Intelligent identification method for microbial pollution source in water body

Abstract

The invention provides an intelligent identification method for a water microbial pollution source, and relates to the technical field of pollution source identification, and the method comprises the steps: connecting a monitoring platform, and obtaining water monitoring data; performing time and space integration arrangement on the water body monitoring data according to the position acquisition information and the time acquisition information, and establishing a space-time relationship between the acquired data; based on the time-space relationship between the collected data, fitting the pollution source time change relationship and the space position change relationship of the microbiological detection data; and according to the pollution source time change characteristics, the pollution source space position change characteristics and the breeding flow characteristics of the microbial pollution source, pollution source classification and identification are carried out, and the pollution source type and position positioning are determined. Through the pollution source identification method and device, the technical problem that in the prior art, multiple pollution sources are difficult to distinguish under the condition that the water body pollution sources are complex, and consequently the pollution source identification accuracy is low can be solved, and the pollution source identification accuracy is improved through the change relation between the water body monitoring position and the microorganisms.

Description

Technical Field

[0001] The present application relates to the technical field of pollution source identification, and in particular to a method for intelligently identifying water microbial pollution sources. Background Art

[0002] The acceleration of urbanization has led to the increasing diversification of water pollution sources and the complexity of pollution types, which has brought great challenges to water quality monitoring and pollution source identification. Existing water pollution source identification methods often rely on a single monitoring indicator or certain water quality indicators, and are limited in distinguishing multiple pollution sources in complex environments. In natural water bodies, pollution sources are often intertwined, which can easily cause pollutants to diffuse and mix, making the characteristics of water pollution more complex, making it difficult for traditional identification methods to reflect the dynamic changes of pollution sources in real time when facing rapidly changing pollution sources, especially under the interaction of different pollution sources. It is difficult to track the specific source of the pollution source, which limits the accurate identification of the pollution source.

[0003] In summary, the prior art has a technical problem in which the accuracy of pollution source identification is low due to the difficulty in distinguishing multiple pollution sources when the water pollution sources are complex. Summary of the invention

[0004] The purpose of this application is to provide an intelligent identification method for water microbial pollution sources, so as to solve the technical problem in the prior art that it is difficult to distinguish multiple pollution sources when the water pollution sources are complex, resulting in low pollution source identification accuracy.

[0005] In view of the above problems, the present application provides a method for intelligent identification of water body microbial pollution sources, wherein the method for intelligent identification of water body microbial pollution sources includes: connecting to a monitoring platform to obtain water body monitoring data, the water body monitoring data including location collection information and time collection information; integrating and arranging the water body monitoring data in time and space according to the location collection information and time collection information, and establishing a spatiotemporal relationship between the collected data; obtaining microbial detection data of the water body monitoring data, and fitting the time change relationship and spatial position change relationship of the pollution source for the microbial detection data based on the spatiotemporal relationship between the collected data, and determining the time change characteristics of the pollution source and the spatial position change characteristics of the pollution source; classifying and identifying the pollution source according to the time change characteristics and spatial position change characteristics of the pollution source, and combining the reproduction and flow characteristics of the microbial pollution source, and determining the type and location of the pollution source.

[0006] The technical solution provided in this application has at least the following technical effects or advantages:

[0007] By connecting to the monitoring platform, water body monitoring data is obtained, and the water body monitoring data includes location collection information and time collection information; the water body monitoring data is integrated and arranged in time and space according to the location collection information and time collection information, and the spatiotemporal relationship between the collected data is established; the microbial detection data of the water body monitoring data is obtained, and based on the spatiotemporal relationship between the collected data, the time change relationship and spatial position change relationship of the pollution source are fitted to the microbial detection data to determine the time change characteristics of the pollution source and the spatial position change characteristics of the pollution source; according to the time change characteristics of the pollution source and the spatial position change characteristics of the pollution source, and in combination with the reproduction and flow characteristics of the microbial pollution source, the pollution source is classified and identified, and the type and location of the pollution source are determined. In other words, by combining the collected location collection information and time collection information, integrating and arranging in time and space, establishing the spatiotemporal relationship between the data, combining the reproduction and flow characteristics of the microorganisms, the pollution source is classified and identified, and different types of pollution sources are accurately classified, and their spatial positions are located, thereby improving the accuracy of identifying water body microbial pollution sources.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0010] Figure 1 This is a flow chart of a method for intelligently identifying sources of microbial contamination in water bodies according to the present application.

