Water quality monitoring method based on edge calculation and related device

Abstract

The embodiment of the invention provides a water quality monitoring method based on edge calculation and a related device, and belongs to the technical field of data prediction. The method comprises the following steps: acquiring initial parameter data of a target water environment under a target sensor and target space information under a positioning sensor; determining the measurement reliability degree of the target sensor according to the initial parameter data, and adjusting the initial parameter data according to the measurement reliability degree to obtain target parameter data; performing data fusion on the target parameter data to obtain target fusion data of the target sensor; sending the target fusion data, the target parameter data and the target space information to a cloud server in communication connection with the edge device, so that the cloud server performs water quality prediction on the target fusion data, the target parameter data and the target space information according to a water quality prediction model to obtain a target prediction result of the target water environment; and receiving a target prediction result sent by the cloud server, and determining a target monitoring result of the target water environment according to the target prediction result.

Description

Technical Field

[0001] The present invention relates to the field of data prediction technology, and in particular to a water quality monitoring method based on edge computing and related devices. Background Art

[0002] Water environment monitoring is a vital part of water resource protection and management. With the development of artificial intelligence technology, researchers can better analyze water environment conditions and make water quality predictions by combining neural network models with sensor data, thereby providing a scientific basis for decision-making. However, although neural network models have shown strong capabilities in data processing and analysis, their prediction effects are still highly dependent on the quality of data collected by sensors. If there are errors in the measurement of the sensor, such errors may be caused by equipment aging, environmental changes, or improper maintenance, resulting in deviations between the sensor information and the actual water quality conditions. Therefore, the methods for water quality monitoring in related technologies still have the problem of inaccurate predictions. Summary of the invention

[0003] The main purpose of the embodiments of the present invention is to provide a water quality monitoring method and related devices based on edge computing, aiming to solve the problem in the related art that errors in sensor measurement data lead to inaccurate prediction of water quality monitoring results.

[0004] In a first aspect, an embodiment of the present invention provides a water quality monitoring method based on edge computing, comprising:

[0005] Obtaining initial parameter data corresponding to the target water environment under the target sensor and target space information corresponding to the target water environment under the positioning sensor;

[0006] Determining the measurement reliability corresponding to the target sensor according to the initial parameter data, and adjusting the initial parameter data according to the measurement reliability to obtain target parameter data;

[0007] Performing data fusion on the target parameter data to obtain target fusion data corresponding to the target sensor;

[0008] The target fusion data, the target parameter data and the target spatial information are sent to a cloud server that is communicatively connected to the edge device, so that the cloud server performs water quality prediction on the target fusion data, the target parameter data and the target spatial information according to a water quality prediction model to obtain a target prediction result corresponding to the target water environment;

[0009] The target prediction result sent by the cloud server is received, and the target monitoring result corresponding to the target water environment is determined according to the target prediction result.

[0010] In a second aspect, an embodiment of the present invention provides a water quality monitoring device based on edge computing, including:

[0011] A data acquisition module, used to obtain initial parameter data corresponding to the target water environment under the target sensor and target space information corresponding to the target water environment under the positioning sensor;

[0012] A data analysis module, used to determine the measurement reliability corresponding to the target sensor according to the initial parameter data, and adjust the initial parameter data according to the measurement reliability to obtain target parameter data;

[0013] A data fusion module, used for performing data fusion on the target parameter data to obtain target fusion data corresponding to the target sensor;

[0014] A data prediction module, used for sending the target fusion data, the target parameter data and the target spatial information to a cloud server that is communicatively connected to the edge device, so that the cloud server performs water quality prediction on the target fusion data, the target parameter data and the target spatial information according to a water quality prediction model, and obtains a target prediction result corresponding to the target water environment;

[0015] The water quality monitoring module is used to receive the target prediction result sent by the cloud server and determine the target monitoring result corresponding to the target water environment according to the target prediction result.

[0016] In a third aspect, an embodiment of the present invention further provides a terminal device, comprising a processor, a memory, a computer program stored in the memory and executable by the processor, and a data bus for realizing connection and communication between the processor and the memory, wherein when the computer program is executed by the processor, the steps of any one of the water quality monitoring methods based on edge computing provided in the specification of the present invention are implemented.

[0017] In a fourth aspect, an embodiment of the present invention further provides a storage medium for computer-readable storage, characterized in that the storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of any one of the water quality monitoring methods based on edge computing provided in the specification of the present invention.

[0018] The embodiment of the present invention provides a water quality monitoring method based on edge computing and a related device, the method comprising: obtaining initial parameter data of a target water environment under a target sensor and corresponding target space information under a positioning sensor. By analyzing the initial parameter data, the measurement reliability of the target sensor can be evaluated. Based on the measurement reliability, the initial parameter data can be adjusted accordingly to obtain more accurate target parameter data. This process can effectively reduce the impact of sensor errors on water quality monitoring results, and provides important support for the prediction accuracy of subsequent water quality monitoring results. The target parameter data is then subjected to data fusion processing to generate target fusion data of the target sensor. The generated target fusion data, target parameter data and target space information are then transmitted to a cloud server that is connected to the edge device for communication. In the cloud server, based on a water quality prediction model, the target fusion data, target parameter data and target space information are comprehensively analyzed to obtain a target prediction result of the target water environment. In this process, the application of artificial intelligence can further improve the accuracy of the target prediction result, while reducing the requirements for the computing power of the edge device and optimizing the use of system resources. Finally, the target prediction result sent by the received cloud server is received, and the target monitoring result corresponding to the target water environment is determined according to the target prediction result. This method effectively solves the problem of inaccurate prediction of water quality monitoring results due to errors in sensor measurement data in related technologies, improves the overall accuracy of water quality monitoring predictions, and provides a good foundation and support for subsequent real-time monitoring and related processing of the target water environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0020] Figure 1 A schematic diagram of a flow chart of a water quality monitoring method based on edge computing provided in an embodiment of the present invention;

[0021] Figure 2 A schematic diagram of the module structure of a water quality monitoring device based on edge computing provided by an embodiment of the present invention;

[0022] Figure 3 A schematic block diagram of the structure of a terminal device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0025] It should be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0026] The embodiment of the present invention provides a water quality monitoring method based on edge computing and a related device. The water quality monitoring method based on edge computing can be applied to a terminal device, which can be an electronic device such as a tablet computer, a laptop computer, a desktop computer, a personal digital assistant, and a wearable device. The terminal device can be a server or a server cluster.

