[0037] If the water quality condition is that the dissolved oxygen concentration exceeds the standard, the biochemical oxygen demand, ammonia nitrogen concentration, total nitrogen content and total phosphorus content are collected; that is, the real-time water quality data collected by the smart water station also includes biochemical oxygen demand, ammonia nitrogen concentration, total nitrogen content and total phosphorus content; biochemical oxygen demand is the amount of oxygen required when organic matter in the water body is decomposed by microorganisms, and ammonia nitrogen concentration is the concentration of nitrogen in the form of ammonia in the water body;

[0038] If the water quality condition is that the turbidity exceeds the standard, the total suspended solids content and the number of cyanobacteria will be collected; that is, the real-time water quality data collected by the smart water station also includes the total suspended solids content and the number of cyanobacteria; the total suspended solids content is the content of solid particles suspended in the water body, and the number of cyanobacteria is the concentration of cyanobacteria in the water body;

[0039] If the water quality condition is that the conductivity exceeds the standard, the total dissolved solid content and heavy metal content will be collected; that is, the real-time water quality data collected by the smart water station also includes the total dissolved solid content and heavy metal content; the total dissolved solid content is the total amount of dissolved solid matter in the water body;

[0040] If the water quality condition is that the chemical oxygen demand exceeds the standard, the biochemical oxygen demand, organic pollutant content and ammonia nitrogen concentration will be collected; that is, the real-time water quality data collected by the smart water station also includes biochemical oxygen demand, organic pollutant content and ammonia nitrogen concentration; indicating that the water body contains more organic pollutants; the organic pollutant content is the content of volatile organic compounds in the water body.

[0041] Furthermore, the method for assessing water quality comprises:

[0042] Compare each change rate in the water quality change data with the corresponding change range value in the data threshold; compare each data in the real-time water quality data with the corresponding numerical range value in the data threshold; if BF i ≤BH i ≤BF i ′∩SZ i′ ≤SJ i′ ≤SZ i ″, the water quality is normal; among them, BH i is the i-th change rate in the water quality change data, SJ i′ is the i′th data in the real-time water quality data, BF iis the minimum value of the change range corresponding to the i-th change rate, BF i ′ is the maximum value of the change range corresponding to the i-th change rate, SZ i′ is the minimum value of the numerical range corresponding to the i′th data, SZ i ″ is the maximum value of the numerical range corresponding to the i′th data, i∈[1,5], i′∈[1,5], ∩ represents and;

[0043] If BF1>BH1∪BH1>BF1′∪SZ1>SJ1∪SJ1>SZ1′, the pH value is abnormal;

[0044] If BF2>BH2∪BH2>BF2′∪SZ2>SJ2∪SJ2>SZ2′, the dissolved oxygen concentration exceeds the standard;

[0045] If BF3>BH3∪BH3>BF3′∪SZ3>SJ3∪SJ3>SZ3′, the turbidity exceeds the standard;

[0046] If BF4>BH4∪BH4>BF4′∪SZ4>SJ4∪SJ4>SZ′4, the conductivity exceeds the standard;

[0047] If BF5>BH5∪BH5>BF5′∪SZ5>SJ5∪SJ5>SZ5′, the chemical oxygen demand exceeds the standard;

[0048] Among them, i=1 means that the first change rate is the change rate corresponding to the pH value, i′=1 means that the first data is the pH value, ∪ means or; i=2 means that the second change rate is the change rate corresponding to the dissolved oxygen concentration, i′=2 means that the second data is the dissolved oxygen concentration; i=3 means that the third change rate is the change rate corresponding to the turbidity, i′=3 means that the third data is turbidity; i=4 means that the fourth change rate is the change rate corresponding to the conductivity, i′=4 means that the fourth data is conductivity; i=5 means that the fifth change rate is the change rate corresponding to the chemical oxygen demand, i′=5 means that the fifth data is the chemical oxygen demand.

[0049] A data-aware smart water station sampling process information transmission system implements any of the data-aware smart water station sampling process information transmission methods described above and is applied to edge devices, including:

[0050] Data collection module, used to receive real-time water quality data collected by smart water stations;

[0051] The data analysis module is used to receive the historical water quality data collected by the smart water station, analyze the real-time water quality data and historical water quality data, and calculate the water quality change data and data threshold;

[0052] A water quality assessment module is used to assess water quality conditions based on water quality change data, data thresholds and real-time water quality data;

[0053] The strategy optimization module is used to optimize the sampling strategy according to the water quality conditions and send it to the smart water station.

[0054] The technical effects and advantages of the data-aware smart water station sampling process information transmission method and system of the present invention are as follows:

[0055] By collecting and analyzing water quality data in real time, the water quality status can be evaluated in real time, and the sampling strategy can be dynamically optimized according to the water quality status. The timeliness and accuracy of water quality monitoring can be improved, and the quality of data monitoring can be improved, so as to effectively respond to water pollution incidents and ensure water resource management and public health. In addition, the deployment of edge devices can transmit information in a timely manner, reduce data transmission delays, improve information transmission efficiency, enhance data security, and improve the overall reliability and adaptability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 This is a schematic diagram of a data-aware smart water station sampling process information transmission system according to Embodiment 1 of the present invention;

[0057] Figure 2 This is a flow chart of the pollution source finding method of Example 1 of the present invention;

[0058] Figure 3 This is a flow chart of the method for transmitting sampling process information of a smart water station based on data perception according to Example 2 of the present invention. DETAILED DESCRIPTION

[0059] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0060] Example 1

[0061] See also Figure 1 As shown, the data-aware smart water station sampling process information transmission system described in this embodiment includes an equipment deployment module, a data acquisition module, a data analysis module, a water quality assessment module, and a strategy optimization module; each module is connected by wired and / or wireless means to realize data transmission between modules;

[0062] Device deployment module, used to deploy smart water stations and edge devices.

[0063] Smart water stations are deployed according to preset sampling distances, which are pre-set by technical personnel in the field based on the actual conditions of the water body. Edge devices are deployed at each smart water station, which are edge computing devices such as Raspberry Pi or customized embedded systems. Smart water stations are water quality monitoring and management systems that integrate modern information technology, sensor technology, and data analysis capabilities.

[0064] It should be noted that the reason why edge devices are deployed in each smart water station is real-time response: edge devices can process data instantly, reducing the delay in data transmission to the central server, and can quickly respond to changes in water quality or abnormal conditions; bandwidth saving: edge devices can directly process and calculate data, and only need to upload processed results or key data, which significantly reduces the demand for network bandwidth; data security: data is processed and stored on edge devices, reducing the risk of data leakage during transmission and enhancing data security; high availability: edge devices can continue to operate when the network is unstable or disconnected, ensuring that data collection and analysis are not affected; strong adaptability: edge devices can perform customized analysis and processing according to the local environment and water characteristics, improving the adaptability and accuracy of monitoring.

[0065] The data acquisition module is used by smart water stations to collect real-time water quality data and send it to the corresponding edge devices.