[0011] Figure 2 This is a schematic diagram of the process of determining change characteristics in a method for intelligently identifying sources of microbial pollution in water bodies in this application. DETAILED DESCRIPTION

[0012] This application provides an intelligent identification method for water microbial pollution sources, which solves the technical problem in the prior art that it is difficult to distinguish multiple pollution sources when the water pollution sources are complex, resulting in low pollution source identification accuracy. By combining the collected location collection information and time collection information, performing spatiotemporal integration and arrangement, establishing the spatiotemporal relationship between the data, combining the reproduction and flow characteristics of microorganisms, classifying and identifying pollution sources, accurately classifying different types of pollution sources, and locating their spatial positions, the accuracy of water microbial pollution source identification is improved.

[0013] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0014] For examples, please see the attached Figure 1 The present application provides a method for intelligently identifying water microbial pollution sources, wherein the method for intelligently identifying water microbial pollution sources specifically comprises the following steps:

[0015] S100: Connecting to a monitoring platform to obtain water body monitoring data, where the water body monitoring data includes location collection information and time collection information.

[0016] Specifically, the monitoring platform is connected through a wireless sensor network or Internet of Things (IoT) device to ensure that the data acquisition device can send real-time data to the platform. The monitoring platform is used to collect, store, process and analyze environmental data (such as water quality data, climate data, microbial data, etc.), including sensors, data acquisition systems, servers and data analysis tools, which are usually connected to the network and transmit data wirelessly or wired. Information about the state and quality of water bodies is collected through sensors, equipment or manual sampling, including physical parameters (such as temperature, pH value, dissolved oxygen), chemical parameters (such as ammonia nitrogen, heavy metal concentration) and biological data (such as microbial species and quantity), reflecting the pollution status of the water body. Each collected data will be accompanied by a timestamp, indicating the time when the data was collected.

[0017] After the connection is established, water monitoring data is downloaded from the monitoring platform in real time or periodically, including location collection information (such as longitude and latitude coordinates) and time collection information (such as the timestamp of data recording). Location collection information is the geographical location information when the data is obtained, usually recorded through GPS or other positioning systems (such as GIS systems) to help determine the geographical location of the pollution source and its changes. Time collection information is the timestamp of the data acquisition, which is used to record the specific time of data collection. By acquiring water monitoring data including location collection information and time collection information, the spatiotemporal dynamic characteristics of the pollution source are analyzed.

[0018] S200: Integrate and arrange the water body monitoring data in time and space according to the location collection information and the time collection information, and establish a spatiotemporal relationship between the collected data.

[0019] Further, the present application S200 includes:

[0020] According to the position acquisition information, coordinate positioning is performed to identify the position-direction relationship; according to the time acquisition information, a time series arrangement relationship is obtained according to the interval time length of the acquisition timestamps; a mapping association is established between the water body monitoring data and the position-direction relationship and the time series arrangement relationship to obtain the spatiotemporal relationship between the acquired data.

[0021] Specifically, coordinate positioning is performed based on the location collection information, that is, the geographical location of each data point is determined by coordinates such as longitude and latitude. The longitude and latitude coordinates are input into the GIS software or map service API to confirm the specific location of the coordinate point on the map. By determining the coordinate difference between two adjacent monitoring points, the position and direction relationship of each monitoring point relative to other monitoring points is marked to identify the position and direction relationship. The position and direction relationship refers to the relative position relationship between monitoring points. The spatial relationship between different monitoring points is determined by calculating the distance or direction (such as north, south, east, and west) between the coordinate points.

[0022] Preprocess the time series data, such as filling missing values, smoothing outliers, etc. Extract the timestamp of each data point from the monitoring data, and sort each monitoring data in the order of the timestamps in the time collection information to form a time series. Map and associate the position direction relationship and the time sequence arrangement relationship, combine the water body monitoring data at each moment with its corresponding geographical location, and form a data set containing time and space information. The spatiotemporal relationship between the collected data refers to the mutual connection between the two dimensions of time and space, that is, the connection between the geographical distribution of the data collection points and the collection time, which can reveal the dynamic process of the pollution source changing over time and space. By establishing the spatiotemporal relationship, determine how the water body monitoring data changes at different time and space locations, and how these changes are related to each other. By establishing the spatiotemporal relationship, accurately track the location changes of the pollution source and its dynamic changes over time.

