Intelligent detection method and system for key risk object sample in water source

Abstract

The invention relates to the field of water source detection, in particular to a water source region key risk object sample intelligent detection method and system. Judging whether the sample detection data of the water source meets abnormal conditions or not; when the water source meets the abnormal condition, analyzing the sample detection data, identifying a target risk parameter and a sample position according to the sample detection data, generating risk data based on the target risk parameter and the sample position, and sending the risk data to the management end; when the water source does not meet the abnormal condition, extracting target features of the sample detection data; predicting the water quality change trend of the water source according to the target characteristics; judging whether the water quality meets an early warning condition or not according to a prediction result; and when the water quality meets an early warning condition, generating a water quality treatment strategy and early warning information based on a prediction result, and sending the strategies and information to a management end. Through multi-dimensional risk assessment, errors caused by limitation of the sensor are reduced.

Description

Technical Field

[0001] The present application relates to the field of water source detection technology, and in particular to a method and system for intelligent detection of key risk object samples in water sources. Background Art

[0002] The process of monitoring and analyzing key pollutants that may harm water quality and aquatic ecology at water sources (such as rivers, lakes, reservoirs, etc.). Its purpose is to sample and test potential pollutants in the water source environment. These pollutants may be introduced into the water body by discharge sources such as industry, agriculture, and domestic sewage, or they may be harmful substances that occur naturally in the water body.

[0003] In the prior art, water quality monitoring methods rely on sensors to directly measure whether the content of risk substances (such as heavy metals, chemical pollutants, dissolved oxygen, etc.) in water sources exceeds the standard. Although this method can obtain water quality data in real time, since the concentration of pollutants in water sources is greatly affected by environmental conditions (such as temperature, pH value, water flow rate, precipitation, etc.), sensors are easily affected by errors, interference or reaction lags, and sensor detection results often have certain errors. For example, the concentration of certain pollutants may fluctuate rapidly or show a gradual change trend in a short period of time, and sensors often cannot capture these subtle changes in time. For another example, when extreme weather events occur, environmental characteristics such as temperature, humidity, and rainfall may change dramatically in a very short period of time. Such changes may cause instantaneous fluctuations in water quality parameters, causing water source detection data to exceed fixed thresholds in a short period of time, but this does not necessarily mean that the water quality of the water source is abnormal, but is caused by environmental disturbances. If the detection model only relies on a fixed threshold and fails to effectively identify and adapt to these extreme environmental changes, it may lead to false alarms or missed alarms.

[0004] Therefore, the prior art has defects and needs to be improved. Summary of the invention

[0005] In order to solve one or several problems in the prior art, the main purpose of this application is to provide an intelligent detection method and system for key risk object samples in water sources.

[0006] In order to achieve the above-mentioned invention object, the present application proposes a method for intelligent detection of key risk object samples in water source areas, the method comprising:

[0007] Obtain sample test data of water sources in real time;

[0008] Determining whether the sample detection data of the water source meets the abnormal condition;

[0009] When the water source meets the abnormal condition, the sample detection data is parsed, the target risk parameter and the sample location are identified according to the sample detection data, and risk data is generated based on the target risk parameter and the sample location and sent to the management end;

[0010] When the water source does not meet the abnormal condition, extracting the target features of the sample detection data;

[0011] Predicting the water quality change trend of the water source based on the target feature;

[0012] Determine whether the water quality meets the early warning conditions according to the prediction results;

[0013] When the water quality meets the warning condition, a water quality treatment strategy and warning information are generated based on the prediction result and sent to the management end.

[0014] The present application also provides a water source key risk object sample intelligent detection system, including:

[0015] An acquisition module is used to obtain sample detection data of water sources in real time;

[0016] A judgment module is used to judge whether the sample detection data of the water source meets the abnormal condition;

[0017] An analysis module, used for analyzing the sample detection data when the water source meets the abnormal condition, identifying the target risk parameter and the sample location according to the sample detection data, generating risk data based on the target risk parameter and the sample location and sending it to the management end;

[0018] An extraction module, used for extracting target features of the sample detection data when the water source does not meet the abnormal condition;

[0019] A prediction module, used to predict the water quality change trend of the water source based on the target feature;

[0020] A second judgment module is used to judge whether the water quality meets the warning condition according to the prediction result;

[0021] The sending module is used to generate a water quality treatment strategy and early warning information based on the prediction result and send them to the management end when the water quality meets the early warning condition.

[0022] The present application also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the above methods when executing the computer program.

[0023] The present application also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of any of the above-mentioned methods are implemented.

[0024] The intelligent detection method and system of key risk samples of water sources in the embodiments of the present application, through real-time monitoring and data analysis, can not only detect the current water quality status, but also predict the future trend of water quality changes. This forward-looking judgment enables the system to identify potential water quality problems in advance, avoiding the possible response lag in traditional monitoring methods, thereby providing managers with more accurate risk assessment and early warning information. By extracting features from sample data, combining historical data and environmental factors, a prediction model for water quality changes can be generated in real time. This predictive capability significantly improves the response speed of the monitoring system, allowing managers to take action in advance and effectively reduce the risks caused by fluctuations in pollutant concentrations. By integrating multiple data sources (such as water quality sample test results, environmental factor data, etc.), multi-dimensional risk assessment is achieved. This data fusion method avoids the limitations of a single sensor due to environmental interference or technical limitations, and can more comprehensively and accurately reflect the true water quality of the water source, reducing the errors caused by sensor limitations. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 A schematic diagram of a flow chart of a method for intelligently detecting samples of key risk objects in a water source area according to an embodiment of the present application;

[0026] Figure 2 A schematic diagram of a flow chart of a method for intelligently detecting samples of key risk objects in a water source area according to an embodiment of the present application;

[0027] Figure 3 This is a schematic block diagram of the structure of an intelligent detection system for key risk object samples in a water source area according to an embodiment of the present application;

[0028] Figure 4 A schematic block diagram of the structure of a computer device according to an embodiment of the present application.