[0066] Real-time water quality data refers to water quality data collected in real time; water quality data includes physical data and chemical data; physical data includes pH value, turbidity and conductivity; chemical data includes dissolved oxygen concentration and chemical oxygen demand; smart water stations include pH sensors, turbidity sensors, conductivity sensors, dissolved oxygen sensors and electrochemical oxidation sensors for collecting water quality data.

[0067] The pH value is an indicator to measure the acidity and alkalinity of water. Different aquatic organisms have different adaptability to the acidity and alkalinity of water. Too high or too low a pH value will cause the death or reduction of aquatic organisms. The pH value affects the solubility and reactivity of chemical substances in water. For example, some heavy metals are more soluble in acidic environments and are toxic to organisms. Therefore, too high or too low a pH value indicates abnormal water quality.

[0068] Turbidity is an indicator of water transparency and reflects the content of suspended matter in the water. High turbidity usually indicates that the water contains a large amount of suspended matter, indicating that the water quality is polluted, affecting the use and safety of water. Turbid water will also hinder the penetration of sunlight, thereby affecting the photosynthesis of aquatic plants and causing ecological imbalance. In addition, water with high turbidity requires more chemicals and time during the treatment process, increasing treatment costs.

[0069] Conductivity is an indicator to measure the concentration of electrolytes in water. It reflects the concentration of dissolved ions (such as salts, minerals, etc.) in water and is used to assess the degree of mineralization of water. The higher the conductivity, the greater the concentration of dissolved ions in the water, indicating that the mineral content in the water is richer or the water is more seriously polluted, and vice versa.

[0070] The dissolved oxygen concentration is the amount of oxygen dissolved in the water in the form of oxygen; dissolved oxygen is a key factor in the survival and reproduction of aquatic organisms (such as fish, shrimp, algae, etc.). The concentration of dissolved oxygen in the water directly affects the health and survival of aquatic organisms; the dissolved oxygen concentration also reflects the self-purification ability of the water. The higher the dissolved oxygen concentration, the healthier the water, and vice versa.

[0071] Chemical oxygen demand is the amount of oxygen required for the oxidation of organic matter in water under the action of strong oxidants. Chemical oxygen demand is an important indicator for measuring the concentration of organic pollutants in water. The higher the chemical oxygen demand, the higher the organic content in the water body and the worse the water quality, and vice versa.

[0072] The data analysis module is used by smart water stations to collect historical water quality data and send it to the corresponding edge devices. The edge devices analyze the real-time water quality data and historical water quality data, and calculate the water quality change data and data thresholds.

[0073] Historical water quality data refers to water quality data collected historically.

[0074] Methods for calculating water quality change data include:

[0075] The water quality data in the historical water quality data that are adjacent to the collection time of the real-time water quality data are marked as adjacent water quality data; the corresponding data in the adjacent water quality data are subtracted from each data in the real-time water quality data to obtain the change amount of each data in the water quality data; the change amount of each data is multiplied by the preset sampling frequency to obtain the change rate of each data in the water quality data, and use it as the water quality change data; the sampling frequency is pre-set by technical personnel in this field according to the actual situation of the water body.

[0076] Methods for calculating data thresholds include:

[0077] The data threshold includes a numerical range value and a variation range value; wherein the numerical range value and the variation range value correspond one-to-one to each data in the water quality data.

[0078] The calculation method of numerical range value includes:

[0079] Count the number of groups of water quality data in the historical water quality data and add 1 as the number of data; take the real-time water quality data and the historical water quality data as the total water quality data; add the data of the same type in the total water quality data in turn, and divide it by the number of data to obtain the data mean corresponding to each data in the water quality data; subtract the corresponding data mean from each data in the total water quality data and then square it to obtain the data difference corresponding to each data in the total water quality data; add the data difference of each data in the total water quality data in turn, and then divide it by the number of data to obtain the data variance corresponding to each data in the water quality data; take the square root of each data variance to obtain the data standard deviation corresponding to each data in the water quality data; add three times the corresponding data standard deviation to the data mean corresponding to each data in the water quality data as the maximum value of the corresponding numerical range, and subtract three times the corresponding data standard deviation from the numerical mean corresponding to each data in the water quality data as the minimum value of the corresponding numerical range.

[0080] The calculation method of the variation range value includes:

[0081] Subtract 1 from the number of data to obtain the number of changes; collect all calculated water quality change data and mark them as total change data; add the same type of change rates in the total change data to an analysis set respectively, and the analysis set corresponds to the change rate in the water quality change data one by one; sort the change rates in each analysis set from small to large to generate a corresponding sorting table; multiply the number of changes by one quarter to obtain the first number; multiply the number of changes by three quarters to obtain the second number; if the first number or the second number is not an integer, round it up; in each sorting table, mark the change rate in the first number as the first change rate, and the change rate in the second number as the second change rate. Denote it as the second change rate; subtract the first change rate from the second change rate corresponding to each analysis set to obtain the change rate interval; multiply the change rate interval of each analysis set by 1.5 to obtain the data interval; subtract the data interval from the first change rate of each analysis set to obtain the minimum value of the corresponding change range value; add the data interval to the second change rate of each analysis set to obtain the maximum value of the corresponding change range value; wherein, since the analysis sets correspond one-to-one to the change rates in the water quality change data, the change rates in the water quality change data correspond one-to-one to the data in the water quality data, and the change range values ​​correspond one-to-one to the data in the water quality data, the analysis sets correspond one-to-one to the change range values.

[0082] The water quality assessment module is used by edge devices to assess water quality conditions based on water quality change data, data thresholds, and real-time water quality data.

[0083] Water quality conditions include normal water quality and abnormal water quality; abnormal water quality includes abnormal pH value, excessive dissolved oxygen concentration, excessive turbidity, excessive conductivity and excessive chemical oxygen demand.

[0084] Methods for assessing water quality conditions include:

[0085] Each change rate in the water quality change data is compared with the corresponding change range value in the data threshold; each data in the real-time water quality data is compared with the corresponding numerical range value in the data threshold.

[0086] If BF i ≤BH i ≤BF i ′∩SZ i′ ≤SJ i′ ≤SZ i ″, the water quality is normal; among them, BH i is the i-th change rate in the water quality change data, SJ i′ is the i′th data in the real-time water quality data, BF i is the minimum value of the change range corresponding to the i-th change rate, BF i ′ is the maximum value of the change range corresponding to the i-th change rate, SZ i′ is the minimum value of the numerical range corresponding to the i′th data, SZ i ″ is the maximum value of the numerical range corresponding to the i′th data, i∈[1,5], i′∈[1,5], ∩ represents and.

[0087] If BF1>BH1∪BH1>BF1′∪SZ1>SJ1∪SJ1>SZ1′, the pH value is abnormal; wherein, i=1 indicates that the first change rate is the change rate corresponding to the pH value, i′=1 indicates that the first data is the pH value, and ∪ indicates or.

[0088] If BF2>BH2∪BH2>BF2′∪SZ2>SJ2∪SJ2>SZ2′, the dissolved oxygen concentration exceeds the standard; wherein, i=2 indicates that the second change rate is the change rate corresponding to the dissolved oxygen concentration, and i′=2 indicates that the second data is the dissolved oxygen concentration.