[0023] S300: Acquire the microbial detection data of the water body monitoring data, and based on the spatiotemporal relationship between the collected data, fit the time variation relationship and spatial position variation relationship of the pollution source to the microbial detection data to determine the time variation characteristics and spatial position variation characteristics of the pollution source.

[0024] Specifically, through professional monitoring equipment (such as microscopes, image recognition technology, test methods, etc.), the microbial data in the water body, including information such as microbial species, concentration, and distribution, is obtained to detect whether the water body is contaminated by microorganisms. For example, the presence of certain bacteria, algae, or other microorganisms can be observed through a microscope. Potential sources of pollution are marked based on abnormal concentrations or types of microorganisms in the water body. Based on the location collection information and time collection information in the acquired water body monitoring data, the spatiotemporal relationship between the data is extracted, that is, the spatiotemporal relationship between the collected data.

[0025] According to the spatiotemporal relationship between the collected data, the pollution source is identified and the pollution source is marked for the microbial detection data. Using the coordinate positioning information, all the time series collected data at the pollution source marked position are screened out, and the longitudinal pollution source characteristics are fitted, that is, the change trend of the pollution source is analyzed and fitted in the time dimension to determine the characteristics of the pollution source changing over time, such as the increase or decrease of pollutant concentration.

[0026] The terrain, water flow position and water body monitoring position are associated to determine the possible impact of terrain and water flow on each position. Combined with the impact of terrain and water flow and the temporal changes of pollution sources, an impact time period is determined, which represents the time required for the pollution source to spread from one point to another. The impact time period is used as a time gradient to align the microbial detection data at different time points and different spatial positions. Using the extracted data, the change characteristics of microbial data at different spatial positions within the same time gradient are analyzed as the spatial position change characteristics of the pollution source. The time change characteristics can be the growth trend or periodic changes in the concentration of the pollution source, and the spatial position change characteristics can be the diffusion path and range of the pollution source. For example, if the concentration of the pollution source continues to rise over a period of time and expands to the surrounding areas, then the time change characteristics are the concentration increase, and the spatial position change characteristics are the diffusion path. By fitting the relationship between time change and spatial position change, the dynamic characteristics of the pollution source can be accurately revealed, the future development trend of the pollution source and its possible diffusion range can be predicted, and countermeasures can be prepared in advance.

[0027] S400: Classify and identify pollution sources according to the temporal variation characteristics of the pollution sources, the spatial location variation characteristics of the pollution sources, and the reproduction and flow characteristics of the microbial pollution sources, and determine the type and location of the pollution sources.

[0028] Specifically, according to the biological characteristics of each microbial pollution source, the reproduction flow characteristics of the microbial pollution source are fitted through computer simulation. Based on the reproduction flow characteristics obtained by simulation, the trajectory of the pollution source expansion over time is predicted, and the position of the pollution source at different time points is recorded to form a diffusion trajectory. Using the predicted diffusion trajectory, the original pollution source time change characteristics and spatial position change characteristics are corrected to restore the true pollution source location to more accurately reflect the actual behavior of the pollution source. The time change characteristics of the pollution source after correction and restoration are analyzed to identify different change characteristic patterns, including continuous change characteristics, periodic change characteristics, and sudden change characteristics.

[0029] The spatial density cluster analysis is performed on the corrected and restored spatial position change characteristics of the pollution sources, and clustering is performed according to the spatial density to obtain the spatial distribution characteristics of the pollution sources. The change characteristic pattern and the spatial distribution characteristics of the pollution sources are used as input data, and the type of pollution source (such as agricultural pollution, industrial pollution, etc.) and the location of the pollution source (such as specific geographical coordinates, pollution area, etc.) are automatically identified through the pre-trained recognition model. The pre-trained recognition model outputs the type of pollution source (such as agricultural pollution, industrial pollution, domestic pollution, etc.) and its specific location based on the input pollution source characteristics. The location can be a specific coordinate point (such as longitude and latitude) or a regional range. By combining multiple factors such as time, space and biological characteristics, the type and location of the pollution source can be accurately identified, which helps to improve the accuracy of pollution source monitoring and management.

[0030] Furthermore, the present application S200 also includes:

[0031] The monitoring position of each water body monitoring data is located according to the coordinates, a position coordinate system is constructed, and the position direction relationship is marked; the time series arrangement relationship is fitted into the position coordinate system, and a time series visual window is constructed.