[0029] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0031] Reference Figure 1 In an embodiment of the present application, a method for intelligently detecting samples of key risk objects in a water source is provided, and the method comprises:

[0032] S1. Obtain sample test data of water source in real time;

[0033] S2. Determine whether the sample detection data of the water source meets the abnormal condition;

[0034] S3. When the water source meets the abnormal condition, the sample detection data is parsed, the target risk parameter and the sample location are identified according to the sample detection data, and risk data is generated based on the target risk parameter and the sample location and sent to the management end;

[0035] S4. When the water source does not meet the abnormal condition, extracting the target features of the sample detection data;

[0036] S5, predicting the water quality change trend of the water source based on the target feature;

[0037] S6. judging whether the water quality meets the early warning condition according to the prediction result;

[0038] S7. When the water quality meets the warning condition, a water quality treatment strategy and warning information are generated based on the prediction result and sent to the management end.

[0039] As described in the above steps S1-S3, the water source can be continuously monitored by sensors, sampling equipment or remote monitoring systems to collect water quality sample data in real time. These data usually include physical and chemical parameters, such as pH, dissolved oxygen, turbidity, temperature, chemical oxygen demand (COD), ammonia nitrogen, etc. This real-time monitoring can be achieved through various technologies, such as sensor networks and Internet of Things (IoT) devices. By analyzing the water quality data collected in real time, it is determined whether the water quality deviates from the normal range. For example, statistical analysis methods, machine learning algorithms or threshold judgments can be used to identify abnormal conditions such as pollution and excessive concentration of pollutants. Abnormal conditions may include excessively high pollutant concentrations, sudden changes in water quality, or sudden events (such as release from upstream pollution sources). By identifying abnormal conditions in real time, potential pollution threats can be discovered in advance and emergency treatment measures can be taken in time. The accuracy of abnormal condition judgment will directly affect the reliability of the system and avoid frequent false alarms. Data analysis is performed on the test data to identify key parameters related to water quality changes, such as specific pollutants, heavy metals, toxic chemicals, etc. These key parameters can be dissolved oxygen, chemical oxygen demand (COD), nitrogen and phosphorus concentrations, etc. In addition, it is also necessary to combine the spatial location information of the water source to identify the pollution sources at different locations in order to more accurately determine the degree of pollution of the water source. By identifying specific risk parameters and sample locations, the area where the pollution source is located can be clearly identified, which is convenient for rapid response. After identifying the risk parameters, targeted emergency response measures can be formulated. It is not a simple "one size fits all", but different treatment measures are taken according to the type and location of the pollution source. Through the accurate identification of the target risk parameters and locations, the severity of the pollution can be effectively assessed and the quality of decision-making can be improved. For example, physical, chemical or biological treatment methods can be adopted according to the type of pollutants, which can help to formulate more appropriate treatment strategies. According to the target risk parameters and sample locations identified in the above steps, comprehensive risk data is generated. Risk data usually contains information such as the type of pollution source, concentration, and pollution area, and is sent to the management end through the data transmission system. Risk data can be presented to water quality monitoring managers in the form of charts, reports, etc. to help them understand the current risk level of water quality.

[0040] As described in the above steps S4-S7, when there is no obvious abnormality in the water quality data, the system will extract target features (such as the historical change trends of various water quality indicators) through data analysis methods (such as machine learning prediction models, time series analysis, etc.), and predict the water quality change trends based on these features. The prediction model can be based on historical data, weather changes, seasonal factors and other factors, combined with feature data to predict the future trend of water quality changes. Based on the prediction results of the previous step, it is judged whether the future water quality changes will reach the warning threshold. The warning conditions may include water quality indicators reaching a set dangerous level (such as a specific pollutant concentration exceeding the standard). The warning algorithm may combine historical data, trend prediction and warning thresholds, and use a rule engine or model to calculate whether the warning conditions are met. By automatically judging whether the warning conditions are met in combination with the prediction results, intelligent management can be achieved and manual judgment errors can be reduced. By predicting whether the water quality meets the warning conditions, treatment measures can be taken before the pollutant concentration increases to prevent water source pollution. Automatic judgment and warning can improve the system's adaptive ability and reduce human intervention. When the water quality meets the warning conditions, the system will generate a corresponding water quality treatment strategy based on the water quality prediction results and notify the management end through warning information. Water quality treatment strategies can include adding chemicals, adjusting water flow, activating pollution source control equipment, etc. According to the type and forecast of water quality problems, formulate appropriate water quality treatment strategies to minimize water pollution. Through real-time feedback of early warning information, managers can understand the problem as soon as possible and take countermeasures quickly.

[0041] As mentioned above, behavioral data and environmental data of poultry in the orchard are collected. This real-time nature ensures that farmers can obtain accurate information at any time, which helps to quickly identify any potential problems. It ensures that poultry health problems can be discovered at the earliest stage and avoid delays. Continuous high-frequency data collection improves the accuracy of monitoring and helps to accurately judge the behavior and physiological status of poultry. When the behavior or physiological data of poultry deviates from the normal range, abnormal behavior can be automatically judged and marked. This process combines the poultry's own data and environmental factors to efficiently and accurately identify health or environmental problems. Poultry abnormalities can be detected in time to avoid the spread of problems, such as disease transmission or environmental stress. It reduces the subjective judgment errors of farmers, and data-driven automated judgment is more reliable. Combined with environmental data (such as temperature, humidity, light, etc.) for comprehensive analysis, the root cause of abnormal behavior can be inferred. This multi-dimensional data fusion capability significantly improves the system's adaptability to complex situations. It avoids misjudgments that may be caused by a single data source and ensures the accuracy of abnormal behavior analysis results. For example, natural behavior changes caused by environmental changes are no longer misjudged as health problems. Through comprehensive analysis, the system can self-learn, improve the accuracy of abnormal behavior recognition, and gradually optimize the monitoring model. After confirming the cause of the abnormal behavior, the system can automatically generate targeted response strategies.