[0089] If BF3>BH3∪BH3>BF3′∪SZ3>SJ3∪SJ3>SZ3′, the turbidity exceeds the standard; among them, i=3 means that the third change rate is the change rate corresponding to the turbidity, and i′=3 means that the third data is turbidity.

[0090] If BF4>BH4∪BH4>BF4′∪SZ4>SJ4∪SJ4>SZ′4, the conductivity exceeds the standard; wherein, i=4 indicates that the fourth change rate is the change rate corresponding to the conductivity, and i′=4 indicates that the fourth data is the conductivity.

[0091] If BF5>BH5∪BH5>BF5′∪SZ5>SJ5∪SJ5>SZ5′, the chemical oxygen demand exceeds the standard; wherein, i=5 indicates that the fifth change rate is the change rate corresponding to the chemical oxygen demand, and i′=5 indicates that the fifth data is the chemical oxygen demand.

[0092] The strategy optimization module is used by edge devices to optimize the sampling strategy according to water quality conditions and send it to the corresponding smart water station. The smart water station performs sampling according to the optimized sampling strategy.

[0093] The sampling strategy includes sampling indicators, sampling frequency and sampling points; the sampling indicators are the types of data in the real-time water quality data collected by the smart water station, the sampling frequency is the number of times the smart water station collects real-time water quality data per unit time, and the sampling point is the location of the smart water station.

[0094] Methods for optimizing sampling indicators include:

[0095] If the water quality condition is abnormal pH value, the redox potential, total nitrogen content, total phosphorus content and heavy metal content will be collected, that is, the real-time water quality data collected by the smart water station also includes redox potential, total nitrogen content, total phosphorus content and heavy metal content; indicating that the water body is polluted or affected by chemical reactions;

[0096] The redox potential is the ratio of oxides and reductants in the water body, which is used to further judge the oxidizability or reductibility of the water body and analyze whether there are oxidizing or reductive pollutants; the redox potential is obtained by the ORP electrode integrated in the smart water station; the total nitrogen content is the content of all nitrogen in the water body, including ammonia nitrogen, nitrite and nitrate, etc., and the total nitrogen content is obtained by the online nitrogen analyzer integrated in the smart water station; the total phosphorus content is the total amount of all phosphorus in the water, and the total phosphorus content is obtained by the online phosphorus analyzer integrated in the smart water station; both the total nitrogen content and the total phosphorus content are used to assess the eutrophication risk of water bodies; the heavy metal content is the concentration of heavy metals (such as lead, mercury, cadmium, etc.) in the water body, which is used to determine whether the water pollution is heavy metal pollution; the heavy metal content is obtained by the heavy metal ion sensor integrated in the smart water station.

[0097] If the water quality condition is that the dissolved oxygen concentration exceeds the standard, the biochemical oxygen demand, ammonia nitrogen concentration, total nitrogen content and total phosphorus content are collected; that is, the real-time water quality data collected by the smart water station also includes biochemical oxygen demand, ammonia nitrogen concentration, total nitrogen content and total phosphorus content; this indicates that there are too many organic matter in the water body, and further analysis of organic pollution and water body self-purification capacity is required;

[0098] Biochemical oxygen demand is the amount of oxygen required when organic matter in water is decomposed by microorganisms, and is used to determine the organic content in water. The biochemical oxygen demand is collected by the electrochemical sensor integrated in the smart water station. The ammonia nitrogen concentration is the concentration of nitrogen in the form of ammonia (NH3) in water, and is used to further evaluate the biodegradability of organic matter in water. The ammonia nitrogen concentration is obtained by the ammonia nitrogen sensor integrated in the smart water station.

[0099] If the water quality condition is that the turbidity exceeds the standard, the total suspended matter content and the number of blue algae will be collected; that is, the real-time water quality data collected by the smart water station also includes the total suspended matter content and the number of blue algae; indicating that the water body contains a large amount of suspended matter or particulate matter;

[0100] The total suspended solids content is the content of solid particles suspended in the water body, which is used to assess whether the increase in turbidity is caused by silt or other particles; the total suspended solids content is obtained by the optical turbidity sensor integrated in the smart water station; the number of cyanobacteria is the concentration of cyanobacteria in the water body, which is used to assess the risk of cyanobacteria outbreaks; the number of cyanobacteria is obtained by the cell counting sensor integrated in the smart water station.

[0101] If the water quality condition is that the conductivity exceeds the standard, the total dissolved solid content and heavy metal content will be collected; that is, the real-time water quality data collected by the smart water station also includes the total dissolved solid content and heavy metal content; this indicates that the concentration of dissolved ions in the water body is too high, which is usually related to mineralization and pollutants;

[0102] The total dissolved solids content is the total amount of dissolved solid matter in the water, including minerals and organic matter, and is used to confirm the cause of increased conductivity; the total dissolved solids are obtained by the total dissolved solids sensor integrated in the smart water station.

[0103] If the water quality condition is that the chemical oxygen demand exceeds the standard, the biochemical oxygen demand, organic pollutant content and ammonia nitrogen concentration will be collected; that is, the real-time water quality data collected by the smart water station also includes biochemical oxygen demand, organic pollutant content and ammonia nitrogen concentration; indicating that the water body contains more organic pollutants;

[0104] The organic pollutant content is the content of volatile organic compounds (VOCs) in the water body, which is used to confirm whether there is industrial organic pollution in the water body; the organic pollutant content is obtained by the volatile organic compound sensor integrated in the smart water station.

[0105] It should be understood that the purpose of optimizing the sampling index is to:

[0106] 1. Improve monitoring efficiency: By selecting sampling indicators related to specific water quality issues, water conditions can be monitored and evaluated more effectively, avoiding unnecessary data redundancy; and the optimized sampling indicators can ensure that relevant data can be quickly obtained and corresponding measures can be taken in a timely manner when water quality is abnormal;

[0107] 2. Enhance data relevance: Collecting relevant indicators based on different water quality conditions can provide more targeted analysis, help determine the specific pollution type or problem of the water body, and more comprehensively assess the ecological health and pollution risks of the water body;

[0108] 3. Improve decision-making support capabilities: Optimizing sampling indicators can provide more scientific data support for water quality management and decision-making, help relevant departments formulate effective water resource management strategies and pollution control measures; and can timely identify and evaluate potential environmental risks and reduce the occurrence of sudden pollution incidents.

[0109] Methods for optimizing the sampling frequency include:

[0110] The real-time water quality data and water quality change data are used as analysis data. The analysis data are input into the trained frequency prediction model to predict the corresponding sampling frequency and mark it as the predicted frequency. The sampling frequency of the smart water station is replaced with the corresponding predicted frequency, and the real-time water quality data is sampled.