[0032] Specifically, based on the location collection information, coordinate positioning is performed for each monitoring point to determine the coordinate points of each water body monitoring data. Based on the coordinate points of each monitoring point, the longitude and latitude coordinates of each monitoring point are converted into points in the coordinate system to construct a position coordinate system, which is usually a two-dimensional coordinate system. In the coordinate system, arrows or other symbols are used to indicate the directional relationship between monitoring points.

[0033] Combine the time series relationship after sorting the time collection information with the corresponding position coordinate system, and display the data changes of different monitoring points along the time axis. In the position coordinate system, add the time dimension to each monitoring point to form a three-dimensional coordinate system, and use time series visualization tools such as time sliders and animations to show the changes of data over time. Present spatiotemporal data (position and time) in a visual way, usually using visualization tools such as charts, heat maps, and time axes to help users quickly understand the data. In the time series visual window, you can clearly see the changes of monitoring data over time, as well as the data comparison between different monitoring points. By combining the position coordinate system with the time series, complex water monitoring data can be converted into easy-to-understand spatiotemporal images to help users discover potential sources of pollution and changing trends.

[0034] Further, as attached Figure 2 As shown, the present application S300 includes:

[0035] The microbial detection data is labeled with pollution sources; based on coordinate positioning, the time-series collected data at the same spatial position is extracted according to the labeled pollution source, and longitudinal pollution source feature fitting is performed to obtain the longitudinal time variation characteristics of the labeled pollution source as the pollution source time variation characteristics; the same time gradient alignment extraction is performed according to the labeled pollution source, and the horizontal pollution source feature fitting is performed on the collected data at the extracted spatial position to obtain the horizontal spatial position variation characteristics of the labeled pollution source as the pollution source spatial position variation characteristics.

[0036] Specifically, for microbial detection data, possible pollution sources are identified and marked and classified. Pollution source marking is to determine the location and nature of the pollution source by analyzing the abnormal concentration or specific types of microorganisms in the water. For example, if a higher concentration of pathogenic bacteria or algae blooms (such as algae) appears in the water at certain monitoring points, it may indicate that there is a pollution source in the area. Through this marking, the preliminary location of the pollution source can be clearly identified.

[0037] According to the coordinate positioning of the marked pollution source, all time series data collected at the same spatial position of the marked pollution source are extracted. Time series data refers to the data collected at different time points at this location, which can show the changes of the pollution source over time. Fitting analysis is performed on the extracted time series data to obtain the characteristics of the pollution source changing over time. Longitudinal pollution source feature fitting is to analyze and fit the changing trend of the pollution source in the time dimension to determine the characteristics of the pollution source changing over time, such as the increase or decrease in pollutant concentration. Longitudinal fitting can reveal the time change law of the pollution source, such as whether the pollution source increases or decreases in a certain period of time. According to the longitudinal pollution source feature fitting, the time change characteristics of the pollution source are obtained, including the start time, duration, intensity change, etc. of the pollution source.

[0038] According to the location information in the water body monitoring data, combined with the geographical topography of the area where the water body is located and the flow characteristics of the water flow, the terrain and water flow influence relationship of the water body location is established. Using the established terrain and water flow influence relationship, the temporal influence of each collection location is deduced and analyzed, and the propagation path and time of pollutants in the water body are simulated to determine the impact time period. Taking the impact time period as the time gradient, the microbial detection data at different time points and different spatial locations are aligned according to the marked pollution source. According to the time gradient, the spatial location collection data related to the pollution source is extracted, and the extracted spatial location collection data is horizontally fitted and analyzed to reveal the change law of the pollution source at different spatial locations, and determine the distribution and change characteristics of the pollution source in space. Through vertical and horizontal fitting, the change law of the pollution source in time and space can be revealed, and the pollution source can be dynamically monitored and predicted to ensure real-time tracking of the change and impact range of the pollution source.

[0039] Furthermore, the present application also includes the following steps:

[0040] Establish the influence relationship between the terrain and water flow of the water body location; deduce and analyze the temporal influence of each collection location based on the terrain and water flow influence relationship to determine the influence time period; use the influence time period as the time gradient, align the microbial data according to the marked pollution source, and extract the collection data of the spatial location; based on the collection data of the spatial location, perform lateral fitting of the change characteristics of each microbial data within the same time gradient.