[0027] Some embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0028] Please refer to Figure 1 , Figure 1 A flow chart of a water quality monitoring method based on edge computing provided in an embodiment of the present invention.

[0029] like Figure 1 As shown, the water quality monitoring method based on edge computing includes steps S101 to S105.

[0030] Step S101: obtaining initial parameter data corresponding to a target water environment under a target sensor and target space information corresponding to the target water environment under a positioning sensor.

[0031] Exemplarily, the target sensor includes but is not limited to sensors for measuring the temperature, solubility, pH value, turbidity, conductivity, ammonia nitrogen and other water quality parameters corresponding to the target water environment, so that the initial parameter data corresponding to the target water environment is collected in real time according to the target sensor. And the position of the target water environment is measured according to the positioning sensor such as GPS equipment to obtain the target space information corresponding to the target water environment.

[0032] Step S102: determining the measurement reliability corresponding to the target sensor according to the initial parameter data, and adjusting the initial parameter data according to the measurement reliability to obtain target parameter data.

[0033] Exemplarily, historical measurement data corresponding to the target sensor is obtained, and then data fitting is performed on the measurement data of the target sensor according to the historical measurement data to obtain a target fitting function corresponding to the target sensor, thereby obtaining the corresponding prediction parameter data at the current moment according to the target fitting function, thereby calculating the difference between the prediction parameter data and the initial parameter data, and when the difference is less than a preset value, it is determined that the measurement reliability corresponding to the target sensor is high; when the difference is greater than or equal to the preset value, it is determined that the measurement reliability corresponding to the target sensor is low.

[0034] Exemplarily, when the measurement reliability is high, the initial parameter data is determined as the target parameter data; when the measurement reliability is low, the initial parameter data corresponding to the target sensor is re-obtained until the measurement reliability corresponding to the initial parameter data is high.

[0035] In some embodiments, the target sensor includes at least a first sensor and a second sensor, the initial parameter data includes at least first parameter data corresponding to the first sensor at a target moment and second parameter data corresponding to the second sensor at the target moment, and determining the measurement reliability corresponding to the target sensor based on the initial parameter data includes: obtaining first historical data corresponding to the first sensor and second historical data corresponding to the second sensor; determining a first distribution function corresponding to the first sensor based on the first historical data and determining a second distribution function corresponding to the second sensor based on the second historical data; determining a target correlation value between the first parameter data and the second parameter data based on the first distribution function, the second distribution function, the first parameter data and the second parameter data; and determining the measurement reliability corresponding to the first sensor and the second sensor based on the target correlation value.

[0036] Exemplarily, the target sensor includes at least a first sensor and a second sensor, and the initial parameter data includes at least first parameter data corresponding to the first sensor at the target time and second parameter data corresponding to the second sensor at the target time. That is, the target sensor is a plurality of sensors, and the initial parameter data includes measurement data corresponding to the plurality of sensors.

[0037] Exemplarily, at least first historical data corresponding to the first sensor and second historical data corresponding to the second sensor are obtained from the database, and then which statistical analysis method is used to estimate the distribution function of the data according to the data distribution characteristics corresponding to the first historical data and the second historical data. The distribution function includes but is not limited to normal distribution, uniform distribution, exponential distribution, etc.

[0038] Exemplarily, a distribution function is determined according to the distribution characteristics of the first historical data, and a first parameter corresponding to the first historical data is obtained by fitting using statistical software such as R or Python's SciPy library, and then a first distribution function corresponding to the first historical data is determined according to the first parameter. Similarly, a distribution function is determined according to the distribution characteristics of the second historical data, and a second parameter corresponding to the second historical data is obtained by fitting using statistical software such as R or Python's SciPy library, and then a second distribution function corresponding to the second historical data is determined according to the second parameter.

[0039] Exemplarily, the first data distance corresponding to the first sensor and the second sensor at the target time is calculated according to the first parameter data and the second parameter data, and the second data distance corresponding to the first sensor and the second sensor at different times is calculated according to the first distribution function and the second distribution function, so as to calculate the ratio between the first data distance and the second data distance, and determine the ratio result as the target correlation value between the first parameter data and the second parameter data. The closer the target correlation value is to 1, the closer the distribution between the first parameter data and the second parameter data is to the distribution between the first distribution function and the second distribution function; the farther the target correlation value is from 1, the farther the distribution between the first parameter data and the second parameter data is from the distribution between the first distribution function and the second distribution function, that is, the greater the measurement error of the first parameter data and the second parameter data.

[0040] Exemplarily, a preset value is determined, and the absolute value of the difference between the target correlation value and 1 is calculated to obtain an absolute error. When the absolute error is less than or equal to the preset value, the corresponding measurement reliability between the first sensor and the second sensor is high, that is, the first sensor and the second sensor can prove to each other that the reliability of each other's sensor measurement results is high; when the absolute error is greater than the preset value, the corresponding measurement reliability between the first sensor and the second sensor is low, that is, the first sensor and the second sensor cannot prove to each other that the reliability of each other's sensor measurement results is high; in other words, there is a problem of large measurement error between the first parameter data and the second parameter data.

[0041] Specifically, by fitting the distribution function with historical data, the measurement characteristics of the sensor can be described more accurately, thereby improving the accuracy of the analysis of the current data. Therefore, by using the first distribution function and the second distribution function to analyze the target correlation value between the first parameter data and the second parameter data, the measurement reliability of the first sensor and the second sensor can be effectively quantified, thereby providing support for the subsequent provision of comprehensive and accurate water quality monitoring information.