[0042] Reference Figure 2 In one embodiment, after the step of determining whether the sample detection data of the water source meets the abnormal condition and before the step of determining when the water source meets the abnormal condition, the method further includes:

[0043] S31, acquiring environmental data of the sample location, and extracting environmental features of the environmental data, wherein the environmental features include temperature features, humidity features, wind speed features, and rainfall features;

[0044] S32, constructing an error analysis model, inputting the environmental characteristics and sample detection data into the error analysis model, evaluating the interference coefficient of the environmental characteristics on the sample detection data within a specific time through the error analysis model, and outputting the evaluation result, wherein the interference coefficient is used to evaluate the amplitude of the water quality parameter deviation from the normal value caused by the environmental characteristics;

[0045] S33, based on the evaluation result, adjusting the detection threshold by the interference coefficient to obtain an adjusted detection threshold;

[0046] S34, judging whether the water source meets the abnormal condition according to the adjusted detection threshold;

[0047] S35. When the sample detection data is greater than the adjusted detection threshold, it is determined that the sample detection data of the water source meets an abnormal condition.

[0048] As mentioned in the above steps, temperature is a key factor affecting water quality parameters. For example, rising temperature increases the rate of consumption of dissolved oxygen in water, accelerates the rate of chemical reactions, and causes changes in some water quality indicators. Changes in temperature can also affect the activity of microorganisms in the water body, thereby affecting water quality. Some water quality testing equipment is very sensitive to temperature, so the readings of the testing equipment may be biased under different temperature conditions. If the impact of temperature on water quality is ignored, some water quality indicators may be misinterpreted. For example, a water body at a higher temperature may show a lower level of dissolved oxygen, but in fact this low value is caused by temperature and does not necessarily mean that the water quality has deteriorated. By inputting temperature data as an environmental feature into the error analysis model, its interference coefficient on water quality data can be evaluated. By dynamically adjusting the detection threshold, the error caused by temperature changes can be corrected to ensure more accurate water quality monitoring. Air humidity has a direct impact on evaporation, precipitation and water level changes in water bodies. Humidity changes usually cause fluctuations in water flow and water level in water sources, which affects the data stability of water quality monitoring. For example, when the air humidity is high, the evaporation of the water source is low, which may cause changes in the nutrient concentration of the water body. In addition, excessively high or low humidity may also lead to unstable performance of water quality monitoring instruments, thus affecting the accuracy of the test results. Changes in humidity may have a significant impact on water quality parameters in the short term. For example, under high humidity conditions, changes in water quality indicators may be misjudged as abnormal water quality. By taking humidity as an input feature, the error analysis model can identify potential errors caused by humidity changes and calculate the corresponding interference coefficient. By adjusting the detection threshold, the interference of humidity changes on water quality monitoring can be eliminated and misjudgments can be reduced. Wind speed has a direct impact on the surface fluctuations, water flow velocity and evaporation process of water bodies, especially in open waters. Changes in wind speed may affect water quality parameters such as oxygen solubility and suspended matter concentration in water bodies. For example, high wind speeds may cause pollutants in water bodies to float up, thereby affecting water quality monitoring data. Changes in wind speed may also interfere with the stability of monitoring equipment (such as buoys, sensors, etc.), resulting in deviations in collected data. High wind speeds may cause large fluctuations in the water surface, thereby affecting the stability and measurement accuracy of water quality sensors. If the impact of wind speed is ignored, water quality changes caused by wind speed may be misjudged as abnormal water quality. By inputting wind speed characteristics, the error analysis model can evaluate the impact of wind speed changes on water quality detection and calculate the interference coefficient. This process can help dynamically adjust the detection threshold to ensure that the impact of wind speed on water quality detection results is properly handled, thereby improving the accuracy of monitoring. The impact of rainfall on water quality: Rainfall directly affects the quality of water sources. Large amounts of precipitation can bring external pollutants (such as silt, chemicals, etc.), causing drastic changes in water quality, especially in a short period of time. For example, when there is too much precipitation, rainwater will wash the ground, bringing a large amount of pollutants, and water quality parameters such as suspended matter concentration and chemical oxygen demand may increase rapidly. Precipitation causes changes in water flow and the washing of pollutants, which may cause significant fluctuations in water quality detection data in the short term.If these factors are not taken into account after precipitation, fluctuations in water quality data may be mistakenly interpreted as abnormal water quality. The role of rainfall in error adjustment: By inputting rainfall data, the error analysis model can identify the short-term impact of precipitation on water quality, calculate the interference coefficient, and then determine whether the water quality is abnormal based on the adjusted detection threshold. This helps ensure that the water quality monitoring system can correctly identify water quality changes after precipitation and avoid misjudgments. In the error analysis model, the core goal of the model is to quantify the impact of various environmental characteristics on water quality test data. By combining environmental characteristics (temperature, humidity, wind speed, rainfall) with sample test data, the model can calculate the interference coefficient of each environmental characteristic on the test data. The interference coefficient measures the extent to which environmental changes cause water quality parameters to deviate from normal values. For example, an increase in temperature may lead to low dissolved oxygen, and the model can quantify this change as an interference coefficient. By evaluating the interference coefficient of environmental characteristics, the error analysis model can dynamically adjust the threshold of water quality detection. With the adjusted threshold, the system can more accurately identify water quality anomalies and avoid misjudgments caused by environmental factors. For example, if the wind speed is high, the system can raise the threshold to eliminate the impact of wind speed changes and ensure a more reliable judgment of water quality.