[0041] Specifically, the geographical topography of the area where the water body is located and the flow characteristics of the water flow are analyzed to determine the relationship between the terrain and water flow, that is, the relationship between the geographical topography of the water body (such as ups and downs, river bends, etc.) and the flow of water. Factors such as topography, river bends, and changes in the depth of lakes will affect the speed and direction of the water flow. The water flow is affected by the terrain, and the flow rate, flow direction, and dynamic changes of the water body will vary due to the terrain. According to the relationship between the terrain and water flow, the water flow impact characteristics of different water body locations at different time points are analyzed to deduce the diffusion trend of the pollution source in time. Through time series data, the specific impact period of water flow on the diffusion of pollution sources can be determined. Time series impact deduction analysis refers to the analysis of the water flow influencing factors at different time points based on time series data, the change law of water flow and pollution sources in different time periods, and the diffusion or impact process of pollution sources in time. The impact time period is the time period of the influence relationship between water flow and terrain on the diffusion, propagation or change of pollution sources. Water flows with different terrains, different flow rates, and different time periods may affect the diffusion range and concentration of pollution sources.

[0042] The impact time period represents the time required for the pollution source to spread from one point to another. Using the determined impact time period as the time gradient, the microbial detection data at different time points and different spatial locations are aligned. In other words, the microbial detection data at different times and spatial locations are unified into the same time scale or spatial coordinate system so that the data can be compared and analyzed with each other. According to the time gradient, the spatial location collection data related to the pollution source is extracted, that is, the water body data collected at different spatial locations, including information such as microbial concentration and pollution source. The extracted spatial location collection data is horizontally fitted and analyzed to reveal the changing laws of pollution sources at different spatial locations. The microbial data at multiple spatial locations in the same time period are fitted and analyzed to reveal their laws of spatial changes over time, such as the diffusion or concentration changes of pollution sources in space. Through the influence of terrain and water flow and time series analysis, the diffusion time and path of pollution sources can be accurately deduced, thereby improving the accuracy of pollution source identification.

[0043] Further, the present application S400 includes:

[0044] The reproduction and flow characteristics of the microbial pollution source are used to correct and restore the time change characteristics and the spatial position change characteristics of the pollution source; the change characteristic pattern is identified based on the corrected and restored time change characteristics of the pollution source, and the change characteristic pattern includes continuous change characteristics, periodic change characteristics, and sudden change characteristics; the pollution source spatial density is clustered based on the corrected and restored spatial position change characteristics of the pollution source to obtain the spatial distribution characteristics of the pollution source; the change characteristic pattern of the pollution source and the spatial distribution characteristics of the pollution source are used as input data, and the pollution source type is identified and the pollution source position is located through a pre-trained recognition model, and the pollution source type and position location are output.

[0045] Specifically, the reproduction mobility of each microbial pollution source is determined based on its characteristics, such as reproduction speed and survival conditions. Combining the relationship between topography and water flow and the reproduction mobility of microorganisms, a computer model is used for simulation fitting to reveal the behavior patterns of microbial pollution sources under different environmental conditions. Using tools such as particle tracking or diffusion models, the diffusion trajectory of pollution sources over time is predicted based on the simulation fitting results. Based on the diffusion trajectory prediction results, the initial pollution source time change characteristics and spatial position change characteristics are corrected, which helps to determine the true source and diffusion process of the pollution source.

[0046] The time-varying characteristics of the pollution sources after correction and restoration are used for change characteristic pattern recognition. By analyzing the time-varying characteristics of the pollution sources, different change patterns are identified, including continuous change characteristics (such as the gradual expansion of pollution), periodic change characteristics (such as periodic fluctuations in pollution) and sudden change characteristics (such as sudden pollution events). Continuous change characteristics refer to the gradual increase or decrease of pollution source concentration over time, periodic change characteristics refer to the regular fluctuation of pollution source concentration over time, and sudden change characteristics refer to the sudden increase or decrease of pollution source concentration in a short period of time.

[0047] Perform spatial density cluster analysis on the corrected and restored spatial position change characteristics of the pollution source, calculate the density of the spatial position, perform cluster analysis on the spatial position of the pollution source, and determine the spatial distribution pattern of the pollution source in the water body. Input the change characteristic pattern of the pollution source and the spatial distribution characteristics of the pollution source into the pre-trained recognition model, and output the type and location identification results of the pollution source. The pre-trained recognition model infers the type of pollution source (such as industrial pollution, agricultural pollution, etc.) and the specific location of the pollution source (such as a coordinate point in the water area or a pollution range area) based on the input change characteristic pattern and spatial distribution characteristics. Through the recognition of time-varying characteristic patterns, the changing trend of the pollution source is clarified, and it is identified whether the pollution source is continuously changing, periodically changing or suddenly changing, so as to realize the intelligent identification of the type and location of the pollution source, and improve the real-time and accuracy of pollution control.