[0042] In some embodiments, determining the target association value between the first parameter data and the second parameter data according to the first distribution function, the second distribution function, the first parameter data, and the second parameter data includes: determining the first variance corresponding to the first sensor according to the first historical data and determining the second variance corresponding to the second sensor according to the second historical data; performing an integral operation on the first distribution function according to the first parameter data and the second parameter data to obtain the first association value between the first parameter data and the second parameter data; performing an integral operation on the second distribution function according to the first parameter data and the second parameter data to obtain the second association value between the first parameter data and the second parameter data; performing information fusion on the first association value and the second association value according to the first variance and the second variance to determine the target association value between the first parameter data and the second parameter data; wherein the target association value is obtained according to the following formula:

[0043] ;

[0044] in, represents the target association value between the first parameter data and the second parameter data, represents the second variance, represents the first variance, abs represents the absolute value, represents the second parameter data, represents the first parameter data, represents the first distribution function, represents the second distribution function, represents the differential, The integral symbol, x represents the integral variable, used to characterize a first average distance corresponding to the first sensor and the second sensor under the first distribution function; The second average distance between the first sensor and the second sensor under the second distribution function is used to represent the second average distance between the first sensor and the second sensor under the second distribution function.

[0045] Exemplarily, a first average value of the first sensor and a second average value of the second sensor are calculated according to the first historical data and the second historical data, respectively. Thus, a first difference is obtained by performing a difference calculation according to the first historical data and the first average value, and the first difference is squared and then summed and averaged to obtain a first variance. Similarly, a second difference is obtained by performing a difference calculation according to the second historical data and the second average value, and the second difference is squared and then summed and averaged to obtain a second variance.

[0046] Exemplarily, the first distribution function is integrated according to the first parameter data and the second parameter data to obtain a first correlation value between the first parameter data and the second parameter data; wherein the first distribution function is used as an integral function, and the first parameter data and the second parameter data are used as an integral range, wherein the calculation formula of the first correlation value is as follows:

[0047]

[0048] in, represents a first association value between the first parameter data and the second parameter data, Represents the second parameter data, Represents the first parameter data, represents the first distribution function, represents the differential, The integral symbol, x, represents the integral variable.

[0049] Exemplarily, performing an integration operation on the first distribution function within the range given by the first parameter data and the second parameter data can calculate the first correlation value corresponding to the range determined by the first parameter data and the second parameter data, thereby reflecting the first current distance between the first parameter data and the second parameter data at the target moment through the first correlation value.

[0050] Similarly, the second distribution function is integrated according to the first parameter data and the second parameter data to obtain a second correlation value between the first parameter data and the second parameter data; wherein the second distribution function is used as an integral function, and the first parameter data and the second parameter data are used as integral ranges, wherein the calculation formula of the second correlation value is as follows:

[0051]

[0052] in, represents a second association value between the first parameter data and the second parameter data, Represents the second parameter data, Represents the first parameter data, represents the second distribution function, represents the differential, The integral symbol, x, represents the integral variable.

[0053] Exemplarily, by performing an integration operation on the second distribution function within the range given by the first parameter data and the second parameter data, a second correlation value corresponding to the range determined by the first parameter data and the second parameter data can be calculated, thereby reflecting the second current distance between the first parameter data and the second parameter data at the target moment through the second correlation value.

[0054] Exemplarily, the first distribution function is integrated according to the first historical parameter and the second historical parameter at the same time in the first historical data and the second historical data to obtain the first historical association value, and then the first historical association value is summed and averaged to obtain the first average distance. And the second distribution function is integrated according to the first historical parameter and the second historical parameter at the same time in the first historical data and the second historical data to obtain the second historical association value, and then the second historical association value is summed and averaged to obtain the second average distance.

[0055] Exemplarily, the target correlation value between the first parameter data and the second parameter data is determined by performing information fusion of the first variance and the second variance on the first correlation value and the second correlation value according to the following formula:

[0056] ;

[0057] in, represents the target association value between the first parameter data and the second parameter data, represents the second variance, represents the first variance, abs represents the absolute value, Represents the second parameter data, Represents the first parameter data, represents the first distribution function, represents the second distribution function, represents the differential, The integral symbol, x represents the integral variable, used to characterize a first average distance corresponding to the first sensor and the second sensor under a first distribution function; The second average distance between the first sensor and the second sensor under the second distribution function is used to represent the second average distance between the first sensor and the second sensor under the second distribution function.

[0058] Exemplarily, the first distribution function and the second distribution function and the corresponding first variance and second variance are comprehensively considered to make the information integration more comprehensive and optimize the calculation process of the target correlation value to make it more consistent with the actual measurement situation. In addition, the calculation result of the target correlation value can be dynamically adjusted by the first average distance, the second average distance, and the first variance and the second variance, so that it can better adapt to changes in the environment and operating conditions. This improves the adaptability to different scenarios and makes subsequent water quality monitoring more flexible and accurate.

[0059] In some embodiments, the adjusting the initial parameter data according to the measurement reliability to obtain the target parameter data includes: performing abnormal screening on the target sensor according to the measurement reliability to obtain an abnormal sensor; obtaining third historical data corresponding to the abnormal sensor, and predicting the measurement value corresponding to the abnormal sensor at the target time according to the third historical data to obtain the target estimated value corresponding to the abnormal sensor at the target time; adjusting the initial parameter data according to the target estimated value to obtain the target parameter data.

[0060] Exemplarily, the measurement reliability between the target sensors is analyzed, and when the number of high reliability between any one of the target sensors and the other remaining sensors is greater than a preset number, the sensor is a sensor with normal measurement, otherwise the sensor is determined to be an abnormal sensor.

[0061] For example, the target sensors include a first sensor, a second sensor, a third sensor, etc., the measurement reliability between the first sensor and the second sensor is high, the measurement reliability between the first sensor and the third sensor is low, and the measurement reliability between the second sensor and the third sensor is high. Then, for the first sensor, only the second sensor can prove that the measurement reliability is high; for the second sensor, both the first sensor and the third sensor can prove that the measurement reliability is high; for the third sensor, only the second sensor can prove that the measurement reliability is high. Therefore, if the preset number is 2, the number of times when only the second sensor has a high reliability is 2, so the second sensor is a sensor with normal measurement, and the first sensor and the third sensor are both abnormal sensors.