[0049] In one embodiment, the method for predicting the water quality change trend of the water source based on the target feature includes:

[0050] Obtain historical sample test data and corresponding historical environment data, and extract target features of the historical sample test data;

[0051] Performing time series analysis on the target characteristics and historical environmental data to identify influencing parameters and hysteresis characteristics of the historical environmental data on the target characteristics;

[0052] Obtain future environmental forecast data;

[0053] Acquire current sample detection data, input the current sample detection data, influencing parameters, hysteresis characteristics and environmental prediction data into a preset trend prediction model, predict the changing trend of water quality under future environmental conditions through the trend prediction model, and output the prediction result;

[0054] Based on the output results, the predicted water quality change trend is obtained.

[0055] As mentioned above, historical sample test data contains information about the changes in water quality indicators (such as dissolved oxygen, chemical oxygen demand, suspended solids concentration, etc.) over time. Extracting target features from these data can help understand the long-term trend and current status of water quality. Historical data provides trend lines and fluctuation patterns for model building, which can reveal past water quality changes. By using these target features, the model can perform pattern recognition on water quality changes, thereby more accurately predicting future water quality trends. Historical data can capture the seasonality and periodicity in water quality changes, helping the model to distinguish normal fluctuations from abnormal fluctuations. Environmental factors (such as temperature, humidity, precipitation, wind speed, etc.) have a significant impact on water quality changes. For example, rainfall may lead to increased pollution of water sources, and temperature changes may affect the dissolved oxygen level of water bodies. By analyzing these historical environmental data, it is possible to identify which environmental characteristics have a significant correlation with water quality changes, and reveal the lag effect between environmental changes and water quality indicators (that is, the impact of environmental changes on water quality may be delayed). The relationship between environmental factors and water quality may not take effect immediately. Changes in certain environmental characteristics (such as increased rainfall) may take some time to significantly affect water quality. Therefore, identifying these hysteresis effects can enable the model to more accurately predict water quality changes, rather than relying solely on current environmental data. Time series analysis can help us understand how environmental data changes over time and predict its impact on water quality by modeling the trends, seasonality, and periodicity of historical data. Hysteresis characteristics refer to the fact that changes in certain environmental data in the past may have an impact on current water quality. Time series analysis helps identify the time span and pattern of such impacts. By capturing hysteresis effects, time series analysis can help the model more accurately predict future water quality changes. For example, identifying the impact of rainfall in the past week on future water quality can improve the accuracy of predicting water quality trends and avoid ignoring the delayed impact of environmental factors. The purpose of obtaining future environmental forecast data is to incorporate upcoming environmental changes into water quality trend forecasts. For example, information such as future temperature and precipitation provided by weather forecasts can help the model understand upcoming environmental changes and adjust its predictions of water quality changes. Future environmental conditions are key inputs to the prediction model, helping the model identify the possible impact of short-term environmental fluctuations on water quality. By combining future environmental data, the model can make predictions about water quality changes in advance. Future predictions combined with environmental data can provide valuable information for water quality management and emergency response, and improve the timeliness and scientificity of decision-making. The current sample test data represents the actual status of water quality at the current moment and can reflect the real-time water quality status of the water source. Using this data as model input can ensure that the prediction results are more in line with the actual water quality conditions, and make reasonable inferences about future water quality changes in combination with other factors (such as environmental data and lag characteristics). The current test data provides immediate feedback on water quality, helping the model to compare and adjust the prediction results, and ensure that the trend of water quality changes reflects the actual situation.As an immediate input, it can effectively correct the prediction errors based on historical data. The model integrates historical data, environmental prediction data and current water quality data, and comprehensively analyzes the impact of various factors on water quality changes. Through training and optimization, the model can identify complex cause-and-effect relationships and predict the changing trend of water quality under specific environmental conditions in the future. Through the preset trend prediction model, the accuracy of the prediction of water quality changes can be maintained under dynamic environmental changes. This method avoids the limitations of traditional static models, can cope with complex natural environmental fluctuations, and improve the reliability of water quality monitoring systems.

[0056] In one embodiment, the method for judging whether the water quality meets the early warning condition according to the prediction result includes:

[0057] Obtaining the predicted water quality change trend;

[0058] Calculating the change range of the target characteristic according to the water quality change trend;

[0059] Based on the change amplitude, determining whether the change amplitude of the target feature is within a preset amplitude range;

[0060] When the change amplitude of the target characteristic exceeds the preset amplitude range, it is judged that the water quality meets the warning condition;

[0061] When the change amplitude of the target feature is within a preset amplitude range, calculating the change rate of the target feature;

[0062] Based on the change rate, determining whether the change rate of the target feature is within a preset rate range;

[0063] When the change rate of the target characteristic exceeds a preset rate range, it is determined that the water quality change trend meets the early warning condition.