[0111] The training process of the frequency prediction model includes:

[0112] A corresponding sampling frequency is set in advance for the analysis data of group A, where A is an integer greater than 1. The sampling frequency corresponding to the analysis data is collected by a technician in the field of history sampling strategy adjustment. The technician in the field analyzes each group of different analysis data in turn according to actual experience, obtains the corresponding sampling frequency, and sets the corresponding sampling frequency for the different analysis data of group A in turn;

[0113] The analysis data and the corresponding sampling frequencies are converted into a corresponding set of feature vectors; each set of feature vectors is used as the input of the structural evaluation model, and the structural evaluation model uses a set of predicted sampling frequencies corresponding to each set of analysis data as output, and uses the actual sampling frequency corresponding to each set of analysis data as the prediction target, where the actual sampling frequency is the sampling frequency corresponding to the analysis data collected in advance; the prediction results are evaluated for model accuracy using the mean absolute percentage error MAPE, and when the calculated MAPE is less than the preset MAPE, the structural evaluation model training is completed; wherein, the calculation formula of MAPE is: Among them, η d is the actual sampling frequency corresponding to the dth group of analysis data, is the predicted sampling frequency corresponding to the dth group of analysis data, d is the group number of the feature vector corresponding to the analysis data, A represents the number of predicted sampling frequencies, d∈[1,A]; a structural evaluation model is generated according to the analysis data to predict the sampling frequency; wherein the structural evaluation model is a deep belief network model, and the preset MAPE is pre-set by technicians in this field according to the accuracy required by the structural evaluation model.

[0114] It should be understood that the purpose of optimizing the sampling frequency is to:

[0115] 1. Timely capture of water quality changes: By adjusting the sampling frequency, important data can be obtained in a timely manner when water quality changes significantly, capturing instantaneous changes; and more frequent monitoring can be ensured at critical moments (such as pollution peaks) to avoid data omissions;

[0116] 2. Improve data accuracy: By dynamically adjusting the sampling frequency, frequent sampling can be reduced when water quality is relatively stable, reducing noise in the data and improving the overall quality of the data; a higher sampling frequency can more accurately reflect water quality conditions when changes are drastic, ensuring that the monitoring data is representative;

[0117] 3. Optimize monitoring resources: Reasonable arrangement of sampling frequency can effectively utilize human, equipment and time resources, reduce costs, and at the same time ensure the effectiveness of monitoring and ensure that key data can still be obtained when resources are limited.

[0118] Methods for optimizing sampling points include:

[0119] The smart water station corresponding to the edge device that evaluates the water quality status as abnormal is marked as an abnormal water station; the water station coordinates of all smart water stations are obtained, and the water station coordinates are obtained by the GPS integrated in the smart water station; the center coordinates of the source of the water body are obtained and marked as the source coordinates, that is, the location of the starting point of the water body is obtained; the source coordinates are measured and obtained by technical personnel in this field at the source of the water body; based on the source coordinates and the water station coordinates, the Euclidean distance between each smart water station and the source coordinates is calculated, and marked as the source distance; the source distances are sorted from small to large to generate a first sequence table; according to the positive order of the first sequence table, each smart water station is sequentially set with an increasing digital number, that is, the range of the digital number is [1, m], and m is the number of smart water stations; among them, if there are multiple smart water stations with equal source distances, the corresponding digital numbers are the same.

[0120] The smart water stations with digital numbers smaller than the corresponding digital numbers of the abnormal water stations are marked as upstream water stations; based on the water station coordinates of the abnormal water station and each upstream water station, the Euclidean distance between each upstream water station and the abnormal water station is calculated and marked as the water station distance; all water station distances are sorted from small to large to generate a second sequence table; the upstream water station corresponding to the first water station distance in the second sequence table is marked as an adjacent water station; based on the water station distance between the adjacent water station and the abnormal water station, the abnormal water station and the adjacent water stations are gradually moved to find the pollution source, and the abnormal water station is moved to the pollution source.

[0121] It should be understood that the purpose of optimizing the sampling points is to:

[0122] 1. Real-time monitoring: Smart water stations located near pollution sources can monitor water quality changes in real time and obtain key data on water quality in a timely manner;

[0123] 2. Optimize management measures: Accurate monitoring data can provide scientific basis for decision makers to help formulate effective pollution control and management measures; it can also evaluate the effectiveness of the governance measures taken and adjust the management strategy in a timely manner;

[0124] 3. Enhance emergency response capabilities: Monitoring equipment at pollution sources can provide a basis for rapid response to sudden pollution incidents and support emergency response; and through real-time monitoring of pollution sources, early warning can be achieved and relevant departments can be notified in time to take measures;

[0125] 4. Protect the ecological environment: By monitoring near the pollution source, water quality problems can be discovered and dealt with in a timely manner to protect the health of the water ecosystem; and effective water quality monitoring helps to achieve sustainable management of water resources and promote a balance between ecological protection and economic development.

[0126] like Figure 2 As shown, the steps to find the source of pollution include:

[0127] Step 1: Preset the initial search interval [a0, b0], where a0=0, b0 is the distance between the adjacent water station and the abnormal water station; set the segmentation ratio h, and in this embodiment, h=0.618 is preferred;

[0128] Step 2: According to the segmentation ratio h, the initial search interval [a0, b0] is divided into two new points ζ1 and ζ2, ζ1 = a0 + (1-h)(b0-a0), ζ2 = a0 + h (b0-a0);

[0129] Step 3: Move the abnormal water station in the direction of the adjacent water station by ζ1, and move the adjacent water station in the direction of the abnormal water station by ζ2; after the movement is completed, according to the water quality condition, the abnormal water station and the adjacent water station respectively collect corresponding data in the real-time water quality data; illustratively, if the water quality condition is an abnormal pH value, the abnormal water station and the adjacent water station respectively collect the pH value; if the water quality condition is an excessive dissolved oxygen concentration, the abnormal water station and the adjacent water station respectively collect the dissolved oxygen concentration; according to the data collected by the abnormal water station and the adjacent water station, respectively calculate the corresponding change rate; mark the change rate corresponding to the abnormal water station as the third change rate, and mark the change rate corresponding to the adjacent water station as the fourth change rate;

[0130] Step 4: If the third change rate is greater than or equal to the fourth change rate, the minimum value of the updated search interval is the minimum value of the search interval before the update, the maximum value is the point corresponding to the fourth change rate, and the abnormal water station is moved to the opposite direction of the adjacent water station by ζ1; if the third change rate is less than the fourth change rate, the minimum value of the updated search interval is the point corresponding to the third change rate, the maximum value is the maximum value of the search interval before the update, and the adjacent water station is moved to the opposite direction of the abnormal water station by ζ2; illustratively, after the initial search interval [a0, b0] is updated through points ζ1 and ζ2, if the third change rate is greater than or equal to the fourth change rate, the updated search interval is [a0, ζ2], if the third change rate is less than the fourth change rate, the updated search interval is [ζ1, b0]; it should be understood that when water quality is affected by pollution sources, the change rate of water quality data usually increases significantly, so by comparing the corresponding change rates of the abnormal water station and the adjacent water station, it can be inferred which smart water station is closer to the pollution source, and the water station with a larger change rate is closer to the pollution source;

[0131] Step 5: preset a width threshold, which is preset by technical personnel in this field according to actual conditions. Subtract the minimum value from the maximum value in the updated search interval to obtain the interval width; compare the interval width with the width threshold; if the interval width is greater than or equal to the width threshold, then re-divide the updated search interval into two new points ζ1 and ζ2 according to the segmentation ratio h, and return to step 3; if the interval width is less than the width threshold, obtain the water station coordinates corresponding to the current abnormal water station and the adjacent water station respectively, and mark them as real-time coordinates; take the mean coordinates of the two real-time coordinates as the pollution source; wherein the horizontal coordinate of the mean coordinate is the mean of the horizontal coordinates of the two real-time coordinates, and the vertical coordinate of the mean coordinate is the mean of the vertical coordinates of the two real-time coordinates.