[0048] Furthermore, the present application also includes the following steps:

[0049] According to the reproduction characteristics of each microbial pollution source, the reproduction fluidity is determined; according to the influence relationship of the terrain and water flow and the reproduction fluidity, simulation fitting is performed to obtain the reproduction flow characteristics of the microbial pollution source; according to the reproduction flow characteristics of the microbial pollution source, the pollution source diffusion trajectory is predicted, and the time change characteristics and spatial position change characteristics of the pollution source are corrected and restored using the diffusion trajectory prediction results.

[0050] Specifically, according to the growth and reproduction characteristics of each microorganism in the water body, its mobility under different conditions is determined, including the growth rate of the microorganism, the mode of transmission (such as water flow, natural movement, etc.) and the mobility in the water flow. The reproduction characteristics of microbial pollution sources refer to the laws of growth and reproduction of microorganisms in water bodies, including their reproduction rate, the influence of environmental conditions (such as water temperature, dissolved oxygen, etc.) and their ability to diffuse, which determine their distribution and changes in the water body. Based on the relationship between the reproduction mobility of microorganisms and the influence of terrain and water flow, simulation fitting is performed, and the diffusion path and change trend of microbial pollution sources in water bodies are established through numerical simulation or simulation methods.

[0051] According to the simulation results, the growth and migration characteristics of water flow and microorganisms are taken into account to predict the diffusion trajectory of pollution sources in water bodies over time. The diffusion path of pollution sources in space is simulated through particle tracking method or diffusion model, and the diffusion range of pollution sources in the future is predicted. Particle tracking method is a simulation method that simulates the diffusion and transmission process of substances by tracking the movement trajectory of single or multiple particles in a fluid. In the prediction of water pollutant diffusion, particle tracking method is used to simulate the movement and diffusion of pollutants (such as microorganisms, chemicals, etc.) with water flow and analyze their propagation path. Diffusion model is a mathematical model used to describe the diffusion of substances in a medium. In the prediction of water pollutant diffusion, the diffusion model mainly simulates the diffusion process of pollutants through mathematical formulas and algorithms.

[0052] Taking the particle tracking method as an example, the relationship between terrain and water flow and the initial position of the pollution source are input, and the movement trajectory of each particle is tracked by releasing multiple virtual particles randomly or at a specified initial position in the water body. The movement of particles is affected by the water flow, and over time, the particles will diffuse downstream or to other areas. The final position and number of particles can be used to determine the path and range of the diffusion of pollution sources. The particle tracking method can simulate the diffusion behavior of pollution sources in complex water bodies in detail, and is particularly suitable for areas with complex water flows. The prediction of the diffusion trajectory of pollution sources refers to predicting the diffusion path of pollution sources by establishing a mathematical model, combining water flow characteristics and microbial reproduction characteristics.

[0053] According to the diffusion trajectory prediction results, the time variation characteristics of the pollution source and the spatial position variation characteristics of the pollution source are corrected or adjusted to make them closer to the actual situation. The restoration process helps to determine the true location of the pollution source and its diffusion pattern, and further improve the accuracy of pollution source identification and control. For example, if the simulation shows that the pollution source diffuses to a certain area at a certain point in time, the time variation curve and spatial position model of the pollution source can be adjusted according to this result to make it more in line with the actual situation. Through simulation fitting and diffusion trajectory prediction, the diffusion path and time variation of the pollution source can be accurately predicted, the true location and changes of the pollution source can be accurately restored, and the accuracy of pollution source positioning can be improved.

[0054] Furthermore, the present application also includes the following steps:

[0055] According to the sample data set of microbial pollution sources of historical water bodies, a training data set and a test data set are constructed, and the training data set and the test data set both include the temporal change characteristic pattern of the microbial pollution source, the spatial distribution characteristics and the pollution source type label, and the pollution source location positioning label; a model architecture is built based on a long short-term memory network, and the network model is trained and converged through the training data set and the test data set to obtain the pre-trained recognition model, which is used to identify the pollution type and locate the position according to the change characteristic pattern of the input pollution source and the spatial distribution characteristics of the pollution source, and output the pollution source type and location positioning recognition results.