[0062] Exemplarily, the third historical data of the abnormal sensor is extracted from the database, and the third historical data is fitted and analyzed using a machine learning or deep learning algorithm. The target fitting function of the abnormal sensor is obtained by data fitting. Next, the target moment is input into the target fitting function as a time parameter to obtain the target estimated value of the abnormal sensor at the target moment, and then the abnormal parameter data corresponding to the abnormal sensor in the initial parameter data is compared with the target estimated value. If the difference between the abnormal parameter data and the target estimated value exceeds a preset threshold, it means that the existing abnormal parameter data may be inaccurate and needs to be recollected. Once the recollected parameter data is obtained, it is replaced with the original abnormal parameter data to obtain the updated target parameter data. On the contrary, if the difference between the abnormal parameter data and the target estimated value is less than or equal to the preset threshold, there is no need to modify the parameter data of the abnormal sensor.

[0063] In some embodiments, predicting the measurement value corresponding to the abnormal sensor at the target time according to the third historical data to obtain the target estimated value corresponding to the abnormal sensor at the target time includes: using a Kalman filter algorithm to determine the state equation, gain matrix and observation equation corresponding to the abnormal sensor according to the third historical data; determining the state prediction value, gain prediction value and observation prediction value corresponding to the abnormal sensor at the target time according to the state equation, the gain matrix and the observation equation; using an error prediction model to determine the target error corresponding to the abnormal sensor at the target time according to the state prediction value, the gain prediction value and the observation prediction value; determining the target prediction value corresponding to the abnormal sensor at the target time according to the state prediction value, the gain prediction value and the observation prediction value; optimizing the target prediction value according to the target error to obtain the target estimated value corresponding to the abnormal sensor at the target time.

[0064] Exemplarily, the Kalman filter algorithm is used to determine the state equation, gain matrix and observation equation corresponding to the abnormal sensor according to the third historical data, wherein the state equation is used to describe the change law of the state of the abnormal sensor over time. The gain matrix is ​​used to represent the relationship between the variables in the process of the state change of the abnormal sensor, and is calculated by the time update step in the Kalman filter algorithm. The observation equation is used to describe the relationship between the observed value and the state of the abnormal sensor.

[0065] Exemplarily, the state equation and the third historical data are used to perform time update to predict the state of the abnormal sensor at the target time to obtain the state prediction value. The gain matrix of the abnormal sensor at the target time is predicted by the calculation formula in the Kalman filter algorithm to obtain the gain prediction value. The observation equation and the state prediction value are used to predict the observation prediction value of the abnormal sensor at the target time.

[0066] Exemplarily, the error prediction model is a neural network model, and the historical state prediction value, historical gain prediction value and historical observation prediction value at each moment are obtained according to the third historical data, and the true error between the observation value and the true value at that moment is obtained, so that the historical state prediction value, historical gain prediction value and historical observation prediction value are used as input of the error prediction model, and the true error is used as output for model training to obtain the error prediction model.

[0067] Exemplarily, the state prediction value, gain prediction value and observation prediction value are used as inputs of the error prediction model, so as to obtain the target error of the abnormal sensor at the target time according to the error prediction model. Then, the target prediction value of the abnormal sensor at the target time is calculated by combining the state prediction value, gain prediction value and observation prediction value according to the Kalman filter. Thus, the target prediction value is optimized according to the target error to obtain a more accurate target estimation value.

[0068] Specifically, the Kalman filter algorithm is effectively used to predict and optimize the data of abnormal sensors, thereby improving the accuracy and reliability of target estimation values.

[0069] Step S103: performing data fusion on the target parameter data to obtain target fusion data corresponding to the target sensor.

[0070] Exemplarily, an actual measurement error corresponding to the target sensor is obtained, thereby determining a fusion weight corresponding to the target parameter data according to the actual measurement error, and then performing data fusion on the target parameter data according to the fusion weight to obtain target fusion data corresponding to the target sensor.

[0071] In some embodiments, performing data fusion on the target parameter data to obtain target fusion data corresponding to the target sensor includes: selecting any sensor among the target sensors to be determined as the third sensor corresponding to the current analysis, and obtaining third parameter data corresponding to the third sensor from the target parameter data; obtaining a parameter prediction value corresponding to the third sensor at the same time as the third parameter data; calculating the mean square error corresponding to the parameter prediction value and the third parameter data, and determining the target weight corresponding to the data fusion of the third parameter data according to the mean square error; performing data fusion on the third parameter data according to the target weight to obtain parameter fusion data corresponding to the third sensor; determining the sensor weight corresponding to the third sensor, and performing data fusion according to the sensor weight and the parameter fusion data to obtain the target fusion data corresponding to the target sensor.

[0072] Exemplarily, the target sensor includes multiple sensors, then any one of the target sensors is selected to be determined as the third sensor corresponding to the current analysis, and third parameter data corresponding to the third sensor is obtained from the target parameter data, wherein the third parameter data includes measurement data corresponding to the third sensor at multiple time moments.

[0073] Exemplarily, a parameter prediction value corresponding to the third sensor at the same time as the third parameter data is obtained according to a parameter prediction model corresponding to the third sensor. The parameter prediction model may be a prediction model obtained through machine learning or deep learning based on historical parameter data corresponding to the third sensor.

[0074] Exemplarily, the mean square error between the parameter prediction value and the third parameter data is calculated, and the inverse of all the mean square errors is taken to obtain the inverse result, and then the inverse results are summed to obtain the inverse sum, and then the ratio between the inverse value of each mean square error and the inverse sum is determined as the target weight.

[0075] Exemplarily, the third parameter data at different times are weightedly fused according to the target weight to obtain parameter fusion data corresponding to the third sensor.

[0076] Exemplarily, an actual measurement error corresponding to the third sensor is obtained, thereby determining a sensor weight corresponding to the third sensor according to the actual measurement error, and then performing weighted summation on the parameter fusion data according to the sensor weight to obtain target fusion data corresponding to the target sensor.