[0132] This embodiment can evaluate the water quality status in real time based on the real-time collection and analysis of water quality data; and dynamically optimize the sampling strategy based on the water quality status; improve the timeliness and accuracy of water quality monitoring, and improve the quality of data monitoring, so as to effectively respond to water pollution incidents and ensure water resource management and public health; in addition, the deployment of edge devices can transmit information in a timely manner, reduce data transmission delays, improve information transmission efficiency, enhance data security, and improve the overall reliability and adaptability of the system.

[0133] Example 2

[0134] See also Figure 3 As shown, the part not described in detail in this embodiment is described in Example 1, and a method for transmitting sampling process information of a smart water station based on data perception is provided, which is applied to an edge device, and the method includes:

[0135] Receive real-time water quality data collected by smart water stations;

[0136] Receive historical water quality data collected by smart water stations, analyze real-time and historical water quality data, and calculate water quality change data and data thresholds;

[0137] Assess water quality status based on water quality change data, data thresholds and real-time water quality data;

[0138] According to the water quality conditions, the sampling strategy is optimized and sent to the smart water station.

[0139] Furthermore, the sampling strategy includes sampling indicators, sampling frequency and sampling points; the sampling indicators are the data types in the real-time water quality data collected by the smart water station, the sampling frequency is the number of times the smart water station collects real-time water quality data per unit time, and the sampling point is the location of the smart water station; edge devices are deployed for each smart water station;

[0140] The method for optimizing the sampling points comprises:

[0141] Mark the smart water station corresponding to the edge device that evaluates the water quality status as abnormal as an abnormal water station; obtain the water station coordinates of all smart water stations; obtain the central coordinates of the water source and mark them as the source coordinates; calculate the Euclidean distance between each smart water station and the source coordinates according to the source coordinates and the water station coordinates, and mark them as the source distance; sort the source distances from small to large to generate a first sequence table; according to the positive order of the first sequence table, set an increasing digital number for each smart water station in turn, that is, the range of the digital number is [1, m], and m is the number of smart water stations; among them, if there are multiple smart water stations with the same source distance, the corresponding digital numbers are the same;

[0142] The smart water stations with digital numbers smaller than the corresponding digital numbers of the abnormal water stations are marked as upstream water stations; based on the water station coordinates of the abnormal water station and each upstream water station, the Euclidean distance between each upstream water station and the abnormal water station is calculated and marked as the water station distance; all water station distances are sorted from small to large to generate a second sequence table; the upstream water station corresponding to the first water station distance in the second sequence table is marked as an adjacent water station; based on the water station distance between the adjacent water station and the abnormal water station, the abnormal water station and the adjacent water stations are gradually moved to find the pollution source, and the abnormal water station is moved to the pollution source.

[0143] Furthermore, the step of finding the pollution source includes:

[0144] Step 1: Preset the initial search interval [a0, b0], where a0 = 0 and b0 is the distance between the adjacent water station and the abnormal water station; set the segmentation ratio h;

[0145] Step 2: According to the segmentation ratio h, the initial search interval [a0, b0] is divided into two new points ζ1 and ζ2, ζ1 = a0 + (1-h)(b0-a0), ζ2 = a0 + h (b0-a0);

[0146] Step 3: Move the abnormal water station in the direction of the adjacent water station by ζ1, and move the adjacent water station in the direction of the abnormal water station by ζ2; after the movement is completed, according to the water quality conditions, the abnormal water station and the adjacent water station respectively collect corresponding data in the real-time water quality data; according to the data collected by the abnormal water station and the adjacent water station, respectively calculate the corresponding change rate; mark the change rate corresponding to the abnormal water station as the third change rate, and mark the change rate corresponding to the adjacent water station as the fourth change rate;

[0147] Step 4: If the third change rate is greater than or equal to the fourth change rate, the minimum value of the updated search interval is the minimum value of the search interval before the update, the maximum value is the point corresponding to the fourth change rate, and the abnormal water station is moved in the opposite direction of the adjacent water station by ζ1; if the third change rate is less than the fourth change rate, the minimum value of the updated search interval is the point corresponding to the third change rate, the maximum value is the maximum value of the search interval before the update, and the adjacent water station is moved in the opposite direction of the abnormal water station by ζ2;

[0148] Step 5: Preset the width threshold, subtract the minimum value from the maximum value in the updated search interval to obtain the interval width; compare the interval width with the width threshold; if the interval width is greater than or equal to the width threshold, then re-divide the updated search interval according to the segmentation ratio h to obtain two new points ζ1 and ζ2, and return to step 3; if the interval width is less than the width threshold, obtain the water station coordinates corresponding to the current abnormal water station and the adjacent water station respectively, and mark them as real-time coordinates; take the mean coordinate of the two real-time coordinates as the pollution source.

[0149] Furthermore, the method for optimizing the sampling frequency includes:

[0150] The real-time water quality data and water quality change data are used as analysis data, and the analysis data is input into the trained frequency prediction model to predict the corresponding sampling frequency and mark it as the predicted frequency; the sampling frequency of the smart water station is replaced with the corresponding predicted frequency, and the real-time water quality data is sampled;

[0151] The training process of the frequency prediction model includes:

[0152] A corresponding sampling frequency is set for group A of analysis data in advance, where A is an integer greater than 1; the analysis data and the corresponding sampling frequency are converted into a corresponding set of feature vectors; each set of feature vectors is used as the input of a structural evaluation model, and the structural evaluation model uses a set of predicted sampling frequencies corresponding to each set of analysis data as output, and uses the actual sampling frequency corresponding to each set of analysis data as a prediction target, where the actual sampling frequency is the sampling frequency corresponding to the analysis data collected in advance; the prediction results are evaluated for model accuracy using the mean absolute percentage error (MAPE), and when the calculated MAPE is less than the preset MAPE, the training of the structural evaluation model is completed; a structural evaluation model that predicts the sampling frequency based on the analysis data is generated; the structural evaluation model is a deep belief network model.

[0153] Furthermore, the real-time water quality data is water quality data collected in real time; the water quality data includes physical data and chemical data; the physical data includes pH value, turbidity and conductivity; the chemical data includes dissolved oxygen concentration and chemical oxygen demand;

[0154] The method for calculating water quality change data comprises:

[0155] The water quality data in the historical water quality data that are adjacent to the collection time of the real-time water quality data are marked as adjacent water quality data; the corresponding data in the adjacent water quality data are subtracted from each data in the real-time water quality data to obtain the change amount of each data in the water quality data; the change amount of each data is multiplied by the preset sampling frequency to obtain the change rate of each data in the water quality data, and used as the water quality change data.