[0056] Specifically, a data set of historical microbial pollution source samples of water bodies is collected, and their temporal variation characteristic patterns, spatial distribution characteristics, pollution source type labels, and pollution source location labels are annotated. The data set of microbial pollution source samples is cleaned, missing values ​​and outliers are processed, and data is standardized. The data set is divided into a training data set and a test data set. A model architecture is built based on a long short-term memory network, including an input layer, an LSTM layer, a fully connected layer, and an output layer. The input layer receives the temporal variation characteristic patterns and spatial distribution characteristics of the pollution source from the training data set as input. The temporal variation characteristics and spatial characteristic data are appropriately preprocessed to form an input form suitable for the LSTM model. The LSTM layer is used to process time series data and extract time-dependent features. LSTM has the ability to remember long time series, so it is particularly effective in capturing the temporal variation characteristics of pollution sources. The output of the LSTM layer will be passed to the fully connected layer, which is used to combine temporal and spatial characteristics to output the prediction results of the pollution source type and pollution source location. The output layer predicts the pollution source type (such as industrial pollution, agricultural pollution, and domestic pollution) and location (such as specific coordinates) based on the labels of the training data set.

[0057] The LSTM model is trained using the training data set. During the training process, the model gradually reduces the error between the predicted results and the actual labels by adjusting the network parameters, and finally converges the model. An appropriate loss function, such as mean square error, is used to optimize the model so that the predicted pollution source type and location are as close to the actual value as possible. During the training process, the model is verified using the test data set to evaluate the generalization ability of the model on unseen data. The model is further optimized based on the test results, and hyperparameters such as learning rate and number of LSTM layers are adjusted to ensure the accuracy and robustness of the model. After the training is completed, the pre-trained LSTM model is used to identify the type and locate the position of the input new pollution source data (including time variation characteristics and spatial distribution characteristics). The type of pollution source (such as industrial, agricultural, domestic pollution source, etc.) and the specific location of the pollution source (for example, the coordinate position of the water area or the pollution area) are output. Model training through long short-term memory network can effectively capture the time series characteristics of the pollution source and accurately identify the type of pollution source (industrial, agricultural, domestic pollution, etc.).

[0058] In summary, the method for intelligently identifying water microbial pollution sources provided by this application has the following technical effects:

[0059] By connecting to the monitoring platform, water body monitoring data is obtained, and the water body monitoring data includes location collection information and time collection information; the water body monitoring data is integrated and arranged in time and space according to the location collection information and time collection information, and the spatiotemporal relationship between the collected data is established; the microbial detection data of the water body monitoring data is obtained, and based on the spatiotemporal relationship between the collected data, the time change relationship and spatial position change relationship of the pollution source are fitted to the microbial detection data to determine the time change characteristics of the pollution source and the spatial position change characteristics of the pollution source; according to the time change characteristics of the pollution source and the spatial position change characteristics of the pollution source, and in combination with the reproduction and flow characteristics of the microbial pollution source, the pollution source is classified and identified, and the type and location of the pollution source are determined. In other words, by combining the collected location collection information and time collection information, integrating and arranging in time and space, establishing the spatiotemporal relationship between the data, combining the reproduction and flow characteristics of the microorganisms, the pollution source is classified and identified, and different types of pollution sources are accurately classified, and their spatial positions are located, thereby improving the accuracy of identifying water body microbial pollution sources.

[0060] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0061] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A method for intelligently identifying water microbial pollution sources, characterized in that: include: Connecting to the monitoring platform to obtain water body monitoring data, wherein the water body monitoring data includes location collection information and time collection information; Integrate and arrange the water body monitoring data in time and space according to the location collection information and time collection information, and establish a spatiotemporal relationship between the collected data; Acquire the microbial detection data of the water body monitoring data, and based on the spatiotemporal relationship between the collected data, fit the time variation relationship and spatial position variation relationship of the pollution source to the microbial detection data to determine the time variation characteristics and spatial position variation characteristics of the pollution source; Based on the temporal variation characteristics of the pollution source, the spatial location variation characteristics of the pollution source, and combined with the reproduction and flow characteristics of the microbial pollution source, the pollution source is classified and identified to determine the type and location of the pollution source.

2. The method for intelligently identifying water microbial pollution sources according to claim 1, characterized in that: The water body monitoring data is arranged in time and space according to the location collection information and the time collection information, and the spatiotemporal relationship between the collected data is established, including: According to the position acquisition information, coordinate positioning is performed to identify the position and direction relationship; According to the time collection information, a time sequence arrangement relationship is obtained according to the time interval length of the collection timestamps; Establish a mapping association between the water body monitoring data and the position direction relationship and the time series arrangement relationship to obtain the spatiotemporal relationship between the collected data.