[0064] As mentioned above, the basic data for early warning is provided by predicting the future trend of water quality changes (such as the rising and falling trends of indicators such as dissolved oxygen, pH value, and chemical oxygen demand). The trend of water quality changes is usually based on historical data, environmental data, and current test data, and is predicted through time series analysis, regression models, or machine learning methods. The predicted trend of water quality changes reflects the expected trend of water quality in a certain period of time in the future. By predicting the trend of water quality changes, it is possible to determine in advance whether there is a risk of water quality deteriorating. Trend prediction helps to identify potential water quality anomalies, prevent possible pollution or environmental disasters, take countermeasures in advance, and ensure the health and safety of water bodies. The prediction of water quality change trends provides a forward-looking perspective for the early warning system. If it is predicted that certain water quality indicators will exceed the standard or deteriorate, warnings can be issued in advance, monitoring can be carried out, or measures can be taken. By predicting future water quality changes, water quality management departments can make more scientific and reasonable decisions based on data, which can not only deal with current water quality problems, but also deal with potential water quality risks. The amplitude of change refers to the maximum fluctuation range of a water quality indicator (target feature) within the prediction period. For example, if a water quality indicator changes significantly, it may indicate unstable water quality or increased risk of pollution. This step evaluates the severity of water quality changes by calculating the maximum range of change of water quality indicators (such as dissolved oxygen concentration, pH value, etc.). The calculation of amplitude helps identify abnormal fluctuations in water quality. When the amplitude of water quality changes exceeds a preset range, it means that the water quality may have changed dramatically, which in turn indicates that the water quality may be polluted or at risk of other environmental disasters. The measurement of amplitude is an important indicator for determining whether the water quality is abnormal. If the amplitude of water quality changes exceeds the preset threshold, it means that the water quality has fluctuated dramatically, which is usually a signal of pollution or changes in environmental conditions. At this time, the system will issue an early warning to remind managers to take timely measures. Through amplitude calculation, water quality fluctuations can be quantitatively analyzed to help decision makers determine whether further investigation and treatment are needed to avoid missing possible water quality crises. This step determines whether the water quality change is within an acceptable range by judging the amplitude of the change. The preset amplitude range is set based on historical data experience, and it represents the fluctuation range of water quality under normal change conditions. If the amplitude of a certain indicator changes beyond this range, it means that the water quality is abnormal. The preset range helps distinguish between normal fluctuations and abnormal fluctuations. In the natural environment, water quality may have some normal fluctuations (such as seasonal changes), but if the amplitude of the change exceeds the normal range, it may mean pollution or other emergencies. By comparing the amplitude of the change with the preset range, it is possible to effectively determine whether there are abnormal changes in water quality and promptly identify possible water pollution or other abnormal events. This helps prevent environmental disasters and avoid health and ecological risks caused by water quality deterioration. The rate of change refers to the speed at which a water quality indicator (target feature) changes in a unit of time. Faster changes may indicate a rapid deterioration in water quality, such as a rapid increase in chemical oxygen demand or a sharp drop in dissolved oxygen.By calculating the rate of change, the dynamics of water quality changes can be more accurately determined. Some water quality changes may be temporary or slow, but some extreme pollution events or sudden environmental changes may cause water quality to deteriorate rapidly. The calculation of the rate of change can help us distinguish between slow changes and rapid changes, and identify abnormalities of rapid changes in a timely manner. The calculation of the rate of change helps detect rapid changes in water quality, especially in sudden pollution events or extreme climate conditions. By monitoring the rate of change of water quality, it is possible to respond quickly, issue early warnings in a timely manner, and avoid major environmental accidents. This step determines whether the water quality change is within a reasonable range by comparing the calculated rate of change with the preset rate range. The preset rate range is usually based on experience, historical data, and environmental characteristics. Rates outside the range mean that the water quality has changed abnormally, and corresponding control measures may need to be taken. The setting of the rate range helps to distinguish between normal water quality changes (such as slow fluctuations of seasonal changes) and abnormal water quality changes (such as sudden pollution, environmental changes, etc.). By combining the characteristics of water quality change trend, amplitude, and rate, it is finally determined whether the water quality meets the early warning conditions. If the amplitude or rate of change exceeds the preset range, it is determined that the water quality is abnormal and meets the early warning conditions. Relying solely on one feature may not fully reflect the true situation of water quality. Comprehensive judgment can help avoid inaccurate warnings caused by a single factor and ensure that the water quality management system is more efficient and reliable.

[0065] In one embodiment, the method further comprises:

[0066] Obtain historical sample test data and corresponding historical environment data, and extract target features of the historical sample test data;

[0067] Performing time series analysis on the target characteristics and historical environmental data to identify the change pattern of the target characteristics under conditions of different historical environmental data;

[0068] Based on the recognition results, a frequency analysis model is constructed to analyze the change pattern under different environmental data conditions, and the frequency of abnormal judgment of the sample detection data is determined according to the change pattern;

[0069] The current environment data is obtained, and the current environment data is input into the frequency analysis model, and the frequency analysis model is used to output the frequency of abnormal judgment of the sample detection data under the current environment data condition.

[0070] As mentioned above, historical sample test data are the test results of certain key indicators or signals, which may be sampled values ​​from certain devices, sensors or systems. They contain status information of the system or device at different time points. Historical environmental data refers to the external environmental conditions when these historical sample test data were collected. These environmental conditions may affect the performance of sample test data, including factors such as temperature, humidity, pressure, and light. Different environmental conditions may cause the performance of test data to be significantly different, so combining environmental data helps capture such changes. By combining historical samples and environmental data, the system can build a comprehensive view of target features, providing an accurate basis for subsequent analysis and modeling. Target features can refer to key indicators that can reveal data anomalies or specific patterns. For example, statistical features such as the mean, variance, maximum value, minimum value, and change trend of test data, or certain complex frequency domain features (such as frequency components, power spectrum, etc.). These features can usually be extracted using statistical methods, signal processing techniques (such as Fourier transform, wavelet transform, etc.) or machine learning algorithms. By extracting target features, the most critical information can be extracted from complex data, data redundancy can be reduced, and the focus can be focused on the part that best reflects system anomalies. The selection and extraction of target features are key to model accuracy, and these features determine the ability of subsequent analysis and prediction. The extracted features can help us identify the behavior patterns of sample data and provide more concise and representative data input for subsequent analysis. Time series analysis is to identify patterns and trends by modeling the patterns of data changes over time. In this case, time series analysis can find out how the target features fluctuate over time under different environmental conditions. In the case of multiple environmental conditions, the changes in target features may show different patterns. Through time series analysis, these patterns can be identified, which in turn provides a useful basis for modeling. Under different environmental conditions, the change patterns of target features may be different. Time series analysis helps to dig out the time dependencies hidden in the data and capture the dynamic characteristics of the system over time. By identifying the change pattern, more accurate anomaly judgment can be made under different environmental conditions. Through time series analysis, we can understand the dynamic change patterns under different environmental conditions, thereby providing strong support for subsequent anomaly detection and frequency analysis. Frequency analysis usually refers to finding the frequency pattern in the signal by analyzing the frequency components of the signal. In this case, the frequency analysis model will try to find the frequency distribution of sample detection data under different environmental conditions based on the identified change pattern. The model can be constructed using frequency domain analysis methods such as Fourier transform and wavelet transform, or using clustering algorithms and regression analysis in machine learning. The frequency distribution of data under environmental conditions can provide additional information. Through frequency analysis, periodic changes or sudden fluctuations in data can be found, which is particularly important for anomaly detection.If the frequency of the target feature is significantly different in different environments, the frequency analysis model can help identify abnormal and normal patterns, thereby improving the accuracy of the judgment. The frequency analysis model can reveal the periodic or regular changes in the data, provide additional basis for abnormal judgment, and help improve the accuracy of the model. The environmental data collected in real time refers to the external environmental conditions at the current moment, such as temperature, humidity, etc. The current environmental data is input into the trained frequency analysis model to predict the abnormal judgment frequency of the sample detection data under the current environmental conditions. The abnormal judgment frequency output by the frequency analysis model refers to the possibility or probability that the sample detection data is abnormal under the current environmental conditions. By inputting real-time environmental data into the model, dynamic abnormal judgment results can be obtained. The advantage of this method is that it can adapt to different environmental changes, monitor whether the data is abnormal in real time, and respond in time.