[0077] Step S104: Send the target fusion data, the target parameter data and the target spatial information to a cloud server that is communicatively connected to the edge device, so that the cloud server performs water quality prediction on the target fusion data, the target parameter data and the target spatial information according to a water quality prediction model to obtain a target prediction result corresponding to the target water environment.

[0078] For example, a suitable communication protocol is selected to realize data transmission between the edge device and the cloud server. The communication protocol includes but is not limited to HTTP, MQTT, CoAP, etc., and the optimal protocol needs to be selected according to the specific application scenario and network conditions.

[0079] Exemplarily, according to the requirements of the communication protocol, the target fusion data, target parameter data and target space information are packaged to obtain a request instruction, and then the request instruction is sent to the cloud server. After the cloud server receives the corresponding request instruction, it parses the request instruction to obtain the target fusion data, target parameter data and target space information, so that the cloud server performs water quality prediction on the target fusion data, target parameter data and target space information according to the water quality prediction model, and obtains the target prediction result corresponding to the target water environment. Among them, the water quality prediction model can be a classification model corresponding to a neural network or machine learning or a water quality score prediction model.

[0080] For example, if the water quality prediction model is a classification model based on a neural network, the target prediction result corresponding to the target water environment can be determined as one of excellent water quality, good water quality, and poor water quality according to the target fusion data, target parameter data, and target spatial information through the water quality prediction model. If the water quality prediction model is a water quality scoring model based on a neural network, the target prediction result corresponding to the target water environment can be determined as a score according to the target fusion data, target parameter data, and target spatial information through the water quality prediction model. The higher the score, the better the water quality of the target water environment.

[0081] Exemplarily, the cloud server performs water quality prediction on the target fusion data, target parameter data and target spatial information according to the water quality prediction model. After obtaining the target prediction results corresponding to the target water environment, the cloud server packages the target prediction results according to the communication protocol and sends them to the edge device.

[0082] Step S105: receiving the target prediction result sent by the cloud server, and determining the target monitoring result corresponding to the target water environment according to the target prediction result.

[0083] Exemplarily, the target prediction result sent by the cloud server is received, and then the target prediction result is compared with the preset information. When the target prediction result meets the preset information, the target monitoring result corresponding to the target water environment is determined to be qualified water quality; when the target prediction result does not meet the preset information, the target monitoring result corresponding to the target water environment is determined to be unqualified water quality.

[0084] For example, when the target prediction result is one of excellent water quality, good water quality, and poor water quality, and the preset information is one of excellent water quality and good water quality, then when the target prediction result is one of excellent water quality and good water quality, the target monitoring result is qualified water quality, otherwise the target monitoring result corresponding to the target water environment is determined to be unqualified water quality. When the target prediction result is a score, the preset information is a preset value, and when the target prediction result is greater than or equal to the preset value, the target monitoring result is qualified water quality, otherwise the target monitoring result corresponding to the target water environment is determined to be unqualified water quality.

[0085] In some embodiments, after obtaining the target monitoring result, the method further includes: obtaining a manual verification result corresponding to the target monitoring result, and sending the manual verification result to the cloud server, so that the cloud server updates the water quality prediction model according to the manual verification result to obtain the updated water quality prediction model.

[0086] Exemplarily, the acquired target monitoring results are submitted to professional water quality analysts or experts in related fields for manual verification to obtain manual verification results. When there is a modification to the target monitoring results in the manual verification results, that is, when the water quality prediction model predicts an error, the manual verification results are sent to the cloud server, so that after the cloud server receives the manual verification results, it performs data analysis and storage. When the number of manual verification results collected reaches a certain number, the cloud server calls the update mechanism of the water quality prediction model. Thus, the water quality prediction model is updated according to the manual verification results to obtain an updated water quality prediction model, and then the cloud server obtains an updated water quality prediction model. The updated water quality prediction model will more accurately reflect the water quality of the target water environment and can more reliably predict future water quality. In addition, the cloud server also needs to deploy the updated water quality prediction model to the cloud server for subsequent edge devices to perform water quality prediction. Thus, by effectively utilizing the manual verification results to continuously optimize the water quality prediction model, the accuracy and reliability of the water quality prediction model can be further improved.

[0087] See also Figure 2 , Figure 2A water quality monitoring device 200 based on edge computing is provided in an embodiment of the present application. The water quality monitoring device 200 based on edge computing includes a data acquisition module 201, a data analysis module 202, a data fusion module 203, a data prediction module 204, and a water quality monitoring module 205, wherein the data acquisition module 201 is used to obtain initial parameter data corresponding to the target water environment under the target sensor and target space information corresponding to the target water environment under the positioning sensor; the data analysis module 202 is used to determine the measurement reliability corresponding to the target sensor according to the initial parameter data, and adjust the initial parameter data according to the measurement reliability to obtain the target parameter data; The data fusion module 203 is used to perform data fusion on the target parameter data to obtain the target fusion data corresponding to the target sensor; the data prediction module 204 is used to send the target fusion data, the target parameter data and the target spatial information to a cloud server that is communicatively connected to the edge device, so that the cloud server performs water quality prediction on the target fusion data, the target parameter data and the target spatial information according to the water quality prediction model to obtain the target prediction result corresponding to the target water environment; the water quality monitoring module 205 is used to receive the target prediction result sent by the cloud server, and determine the target monitoring result corresponding to the target water environment according to the target prediction result.

[0088] In some implementations, the target sensor includes at least a first sensor and a second sensor, the initial parameter data includes at least first parameter data corresponding to the first sensor at the target time and second parameter data corresponding to the second sensor at the target time, and the data analysis module 202 performs, in the process of determining the measurement reliability corresponding to the target sensor according to the initial parameter data:

[0089] Obtaining first historical data corresponding to the first sensor and second historical data corresponding to the second sensor;

[0090] Determine a first distribution function corresponding to the first sensor according to the first historical data and determine a second distribution function corresponding to the second sensor according to the second historical data;

[0091] determining a target correlation value between the first parameter data and the second parameter data according to the first distribution function, the second distribution function, the first parameter data, and the second parameter data;

[0092] The corresponding measurement reliability between the first sensor and the second sensor is determined according to the target association value.