[0156] Furthermore, the method for calculating the data threshold includes:

[0157] The data threshold includes a numerical range value and a variation range value; wherein the numerical range value and the variation range value correspond one-to-one to each data in the water quality data;

[0158] The calculation method of numerical range value includes:

[0159] Count the number of groups of water quality data in the historical water quality data and add 1 as the number of data; take the real-time water quality data and the historical water quality data as the total water quality data; add the data of the same type in the total water quality data in turn, and divide it by the number of data to obtain the data mean corresponding to each data in the water quality data; subtract the corresponding data mean from each data in the total water quality data and then square it to obtain the data difference corresponding to each data in the total water quality data; add the data difference of each data in the total water quality data in turn, and then divide it by the number of data to obtain the data variance corresponding to each data in the water quality data; take the square root of each data variance to obtain the data standard deviation corresponding to each data in the water quality data; add three times the corresponding data standard deviation to the data mean corresponding to each data in the water quality data as the maximum value of the corresponding numerical range, and subtract three times the corresponding data standard deviation from the numerical mean corresponding to each data in the water quality data as the minimum value of the corresponding numerical range.

[0160] Furthermore, the calculation method of the variation range value includes:

[0161] Subtract 1 from the number of data as the number of changes; collect all calculated water quality change data and mark them as total change data; add the same type of change rates in the total change data to an analysis set respectively, and the analysis set corresponds to the change rate in the water quality change data one by one; sort the change rates in each analysis set from small to large to generate a corresponding sorting table; multiply the number of changes by one quarter to obtain the first number; multiply the number of changes by three quarters to obtain the second number; wherein, if the first number or the second number is not an integer, round up; in each sorting table, mark the change rate in the first number as the first change rate, and mark the change rate in the second number as the second change rate; subtract the first change rate from the second change rate corresponding to each analysis set to obtain the change rate interval; multiply the change rate interval of each analysis set by 1.5 to obtain the data interval; subtract the data interval from the first change rate of each analysis set as the minimum value of the corresponding change range value; add the data interval to the second change rate of each analysis set as the maximum value of the corresponding change range value.

[0162] Furthermore, the water quality condition includes normal water quality and abnormal water quality; the abnormal water quality includes abnormal pH value, excessive dissolved oxygen concentration, excessive turbidity, excessive conductivity and excessive chemical oxygen demand;

[0163] The method for optimizing the sampling index comprises:

[0164] If the water quality condition is an abnormal pH value, the redox potential, total nitrogen content, total phosphorus content and heavy metal content are collected, that is, the real-time water quality data collected by the smart water station also includes redox potential, total nitrogen content, total phosphorus content and heavy metal content; the redox potential is the ratio of oxides and reductants in the water body, the total nitrogen content is the content of all nitrogen in the water body, the total phosphorus content is the total amount of all phosphorus in the water, and the heavy metal content is the concentration of heavy metals in the water body;

[0165] If the water quality condition is that the dissolved oxygen concentration exceeds the standard, the biochemical oxygen demand, ammonia nitrogen concentration, total nitrogen content and total phosphorus content are collected; that is, the real-time water quality data collected by the smart water station also includes biochemical oxygen demand, ammonia nitrogen concentration, total nitrogen content and total phosphorus content; biochemical oxygen demand is the amount of oxygen required when organic matter in the water body is decomposed by microorganisms, and ammonia nitrogen concentration is the concentration of nitrogen in the form of ammonia in the water body;

[0166] If the water quality condition is that the turbidity exceeds the standard, the total suspended solids content and the number of cyanobacteria will be collected; that is, the real-time water quality data collected by the smart water station also includes the total suspended solids content and the number of cyanobacteria; the total suspended solids content is the content of solid particles suspended in the water body, and the number of cyanobacteria is the concentration of cyanobacteria in the water body;

[0167] If the water quality condition is that the conductivity exceeds the standard, the total dissolved solid content and heavy metal content will be collected; that is, the real-time water quality data collected by the smart water station also includes the total dissolved solid content and heavy metal content; the total dissolved solid content is the total amount of dissolved solid matter in the water body;

[0168] If the water quality condition is that the chemical oxygen demand exceeds the standard, the biochemical oxygen demand, organic pollutant content and ammonia nitrogen concentration will be collected; that is, the real-time water quality data collected by the smart water station also includes biochemical oxygen demand, organic pollutant content and ammonia nitrogen concentration; indicating that the water body contains more organic pollutants; the organic pollutant content is the content of volatile organic compounds in the water body.

[0169] Furthermore, the method for assessing water quality comprises:

[0170] Compare each change rate in the water quality change data with the corresponding change range value in the data threshold; compare each data in the real-time water quality data with the corresponding numerical range value in the data threshold; if BF i ≤BH i ≤BF i ′∩SZ i′ ≤SJ i′ ≤SZ i ″, the water quality is normal; among them, BH i is the i-th change rate in the water quality change data, SJ i′ is the i′th data in the real-time water quality data, BF iis the minimum value of the change range corresponding to the i-th change rate, BF i ′ is the maximum value of the change range corresponding to the i-th change rate, SZ i′ is the minimum value of the numerical range corresponding to the i′th data, SZ i ″ is the maximum value of the numerical range corresponding to the i′th data, i∈[1,5], i′∈[1,5], ∩ represents and;

[0171] If BF1>BH1∪BH1>BF1′∪SZ1>SJ1∪SJ1>SZ1′, the pH value is abnormal;

[0172] If BF2>BH2∪BH2>BF2′∪SZ2>SJ2∪SJ2>SZ2′, the dissolved oxygen concentration exceeds the standard;

[0173] If BF3>BH3∪BH3>BF3′∪SZ3>SJ3∪SJ3>SZ3′, the turbidity exceeds the standard;

[0174] If BF4>BH4∪BH4>BF4′∪SZ4>SJ4∪SJ4>SZ′4, the conductivity exceeds the standard;

[0175] If BF5>BH5∪BH5>BF5′∪SZ5>SJ5∪SJ5>SZ5′, the chemical oxygen demand exceeds the standard;

[0176] Among them, i=1 means that the first change rate is the change rate corresponding to the pH value, i′=1 means that the first data is the pH value, ∪ means or; i=2 means that the second change rate is the change rate corresponding to the dissolved oxygen concentration, i′=2 means that the second data is the dissolved oxygen concentration; i=3 means that the third change rate is the change rate corresponding to the turbidity, i′=3 means that the third data is turbidity; i=4 means that the fourth change rate is the change rate corresponding to the conductivity, i′=4 means that the fourth data is conductivity; i=5 means that the fifth change rate is the change rate corresponding to the chemical oxygen demand, i′=5 means that the fifth data is the chemical oxygen demand.

[0177] Example 3

[0178] The present application also provides an electronic device. The electronic device may include one or more processors and one or more memories. The memories store computer-readable codes, and when the computer-readable codes are executed by one or more processors, the above-mentioned method for transmitting the sampling process information of the smart water station based on data perception can be executed.

[0179] The electronic device may include a bus, one or more CPUs 502, ROM 503, RAM 504, a communication port connected to a network, input / output, a hard disk, etc. A storage device in the electronic device, such as a ROM or a hard disk, may store the data-aware smart water station sampling process information transmission method provided in the present application. Furthermore, the electronic device may also include a user interface.