[0071] In one embodiment, after the step of generating a water quality treatment strategy based on the prediction result and sending the warning information to the management end, the method further includes:

[0072] Obtain water source sample testing data after water quality treatment strategy;

[0073] According to the time series, the water source sample test data after the water quality treatment strategy is compared with the initial water source sample test data to extract the change parameters of the target risk parameters;

[0074] According to the change parameter of the target risk parameter, judging whether the water quality treatment strategy meets the replacement condition;

[0075] If the change parameter is less than a preset change threshold, it is determined that the water quality treatment strategy meets the replacement condition.

[0076] As mentioned above, source water sample test data refers to the water quality test data collected after the water quality treatment strategy is applied. These data usually include multiple water quality indicators, such as pH, turbidity, dissolved oxygen, bacterial content, chemical oxygen demand (COD), etc. Water quality treatment strategies are usually aimed at improving water quality problems, such as removing harmful substances and adjusting the pH of water. Obtaining treated water data refers to retesting the water source after a specific treatment plan is implemented to obtain these indicator data for subsequent analysis. Time series analysis is to identify trends and patterns by comparing data changes at different time points. In this step, the source water sample data before and after treatment are compared. Initial source water sample data refers to the original water quality data before any treatment is implemented. The treated water sample data is the water quality data of the same period or close to the period after the water quality treatment. Target risk parameters These parameters refer to the key factors in water quality indicators, such as pH, dissolved oxygen, heavy metal content, pathogenic microorganism content, etc., which directly affect the safety of water quality and the health standards of drinking water. By comparing the initial water source data with the treated data, specific changes are extracted, which usually indicate an improvement or deterioration in water quality. The change parameter is the change in certain key water quality indicators in the sample data before and after treatment. For example, a reduction in certain indicators indicates that the water quality treatment is effective, otherwise there may be problems. If the change parameter does not meet the preset standard or threshold, the water quality treatment strategy may need to be replaced. For example, if the concentration of certain pollutants is still high after treatment, or if certain important water quality indicators have not changed significantly, it indicates that the current treatment strategy may be ineffective or ineffective. Monitor the changes in water quality indicators to help determine the effectiveness of the current treatment strategy. If the target risk parameter does not change much after water quality treatment, it means that the water quality improvement is not obvious, and the strategy may need to be adjusted. If the change parameter is less than the threshold, it means that the water quality improvement is not obvious, and the treatment strategy may need to be adjusted or replaced. These thresholds can be set based on water quality safety standards, actual water quality improvement needs, or historical experience. The setting of thresholds must be scientific and reasonable to ensure that problems in water quality treatment can be discovered in a timely manner.

[0077] In one embodiment, before the step of determining whether the sample detection data of the water source meets the abnormal condition, the method further includes:

[0078] Identifying target risk parameters of the sample test data;

[0079] Identify the target risk parameters according to their categories, and perform overlapping signal detection according to each category of target risk objects;

[0080] According to the result of the overlapping signal detection, determining whether the target risk object has an interference;

[0081] When there are interferents on the target risk object, identifying the type of the interferents;

[0082] Separating the overlapping signals according to the type of the interferents, and inferring the content of the interferents according to the signals of the interferents;

[0083] Generate compensation parameters for target risk substances according to the content of interfering substances;

[0084] The content of the target risk substance is adjusted based on the compensation parameter.