3. The method for intelligently identifying water microbial pollution sources according to claim 2, characterized in that: Also includes: Locate the monitoring position of each water body monitoring data according to the coordinates, construct a position coordinate system, and mark the position direction relationship; The time series arrangement relationship is fitted into the position coordinate system to construct a time series visual window.

4. The method for intelligently identifying water microbial pollution sources according to claim 2, characterized in that: Fitting the time variation relationship and spatial position variation relationship of the pollution source to the microbial detection data to determine the time variation characteristics and spatial position variation characteristics of the pollution source, including: Marking the contamination sources of the microbial detection data; Based on coordinate positioning, extract the time series data of the same spatial position according to the marked pollution source, perform longitudinal pollution source feature fitting, and obtain the longitudinal time variation characteristics of the marked pollution source as the pollution source time variation characteristics; The same time gradient alignment extraction is performed according to the annotated pollution source, and the lateral pollution source feature fitting is performed on the collected data of the extracted spatial position to obtain the lateral spatial position change feature of the annotated pollution source as the spatial position change feature of the pollution source.

5. The method for intelligently identifying water microbial pollution sources according to claim 4, characterized in that: Performing the same time gradient alignment extraction according to the marked pollution source, and performing lateral pollution source feature fitting on the collected data of the extracted spatial position, including: Establish the relationship between topography and flow influence of water body location; Determine the time period of influence based on the temporal influence of the terrain and water flow relationship on each acquisition location; Using the impact time period as the time gradient, aligning the microbial data according to the marked pollution source, and extracting the collected data of the spatial position; Based on the collected data at the spatial position, a lateral fitting of the change characteristics of each microbial data within the same time gradient is performed.

6. The method for intelligently identifying water microbial pollution sources according to claim 5, characterized in that: According to the time variation characteristics of the pollution source, the spatial location variation characteristics of the pollution source, and combined with the reproduction and flow characteristics of the microbial pollution source, the pollution source is classified and identified to determine the type and location of the pollution source, including: Using the propagation flow characteristics of the microbial pollution source, the temporal variation characteristics and the spatial position variation characteristics of the pollution source are corrected and restored; According to the corrected and restored pollution source time change characteristics, change characteristic patterns are identified, wherein the change characteristic patterns include continuous change characteristics, periodic change characteristics, and sudden change characteristics; Performing spatial density clustering of pollution sources according to the corrected and restored spatial position change characteristics of the pollution sources to obtain spatial distribution characteristics of the pollution sources; The changing characteristic pattern of the pollution source and the spatial distribution characteristics of the pollution source are used as input data, and the pollution source type and pollution source location are identified through a pre-trained recognition model, and the pollution source type and location are output.

7. The method for intelligently identifying water microbial pollution sources according to claim 6, characterized in that: The time variation characteristics and spatial position variation characteristics of the pollution source are corrected and restored by using the reproduction flow characteristics of the microbial pollution source, including: Determine the reproduction mobility based on the reproduction characteristics of each microbial pollution source; Perform simulation fitting based on the relationship between the terrain and water flow and the reproduction flowability to obtain the reproduction flow characteristics of the microbial pollution source; According to the reproduction and flow characteristics of the microbial pollution source, the pollution source diffusion trajectory is predicted, and the time change characteristics and spatial position change characteristics of the pollution source are corrected and restored using the diffusion trajectory prediction results.

8. The method for intelligently identifying water microbial pollution sources according to claim 6, characterized in that: The method of identifying the pollution source type and locating the pollution source position by using the pre-trained recognition model and outputting the pollution source type and position positioning includes: Based on the historical water body microbial pollution source sample data set, a training data set and a test data set are constructed, wherein the training data set and the test data set both include the temporal variation characteristic pattern of the microbial pollution source, the spatial distribution characteristics and the pollution source type label, and the pollution source location positioning label; A model architecture is built based on a long short-term memory network, and the network model is trained and converged through the training data set and the test data set to obtain the pre-trained recognition model. The pre-trained recognition model is used to identify the pollution type and locate the position according to the changing characteristic pattern of the input pollution source and the spatial distribution characteristics of the pollution source, and output the pollution source type and position identification results.

Images