[0180] Example 4

[0181] One embodiment of the present application discloses a computer-readable storage medium. Computer-readable instructions are stored on the computer-readable storage medium. When the computer-readable instructions are executed by the processor, the data-aware smart water station sampling process information transmission method according to the embodiment of the present application described with reference to the above figures can be executed. The storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory (cache), etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0182] In addition, according to the implementation of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the present application provides a non-transitory machine-readable storage medium, which stores machine-readable instructions, and the machine-readable instructions can be executed by a processor to execute instructions corresponding to the method steps provided in the present application, for example: a method for transmitting information on the sampling process of a smart water station based on data perception. When the computer program is executed by a central processing unit (CPU), the above functions defined in the method of the present application are executed.

[0183] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0184] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A data-aware smart water station sampling process information transmission method, characterized in that: Applied to edge devices, including: Receive real-time water quality data collected by smart water stations; Receive historical water quality data collected by smart water stations, analyze real-time and historical water quality data, and calculate water quality change data and data thresholds; Assess water quality status based on water quality change data, data thresholds and real-time water quality data; According to the water quality conditions, the sampling strategy is optimized and sent to the smart water station.

2. According to the data perception-based smart water station sampling process information transmission method of claim 1, it is characterized in that: The sampling strategy includes sampling indicators, sampling frequency and sampling points; the sampling indicators are the data types in the real-time water quality data collected by the smart water station, the sampling frequency is the number of times the smart water station collects real-time water quality data per unit time, and the sampling point is the location of the smart water station; Edge devices are deployed at each smart water station; The method for optimizing the sampling points comprises: Mark the smart water station corresponding to the edge device that evaluates the water quality status as abnormal as an abnormal water station; obtain the water station coordinates of all smart water stations; obtain the central coordinates of the water source and mark them as the source coordinates; calculate the Euclidean distance between each smart water station and the source coordinates according to the source coordinates and the water station coordinates, and mark them as the source distance; sort the source distances from small to large to generate a first sequence table; according to the positive order of the first sequence table, set an increasing digital number for each smart water station in turn, that is, the range of the digital number is [1, m], and m is the number of smart water stations; among them, if there are multiple smart water stations with the same source distance, the corresponding digital numbers are the same; The smart water stations with digital numbers smaller than the corresponding digital numbers of the abnormal water stations are marked as upstream water stations; based on the water station coordinates of the abnormal water station and each upstream water station, the Euclidean distance between each upstream water station and the abnormal water station is calculated and marked as the water station distance; all water station distances are sorted from small to large to generate a second sequence table; the upstream water station corresponding to the first water station distance in the second sequence table is marked as an adjacent water station; based on the water station distance between the adjacent water station and the abnormal water station, the abnormal water station and the adjacent water stations are gradually moved to find the pollution source, and the abnormal water station is moved to the pollution source.

3. The method for transmitting sampling process information of a smart water station based on data perception according to claim 2 is characterized in that: The steps of finding the pollution source include: Step 1: Preset the initial search interval [a0, b0], where a0 = 0 and b0 is the distance between the adjacent water station and the abnormal water station; set the segmentation ratio h; Step 2: According to the segmentation ratio h, the initial search interval [a0, b0] is divided into two new points ζ1 and ζ2, ζ1 = a0 + (1-h)(b0-a0), ζ2 = a0 + h (b0-a0); Step 3: Move the abnormal water station in the direction of the adjacent water station by ζ1, and move the adjacent water station in the direction of the abnormal water station by ζ2; after the movement is completed, according to the water quality conditions, the abnormal water station and the adjacent water station respectively collect corresponding data in the real-time water quality data; according to the data collected by the abnormal water station and the adjacent water station, respectively calculate the corresponding change rate; mark the change rate corresponding to the abnormal water station as the third change rate, and mark the change rate corresponding to the adjacent water station as the fourth change rate; Step 4: If the third change rate is greater than or equal to the fourth change rate, the minimum value of the updated search interval is the minimum value of the search interval before the update, the maximum value is the point corresponding to the fourth change rate, and the abnormal water station is moved in the opposite direction of the adjacent water station by ζ1; if the third change rate is less than the fourth change rate, the minimum value of the updated search interval is the point corresponding to the third change rate, the maximum value is the maximum value of the search interval before the update, and the adjacent water station is moved in the opposite direction of the abnormal water station by ζ2; Step 5: Preset the width threshold, subtract the minimum value from the maximum value in the updated search interval to obtain the interval width; compare the interval width with the width threshold; if the interval width is greater than or equal to the width threshold, then re-divide the updated search interval according to the segmentation ratio h to obtain two new points ζ1 and ζ2, and return to step 3; if the interval width is less than the width threshold, obtain the water station coordinates corresponding to the current abnormal water station and the adjacent water station respectively, and mark them as real-time coordinates; take the mean coordinate of the two real-time coordinates as the pollution source.

4. The method for transmitting sampling process information of a smart water station based on data perception according to claim 3 is characterized in that: The method for optimizing the sampling frequency comprises: The real-time water quality data and water quality change data are used as analysis data, and the analysis data is input into the trained frequency prediction model to predict the corresponding sampling frequency and mark it as the predicted frequency; the sampling frequency of the smart water station is replaced with the corresponding predicted frequency, and the real-time water quality data is sampled; The training process of the frequency prediction model includes: A corresponding sampling frequency is set for group A of analysis data in advance, where A is an integer greater than 1; the analysis data and the corresponding sampling frequency are converted into a corresponding set of feature vectors; each set of feature vectors is used as the input of a structural evaluation model, and the structural evaluation model uses a set of predicted sampling frequencies corresponding to each set of analysis data as output, and uses the actual sampling frequency corresponding to each set of analysis data as a prediction target, where the actual sampling frequency is the sampling frequency corresponding to the analysis data collected in advance; the prediction results are evaluated for model accuracy using the mean absolute percentage error (MAPE), and when the calculated MAPE is less than the preset MAPE, the training of the structural evaluation model is completed; a structural evaluation model that predicts the sampling frequency based on the analysis data is generated; the structural evaluation model is a deep belief network model.

5. The method for transmitting sampling process information of a smart water station based on data perception according to claim 4 is characterized in that: The real-time water quality data is water quality data collected in real time; the water quality data includes physical data and chemical data; the physical data includes pH value, turbidity and conductivity; the chemical data includes dissolved oxygen concentration and chemical oxygen demand; The method for calculating water quality change data comprises: The water quality data in the historical water quality data that are adjacent to the collection time of the real-time water quality data are marked as adjacent water quality data; the corresponding data in the adjacent water quality data are subtracted from each data in the real-time water quality data to obtain the change amount of each data in the water quality data; the change amount of each data is multiplied by the preset sampling frequency to obtain the change rate of each data in the water quality data, and used as the water quality change data.