[0093] In some implementations, the data analysis module 202, in the process of determining the target association value between the first parameter data and the second parameter data according to the first distribution function, the second distribution function, the first parameter data, and the second parameter data, performs:

[0094] Determine a first variance corresponding to the first sensor according to the first historical data and determine a second variance corresponding to the second sensor according to the second historical data;

[0095] Performing an integral operation on the first distribution function according to the first parameter data and the second parameter data to obtain a first correlation value between the first parameter data and the second parameter data;

[0096] Performing an integral operation on the second distribution function according to the first parameter data and the second parameter data to obtain a second correlation value between the first parameter data and the second parameter data;

[0097] Performing information fusion on the first association value and the second association value according to the first variance and the second variance to determine the target association value between the first parameter data and the second parameter data;

[0098] The target correlation value is obtained according to the following formula:

[0099] ;

[0100] in, represents the target association value between the first parameter data and the second parameter data, represents the second variance, represents the first variance, abs represents the absolute value, represents the second parameter data, represents the first parameter data, represents the first distribution function, represents the second distribution function, represents the differential, The integral symbol, x represents the integral variable, used to characterize a first average distance corresponding to the first sensor and the second sensor under the first distribution function; The second average distance between the first sensor and the second sensor under the second distribution function is used to represent the second average distance between the first sensor and the second sensor under the second distribution function.

[0101] In some implementations, during the process of adjusting the initial parameter data according to the measurement reliability to obtain the target parameter data, the data analysis module 202 performs:

[0102] Performing abnormal screening on the target sensor according to the measurement reliability to obtain an abnormal sensor;

[0103] Obtaining third historical data corresponding to the abnormal sensor, and predicting a measurement value corresponding to the abnormal sensor at a target time according to the third historical data, to obtain a target estimated value corresponding to the abnormal sensor at the target time;

[0104] The target parameter data is obtained by adjusting the initial parameter data according to the target estimated value.

[0105] In some implementations, the data analysis module 202 performs, during the process of predicting the measurement value corresponding to the abnormal sensor at the target time according to the third historical data and obtaining the target estimated value corresponding to the abnormal sensor at the target time:

[0106] Determine the state equation, gain matrix and observation equation corresponding to the abnormal sensor according to the third historical data using a Kalman filter algorithm;

[0107] Determine the state prediction value, gain prediction value and observation prediction value corresponding to the abnormal sensor at the target time according to the state equation, the gain matrix and the observation equation;

[0108] Determine the target error corresponding to the abnormal sensor at the target time according to the state prediction value, the gain prediction value and the observation prediction value using an error prediction model;

[0109] Determine a target prediction value corresponding to the abnormal sensor at the target time according to the state prediction value, the gain prediction value and the observation prediction value;

[0110] The target prediction value is optimized according to the target error to obtain the target estimated value corresponding to the abnormal sensor at the target time.

[0111] In some implementations, the data fusion module 203 performs, during the process of fusing the target parameter data to obtain the target fusion data corresponding to the target sensor:

[0112] Select any sensor among the target sensors to be determined as the third sensor corresponding to the current analysis, and obtain the third parameter data corresponding to the third sensor from the target parameter data;

[0113] Obtaining a parameter prediction value corresponding to the third sensor at the same time as the third parameter data;

[0114] Calculating the mean square error between the parameter prediction value and the third parameter data, and determining the target weight corresponding to the third parameter data when performing data fusion according to the mean square error;

[0115] Performing data fusion on the third parameter data according to the target weight to obtain parameter fusion data corresponding to the third sensor;

[0116] A sensor weight corresponding to the third sensor is determined, and data fusion is performed according to the sensor weight and the parameter fusion data to obtain the target fusion data corresponding to the target sensor.

[0117] In some implementations, after obtaining the target monitoring result, the water quality monitoring module 205 further performs:

[0118] Obtain a manual verification result corresponding to the target monitoring result, and send the manual verification result to the cloud server, so that the cloud server updates the water quality prediction model according to the manual verification result to obtain the updated water quality prediction model.

[0119] In some embodiments, the water quality monitoring device 200 based on edge computing can be applied to a terminal device.

[0120] It should be noted that technical personnel in the relevant field can clearly understand that, for the convenience and conciseness of description, the specific working process of the water quality monitoring device 200 based on edge computing described above can refer to the corresponding process in the aforementioned water quality monitoring method based on edge computing embodiment, and will not be repeated here.

[0121] See also Figure 3 , Figure 3 A schematic block diagram of the structure of a terminal device provided in an embodiment of the present invention.

[0122] like Figure 3 As shown, the terminal device 300 includes a processor 301 and a memory 302 , and the processor 301 and the memory 302 are connected via a bus 303 , such as an I2C (Inter-integrated Circuit) bus.

[0123] Specifically, the processor 301 is used to provide computing and control capabilities to support the operation of the entire terminal device. The processor 301 can be a central processing unit (CPU), and the processor 301 can also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.

[0124] Specifically, the memory 302 may be a Flash chip, a read-only memory (ROM) disk, an optical disk, a USB flash drive, or a mobile hard disk.

[0125] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a partial structure related to the embodiment of the present invention, and does not constitute a limitation on the terminal device to which the embodiment of the present invention is applied. The specific server may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0126] The processor is used to run a computer program stored in a memory, and implement any one of the edge computing-based water quality monitoring methods provided in an embodiment of the present invention when executing the computer program.

[0127] In one embodiment, the processor is used to run a computer program stored in the memory, and implements the following steps when executing the computer program:

[0128] Obtaining initial parameter data corresponding to the target water environment under the target sensor and target space information corresponding to the target water environment under the positioning sensor;

[0129] Determining the measurement reliability corresponding to the target sensor according to the initial parameter data, and adjusting the initial parameter data according to the measurement reliability to obtain target parameter data;

[0130] Performing data fusion on the target parameter data to obtain target fusion data corresponding to the target sensor;

[0131] The target fusion data, the target parameter data and the target spatial information are sent to a cloud server that is communicatively connected to the edge device, so that the cloud server performs water quality prediction on the target fusion data, the target parameter data and the target spatial information according to a water quality prediction model to obtain a target prediction result corresponding to the target water environment;

[0132] The target prediction result sent by the cloud server is received, and the target monitoring result corresponding to the target water environment is determined according to the target prediction result.