[0085] As mentioned above, target risk parameter identification is to determine which water quality indicators are key and potentially risky parameters in the water quality sample test data. For example, heavy metals, bacterial content, chemical oxygen demand (COD), etc. may be target risk parameters that need special attention. Identifying target risk parameters is the first step in data analysis. It helps focus on the most critical indicators, so that we can concentrate on in-depth analysis and processing of these parameters. By accurately identifying target risk parameters, the system can process water quality monitoring data in a targeted manner to ensure that key indicators that may pose a threat to water quality safety and human health are paid attention to. According to the characteristics of target risk parameters, they are divided into different categories. For example, some indicators may be related to toxic chemicals, while others may be related to microorganisms or physical properties (such as turbidity). During the measurement process, the signals of some risk objects may overlap with each other, resulting in confusion in the measurement data. For example, some chemicals or microorganisms may affect the same water quality parameter at the same time. Overlapping signals may lead to misunderstanding or misjudgment of data, so it is necessary to identify and analyze such mixed signals by detecting overlapping signals to ensure the accuracy of the data. Overlapping signal detection can improve the accuracy of data analysis, avoid interference between different substances, and ensure the correct identification and analysis of target risk objects. By detecting overlapping signals, the system can determine whether there are other substances (interfering substances) that affect the detection of the target risk object. These interfering substances may cause the signal reading to deviate from the true value. In actual measurements, other chemicals or physical factors in the environment may interfere with the detection of the target substance. Therefore, identifying and distinguishing interfering substances is crucial to ensure the accuracy of the measurement results. Identifying interfering substances can help eliminate the source of error, thereby improving the credibility and accuracy of the test results. After confirming the existence of interfering substances, the next step is to determine the specific type of interfering substance. For example, certain organic compounds or minerals may affect the detection of heavy metals. Different types of interfering substances may have different effects on the signal, and identifying the type of interfering substance can provide necessary information for subsequent signal separation and compensation. Identifying the type of interfering substance can help accurately locate the root cause of the problem and provide effective guidance for subsequent processing and correction. Once the interfering substance and its type are identified, the next step is to separate the interfering signal from the target signal through a specific algorithm or signal processing technology. The method may include frequency analysis, filtering and other technologies. Through the separated interfering signal, combined with a certain mathematical model and experimental data, the content of the interfering substance can be inferred. Signal separation and inferred content help to accurately assess the true content of the target risk object and avoid the influence of interfering substances on the data. This allows for a more accurate recovery of the actual content of the target risk substance. By separating the interference signal and estimating the content of the interfering substance, the original data can be accurately corrected to ensure that the estimated content of the target risk substance is closer to the true value. Once the content of the interfering substance is estimated, compensation parameters can be generated based on this information to correct the content of the target risk substance.The presence of interferents may cause the measured value of the target risk object to be underestimated or overestimated, so compensation needs to be made according to the content of the interferents to ensure that the final data accurately reflects the water quality. By generating compensation parameters, the measurement errors caused by interferents can be effectively corrected and the accuracy of the test results can be improved. The compensation parameters are applied to correct the content of the target risk object to offset the impact of the interferents.

[0086] As mentioned above, through real-time monitoring and data analysis, not only can the current water quality status be detected, but also the trend of water quality changes in the future can be predicted. This forward-looking judgment enables the system to identify potential water quality problems in advance, avoiding the possible response lag in traditional monitoring methods, thereby providing managers with more accurate risk assessment and early warning information. By extracting features from sample data and combining historical data and environmental factors, a prediction model for water quality changes can be generated in real time. This predictive capability significantly improves the response speed of the monitoring system, allowing managers to take action in advance and effectively reduce the risks brought by fluctuations in pollutant concentrations. By integrating multiple data sources (such as water quality sample test results, environmental factor data, etc.), multi-dimensional risk assessment is achieved. This data fusion method avoids the limitations of a single sensor due to environmental interference or technical limitations, and can more comprehensively and accurately reflect the true water quality of the water source, reducing errors caused by sensor limitations.

[0087] Reference Figure 3 In the embodiment of the present application, a water source key risk object sample intelligent detection system is also provided, including:

[0088] Acquisition module 1, used to acquire sample detection data of water source in real time;

[0089] A first judgment module 2 is used to judge whether the sample detection data of the water source meets the abnormal condition;

[0090] The analysis module 3 is used to analyze the sample detection data when the water source meets the abnormal condition, identify the target risk parameters and the sample location according to the sample detection data, generate risk data based on the target risk parameters and the sample location, and send it to the management end;

[0091] An extraction module 4, used for extracting target features of the sample detection data when the water source does not meet the abnormal condition;

[0092] Prediction module 5, used to predict the water quality change trend of the water source based on the target feature;

[0093] The second judgment module 6 is used to judge whether the water quality meets the warning condition according to the prediction result;

[0094] The sending module 7 is used to generate a water quality treatment strategy and warning information based on the prediction result and send them to the management end when the water quality meets the warning condition.

[0095] As described above, it can be understood that the various components of the intelligent detection system for key risk object samples in water sources proposed in this application can realize the functions of any of the intelligent detection methods for key risk object samples in water sources described above, and the specific structure will not be repeated.

[0096] Reference Figure 4 In an embodiment of the present application, a computer device is also provided. The computer device may be a server, and its internal structure may be as follows: Figure 4 As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data such as monitoring data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for intelligent detection of key risk samples in a water source is implemented.

[0097] The above-mentioned processor executes the above-mentioned intelligent detection method for samples of key risk objects in water sources, including: acquiring sample detection data of water sources in real time; judging whether the sample detection data of the water source meets abnormal conditions; when the water source meets abnormal conditions, parsing the sample detection data, identifying target risk parameters and sample positions according to the sample detection data, generating risk data based on the target risk parameters and sample positions and sending it to the management end; when the water source does not meet the abnormal conditions, extracting target features of the sample detection data; predicting the water quality change trend of the water source based on the target features; judging whether the water quality meets the warning conditions based on the prediction results; when the water quality meets the warning conditions, generating a water quality treatment strategy and warning information based on the prediction results and sending them to the management end.

[0098] An embodiment of the present application also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, a method for intelligently detecting samples of key risk objects in a water source is implemented, comprising the following steps: acquiring sample detection data of a water source in real time; determining whether the sample detection data of the water source meets an abnormal condition; when the water source meets the abnormal condition, parsing the sample detection data, identifying target risk parameters and sample positions according to the sample detection data, generating risk data based on the target risk parameters and sample positions and sending the data to a management end; when the water source does not meet the abnormal condition, extracting target features of the sample detection data; predicting a water quality change trend of the water source based on the target features; determining whether the water quality meets an early warning condition based on the predicted result; when the water quality meets the early warning condition, generating a water quality treatment strategy and early warning information based on the predicted result and sending the data to a management end.

[0099] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided in this application and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0100] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the existence of other identical elements in the process, device, article or method including the element.