6. The method for transmitting sampling process information of a smart water station based on data perception according to claim 5 is characterized in that: The method for calculating the data threshold comprises: The data threshold includes a numerical range value and a variation range value; wherein the numerical range value and the variation range value correspond one-to-one to each data in the water quality data; The calculation method of numerical range value includes: Count the number of groups of water quality data in the historical water quality data and add 1 as the number of data; take the real-time water quality data and the historical water quality data as the total water quality data; add the data of the same type in the total water quality data in turn, and divide it by the number of data to obtain the data mean corresponding to each data in the water quality data; subtract the corresponding data mean from each data in the total water quality data and then square it to obtain the data difference corresponding to each data in the total water quality data; add the data difference of each data in the total water quality data in turn, and then divide it by the number of data to obtain the data variance corresponding to each data in the water quality data; take the square root of each data variance to obtain the data standard deviation corresponding to each data in the water quality data; add three times the corresponding data standard deviation to the data mean corresponding to each data in the water quality data as the maximum value of the corresponding numerical range, and subtract three times the corresponding data standard deviation from the numerical mean corresponding to each data in the water quality data as the minimum value of the corresponding numerical range.

7. The method for transmitting sampling process information of a smart water station based on data perception according to claim 6 is characterized in that: The calculation method of the variation range value includes: Subtract 1 from the number of data as the number of changes; collect all calculated water quality change data and mark them as total change data; add the same type of change rates in the total change data to an analysis set respectively, and the analysis set corresponds to the change rate in the water quality change data one by one; sort the change rates in each analysis set from small to large to generate a corresponding sorting table; multiply the number of changes by one quarter to obtain the first number; multiply the number of changes by three quarters to obtain the second number; wherein, if the first number or the second number is not an integer, round up; in each sorting table, mark the change rate in the first number as the first change rate, and mark the change rate in the second number as the second change rate; subtract the first change rate from the second change rate corresponding to each analysis set to obtain the change rate interval; multiply the change rate interval of each analysis set by 1.5 to obtain the data interval; subtract the data interval from the first change rate of each analysis set as the minimum value of the corresponding change range value; add the data interval to the second change rate of each analysis set as the maximum value of the corresponding change range value.

8. The method for transmitting sampling process information of a smart water station based on data perception according to claim 7 is characterized in that: The water quality conditions include normal water quality and abnormal water quality; the abnormal water quality includes abnormal pH value, excessive dissolved oxygen concentration, excessive turbidity, excessive conductivity and excessive chemical oxygen demand; The method for optimizing the sampling index comprises: If the water quality condition is an abnormal pH value, the redox potential, total nitrogen content, total phosphorus content and heavy metal content are collected, that is, the real-time water quality data collected by the smart water station also includes redox potential, total nitrogen content, total phosphorus content and heavy metal content; the redox potential is the ratio of oxides and reductants in the water body, the total nitrogen content is the content of all nitrogen in the water body, the total phosphorus content is the total amount of all phosphorus in the water, and the heavy metal content is the concentration of heavy metals in the water body; If the water quality condition is that the dissolved oxygen concentration exceeds the standard, the biochemical oxygen demand, ammonia nitrogen concentration, total nitrogen content and total phosphorus content are collected; that is, the real-time water quality data collected by the smart water station also includes biochemical oxygen demand, ammonia nitrogen concentration, total nitrogen content and total phosphorus content; biochemical oxygen demand is the amount of oxygen required when organic matter in the water body is decomposed by microorganisms, and ammonia nitrogen concentration is the concentration of nitrogen in the form of ammonia in the water body; If the water quality condition is that the turbidity exceeds the standard, the total suspended solids content and the number of cyanobacteria will be collected; that is, the real-time water quality data collected by the smart water station also includes the total suspended solids content and the number of cyanobacteria; the total suspended solids content is the content of solid particles suspended in the water body, and the number of cyanobacteria is the concentration of cyanobacteria in the water body; If the water quality condition is that the conductivity exceeds the standard, the total dissolved solid content and heavy metal content will be collected; that is, the real-time water quality data collected by the smart water station also includes the total dissolved solid content and heavy metal content; the total dissolved solid content is the total amount of dissolved solid matter in the water body; If the water quality condition is that the chemical oxygen demand exceeds the standard, the biochemical oxygen demand, organic pollutant content and ammonia nitrogen concentration will be collected; that is, the real-time water quality data collected by the smart water station also includes biochemical oxygen demand, organic pollutant content and ammonia nitrogen concentration; indicating that the water body contains more organic pollutants; the organic pollutant content is the content of volatile organic compounds in the water body.

9. The method for transmitting sampling process information of a smart water station based on data perception according to claim 8 is characterized in that: The methods for assessing water quality conditions include: Compare each change rate in the water quality change data with the corresponding change range value in the data threshold; compare each data in the real-time water quality data with the corresponding numerical range value in the data threshold; if BF i ≤BH i ≤BF i ′∩SZ i′ ≤SJ i′ ≤SZ i ″, the water quality is normal; among them, BH i is the i-th change rate in the water quality change data, SJ i′ is the i′th data in the real-time water quality data, BF i is the minimum value of the change range corresponding to the i-th change rate, BF i ′ is the maximum value of the change range corresponding to the i-th change rate, SZ i′ is the minimum value of the numerical range corresponding to the i′th data, SZ i ″ is the maximum value of the numerical range corresponding to the i′th data, i∈[1,5], i′∈[1,5], ∩ represents and; If BF1>BH1∪BH1>BF1′∪SZ1>SJ1∪SJ1>SZ1′, the pH value is abnormal; If BF2>BH2∪BH2>BF2′∪SZ2>SJ2∪SJ2>SZ2′, the dissolved oxygen concentration exceeds the standard; If BF3>BH3∪BH3>BF3′∪SZ3>SJ3∪SJ3>SZ3′, the turbidity exceeds the standard; If BF4>BH4∪BH4>BF4′∪SZ4>SJ4∪SJ4>SZ′4, the conductivity exceeds the standard; If BF5>BH5∪BH5>BF5′∪SZ5>SJ5∪SJ5>SZ5′, the chemical oxygen demand exceeds the standard; Among them, i=1 means that the first change rate is the change rate corresponding to the pH value, i′=1 means that the first data is the pH value, ∪ means or; i=2 means that the second change rate is the change rate corresponding to the dissolved oxygen concentration, i′=2 means that the second data is the dissolved oxygen concentration; i=3 means that the third change rate is the change rate corresponding to the turbidity, i′=3 means that the third data is turbidity; i=4 means that the fourth change rate is the change rate corresponding to the conductivity, i′=4 means that the fourth data is conductivity; i=5 means that the fifth change rate is the change rate corresponding to the chemical oxygen demand, i′=5 means that the fifth data is the chemical oxygen demand.

10. A data-aware smart water station sampling process information transmission system, implementing the data-aware smart water station sampling process information transmission method according to any one of claims 1 to 9, characterized in that: Applied to edge devices, including: Data collection module, used to receive real-time water quality data collected by smart water stations; The data analysis module is used to receive the historical water quality data collected by the smart water station, analyze the real-time water quality data and historical water quality data, and calculate the water quality change data and data threshold; A water quality assessment module is used to assess water quality conditions based on water quality change data, data thresholds and real-time water quality data; The strategy optimization module is used to optimize the sampling strategy according to the water quality conditions and send it to the smart water station.

Images