[0133] In some implementations, the target sensor includes at least a first sensor and a second sensor, the initial parameter data includes at least first parameter data corresponding to the first sensor at the target moment and second parameter data corresponding to the second sensor at the target moment, and the processor 301, in the process of determining the measurement reliability corresponding to the target sensor according to the initial parameter data, executes:

[0134] Obtaining first historical data corresponding to the first sensor and second historical data corresponding to the second sensor;

[0135] Determine a first distribution function corresponding to the first sensor according to the first historical data and determine a second distribution function corresponding to the second sensor according to the second historical data;

[0136] determining a target correlation value between the first parameter data and the second parameter data according to the first distribution function, the second distribution function, the first parameter data, and the second parameter data;

[0137] The corresponding measurement reliability between the first sensor and the second sensor is determined according to the target association value.

[0138] In some implementations, the processor 301, in the process of determining the target association value between the first parameter data and the second parameter data according to the first distribution function, the second distribution function, the first parameter data, and the second parameter data, executes:

[0139] determining a first variance corresponding to the first sensor according to the first historical data and determining a second variance corresponding to the second sensor according to the second historical data;

[0140] Performing an integral operation on the first distribution function according to the first parameter data and the second parameter data to obtain a first correlation value between the first parameter data and the second parameter data;

[0141] Performing an integral operation on the second distribution function according to the first parameter data and the second parameter data to obtain a second correlation value between the first parameter data and the second parameter data;

[0142] Performing information fusion on the first association value and the second association value according to the first variance and the second variance to determine the target association value between the first parameter data and the second parameter data;

[0143] The target correlation value is obtained according to the following formula:

[0144] ;

[0145] in, represents the target association value between the first parameter data and the second parameter data, represents the second variance, represents the first variance, abs represents the absolute value, represents the second parameter data, represents the first parameter data, represents the first distribution function, represents the second distribution function, represents the differential, The integral symbol, x represents the integral variable, used to characterize a first average distance corresponding to the first sensor and the second sensor under the first distribution function; The second average distance between the first sensor and the second sensor under the second distribution function is used to represent the second average distance between the first sensor and the second sensor under the second distribution function.

[0146] In some implementations, during the process of adjusting the initial parameter data according to the measurement reliability to obtain the target parameter data, the processor 301 executes:

[0147] Performing abnormal screening on the target sensor according to the measurement reliability to obtain an abnormal sensor;

[0148] Obtaining third historical data corresponding to the abnormal sensor, and predicting a measurement value corresponding to the abnormal sensor at a target time according to the third historical data, to obtain a target estimated value corresponding to the abnormal sensor at the target time;

[0149] The target parameter data is obtained by adjusting the initial parameter data according to the target estimated value.

[0150] In some implementations, the processor 301, in the process of predicting the measurement value corresponding to the abnormal sensor at the target time according to the third historical data and obtaining the target estimated value corresponding to the abnormal sensor at the target time, executes:

[0151] Determine the state equation, gain matrix and observation equation corresponding to the abnormal sensor according to the third historical data using a Kalman filter algorithm;

[0152] Determine the state prediction value, gain prediction value and observation prediction value corresponding to the abnormal sensor at the target time according to the state equation, the gain matrix and the observation equation;

[0153] Determine the target error corresponding to the abnormal sensor at the target time according to the state prediction value, the gain prediction value and the observation prediction value using an error prediction model;

[0154] Determine a target prediction value corresponding to the abnormal sensor at the target time according to the state prediction value, the gain prediction value and the observation prediction value;

[0155] The target prediction value is optimized according to the target error to obtain the target estimated value corresponding to the abnormal sensor at the target time.

[0156] In some implementations, the processor 301, during the process of performing data fusion on the target parameter data to obtain the target fusion data corresponding to the target sensor, executes:

[0157] Select any sensor among the target sensors to be determined as the third sensor corresponding to the current analysis, and obtain the third parameter data corresponding to the third sensor from the target parameter data;

[0158] Obtaining a parameter prediction value corresponding to the third sensor at the same time as the third parameter data;

[0159] Calculating the mean square error between the parameter prediction value and the third parameter data, and determining the target weight corresponding to the third parameter data when performing data fusion according to the mean square error;

[0160] Performing data fusion on the third parameter data according to the target weight to obtain parameter fusion data corresponding to the third sensor;

[0161] A sensor weight corresponding to the third sensor is determined, and data fusion is performed according to the sensor weight and the parameter fusion data to obtain the target fusion data corresponding to the target sensor.

[0162] In some implementations, after obtaining the target monitoring result, the processor 301 further executes:

[0163] Obtain a manual verification result corresponding to the target monitoring result, and send the manual verification result to the cloud server, so that the cloud server updates the water quality prediction model according to the manual verification result to obtain the updated water quality prediction model.

[0164] It should be noted that technical personnel in the relevant field can clearly understand that for the convenience and conciseness of description, the specific working process of the terminal device described above can refer to the corresponding process in the aforementioned edge computing-based water quality monitoring method embodiment, and will not be repeated here.

[0165] An embodiment of the present invention also provides a storage medium for computer-readable storage, wherein the storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of any water quality monitoring method based on edge computing provided in the description of the embodiment of the present invention.

[0166] The storage medium may be an internal storage unit of the terminal device described in the foregoing embodiment, such as a hard disk or memory of the terminal device. The storage medium may also be an external storage device of the terminal device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc., equipped on the terminal device.

[0167] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In a hardware embodiment, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or transient medium). As is known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those skilled in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0168] It should be understood that the term "and / or" used in the present specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, including these combinations. It should be noted that, in this article, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "including a..." does not exclude the presence of other identical elements in the process, method, article or system including the element.