[0101] The above description is only a preferred embodiment of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for intelligent detection of key risk samples in water sources, characterized in that: The method comprises: Obtain sample test data of water sources in real time; Determining whether the sample detection data of the water source meets the abnormal condition; When the water source meets the abnormal condition, the sample detection data is parsed, the target risk parameter and the sample location are identified according to the sample detection data, and risk data is generated based on the target risk parameter and the sample location and sent to the management end; When the water source does not meet the abnormal condition, extracting the target features of the sample detection data; Predicting the water quality change trend of the water source based on the target feature; Determine whether the water quality meets the early warning condition according to the prediction result; When the water quality meets the warning condition, a water quality treatment strategy and warning information are generated based on the prediction result and sent to the management end.

2. The intelligent detection method for key risk samples of water sources according to claim 1 is characterized in that: After the step of determining whether the sample detection data of the water source meets the abnormal condition and before the step of determining when the water source meets the abnormal condition, the method further includes: Acquire environmental data of the sample location, and extract environmental features of the environmental data, wherein the environmental features include temperature features, humidity features, wind speed features, and rainfall features; Constructing an error analysis model, inputting the environmental characteristics and sample detection data into the error analysis model, evaluating the interference coefficient of the environmental characteristics on the sample detection data within a specific time through the error analysis model, and outputting the evaluation result, wherein the interference coefficient is used to evaluate the amplitude of the water quality parameter deviation from the normal value caused by the environmental characteristics; Based on the evaluation result, adjusting the detection threshold by the interference coefficient to obtain an adjusted detection threshold; According to the adjusted detection threshold, determining whether the water source meets an abnormal condition; When the sample detection data is greater than the adjusted detection threshold, it is determined that the sample detection data of the water source meets an abnormal condition.

3. The intelligent detection method for key risk object samples in water sources according to claim 2 is characterized in that: The method for predicting the water quality change trend of the water source based on the target feature includes: Obtain historical sample test data and corresponding historical environment data, and extract target features of the historical sample test data; Performing time series analysis on the target characteristics and historical environmental data to identify influencing parameters and hysteresis characteristics of the historical environmental data on the target characteristics; Obtain future environmental forecast data; Acquire current sample detection data, input the current sample detection data, influencing parameters, hysteresis characteristics and environmental prediction data into a preset trend prediction model, predict the changing trend of water quality under future environmental conditions through the trend prediction model, and output the prediction result; Based on the output results, the predicted water quality change trend is obtained.

4. The intelligent detection method for key risk object samples in water sources according to claim 3 is characterized in that: The method of judging whether the water quality meets the early warning condition according to the prediction result includes: Obtaining the predicted water quality change trend; Calculating the change range of the target characteristic according to the water quality change trend; Based on the change amplitude, determining whether the change amplitude of the target feature is within a preset amplitude range; When the change amplitude of the target characteristic exceeds the preset amplitude range, it is judged that the water quality meets the warning condition; When the change amplitude of the target feature is within a preset amplitude range, calculating the change rate of the target feature; Based on the change rate, determining whether the change rate of the target feature is within a preset rate range; When the change rate of the target characteristic exceeds a preset rate range, it is determined that the water quality change trend meets the early warning condition.

5. The intelligent detection method for key risk object samples in water sources according to claim 3 is characterized in that: The method further comprises: Obtain historical sample test data and corresponding historical environment data, and extract target features of the historical sample test data; Performing time series analysis on the target characteristics and historical environmental data to identify the change pattern of the target characteristics under conditions of different historical environmental data; Based on the recognition results, a frequency analysis model is constructed to analyze the change pattern under different environmental data conditions, and the frequency of abnormal judgment of the sample detection data is determined according to the change pattern; The current environment data is obtained, and the current environment data is input into the frequency analysis model, and the frequency analysis model is used to output the frequency of abnormal judgment of the sample detection data under the current environment data condition.

6. The intelligent detection method for key risk object samples in water sources according to claim 1 is characterized in that: After the step of generating a water quality treatment strategy and early warning information based on the prediction result and sending it to the management end, the method further includes: Obtain water source sample testing data after water quality treatment strategy; According to the time series, the water source sample test data after the water quality treatment strategy is compared with the initial water source sample test data to extract the change parameters of the target risk parameters; According to the change parameter of the target risk parameter, judging whether the water quality treatment strategy meets the replacement condition; If the change parameter is less than a preset change threshold, it is determined that the water quality treatment strategy meets the replacement condition.

7. The intelligent detection method for key risk object samples in water sources according to claim 1 is characterized in that: Before the step of determining whether the sample detection data of the water source meets the abnormal condition, the method further includes: Identifying target risk parameters of the sample test data; Identify the target risk parameters according to their categories, and perform overlapping signal detection according to each category of target risk objects; According to the result of the overlapping signal detection, determining whether the target risk object has an interference; When there are interferents on the target risk object, identifying the type of the interferents; Separating the overlapping signals according to the type of the interferents, and inferring the content of the interferents according to the signals of the interferents; Generate compensation parameters for target risk substances according to the content of interfering substances; The content of the target risk substance is adjusted based on the compensation parameter.

8. An intelligent detection system for key risk samples in water sources, characterized in that: include: An acquisition module is used to obtain sample detection data of water sources in real time; A first judgment module, used to judge whether the sample detection data of the water source meets the abnormal condition; An analysis module, used for analyzing the sample detection data when the water source meets the abnormal condition, identifying the target risk parameter and the sample location according to the sample detection data, generating risk data based on the target risk parameter and the sample location and sending it to the management end; An extraction module, used for extracting target features of the sample detection data when the water source does not meet the abnormal condition; A prediction module, used to predict the water quality change trend of the water source based on the target feature; A second judgment module is used to judge whether the water quality meets the warning condition according to the prediction result; The sending module is used to generate a water quality treatment strategy and early warning information based on the prediction result and send them to the management end when the water quality meets the early warning condition.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images