[0169] The serial numbers of the embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments. The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A water quality monitoring method based on edge computing, characterized in that: The method comprises: Obtaining initial parameter data corresponding to the target water environment under the target sensor and target space information corresponding to the target water environment under the positioning sensor; Determining the measurement reliability corresponding to the target sensor according to the initial parameter data, and adjusting the initial parameter data according to the measurement reliability to obtain target parameter data; Performing data fusion on the target parameter data to obtain target fusion data corresponding to the target sensor; The target fusion data, the target parameter data and the target spatial information are sent to a cloud server that is communicatively connected to the edge device, so that the cloud server performs water quality prediction on the target fusion data, the target parameter data and the target spatial information according to a water quality prediction model to obtain a target prediction result corresponding to the target water environment; The target prediction result sent by the cloud server is received, and the target monitoring result corresponding to the target water environment is determined according to the target prediction result.

2. The method according to claim 1, characterized in that The target sensor includes at least a first sensor and a second sensor, the initial parameter data includes at least first parameter data corresponding to the first sensor at a target time and second parameter data corresponding to the second sensor at the target time, and determining the measurement reliability corresponding to the target sensor according to the initial parameter data includes: Obtaining first historical data corresponding to the first sensor and second historical data corresponding to the second sensor; Determine a first distribution function corresponding to the first sensor according to the first historical data and determine a second distribution function corresponding to the second sensor according to the second historical data; determining a target correlation value between the first parameter data and the second parameter data according to the first distribution function, the second distribution function, the first parameter data, and the second parameter data; The corresponding measurement reliability between the first sensor and the second sensor is determined according to the target association value.

3. The method according to claim 2, characterized in that The determining a target correlation value between the first parameter data and the second parameter data according to the first distribution function, the second distribution function, the first parameter data, and the second parameter data comprises: determining a first variance corresponding to the first sensor according to the first historical data and determining a second variance corresponding to the second sensor according to the second historical data; Performing an integral operation on the first distribution function according to the first parameter data and the second parameter data to obtain a first correlation value between the first parameter data and the second parameter data; Performing an integral operation on the second distribution function according to the first parameter data and the second parameter data to obtain a second correlation value between the first parameter data and the second parameter data; Performing information fusion on the first association value and the second association value according to the first variance and the second variance to determine the target association value between the first parameter data and the second parameter data; The target correlation value is obtained according to the following formula: ； in, represents the target association value between the first parameter data and the second parameter data, represents the second variance, represents the first variance, abs represents the absolute value, represents the second parameter data, represents the first parameter data, represents the first distribution function, represents the second distribution function, represents the differential, The integral symbol, x represents the integral variable, used to characterize a first average distance corresponding to the first sensor and the second sensor under the first distribution function; The second average distance between the first sensor and the second sensor under the second distribution function is used to represent the second average distance between the first sensor and the second sensor under the second distribution function.

4. The method according to claim 1, characterized in that The adjusting the initial parameter data according to the measurement reliability to obtain the target parameter data includes: Performing abnormal screening on the target sensor according to the measurement reliability to obtain an abnormal sensor; Obtaining third historical data corresponding to the abnormal sensor, and predicting a measurement value corresponding to the abnormal sensor at a target time according to the third historical data, to obtain a target estimated value corresponding to the abnormal sensor at the target time; The target parameter data is obtained by adjusting the initial parameter data according to the target estimated value.

5. The method according to claim 4, characterized in that The step of predicting the measurement value corresponding to the abnormal sensor at the target time according to the third historical data to obtain the target estimated value corresponding to the abnormal sensor at the target time includes: Determine the state equation, gain matrix and observation equation corresponding to the abnormal sensor according to the third historical data using a Kalman filter algorithm; Determine the state prediction value, gain prediction value and observation prediction value corresponding to the abnormal sensor at the target time according to the state equation, the gain matrix and the observation equation; Determine the target error corresponding to the abnormal sensor at the target time according to the state prediction value, the gain prediction value and the observation prediction value using an error prediction model; Determine a target prediction value corresponding to the abnormal sensor at the target time according to the state prediction value, the gain prediction value and the observation prediction value; The target prediction value is optimized according to the target error to obtain the target estimated value corresponding to the abnormal sensor at the target time.

6. The method according to claim 1, characterized in that The step of fusing the target parameter data to obtain target fusion data corresponding to the target sensor includes: Select any sensor among the target sensors to be determined as the third sensor corresponding to the current analysis, and obtain the third parameter data corresponding to the third sensor from the target parameter data; Obtaining a parameter prediction value corresponding to the third sensor at the same time as the third parameter data; Calculating the mean square error between the parameter prediction value and the third parameter data, and determining the target weight corresponding to the third parameter data when performing data fusion according to the mean square error; Performing data fusion on the third parameter data according to the target weight to obtain parameter fusion data corresponding to the third sensor; A sensor weight corresponding to the third sensor is determined, and data fusion is performed according to the sensor weight and the parameter fusion data to obtain the target fusion data corresponding to the target sensor.

7. The method according to any one of claims 1 to 6, characterized in that After obtaining the target monitoring result, the method further includes: Obtain a manual verification result corresponding to the target monitoring result, and send the manual verification result to the cloud server, so that the cloud server updates the water quality prediction model according to the manual verification result to obtain the updated water quality prediction model.

8. A water quality monitoring device based on edge computing, characterized in that: include: A data acquisition module, used to obtain initial parameter data corresponding to the target water environment under the target sensor and target space information corresponding to the target water environment under the positioning sensor; A data analysis module, used to determine the measurement reliability corresponding to the target sensor according to the initial parameter data, and adjust the initial parameter data according to the measurement reliability to obtain target parameter data; A data fusion module, used for performing data fusion on the target parameter data to obtain target fusion data corresponding to the target sensor; A data prediction module, used for sending the target fusion data, the target parameter data and the target spatial information to a cloud server that is communicatively connected to the edge device, so that the cloud server performs water quality prediction on the target fusion data, the target parameter data and the target spatial information according to a water quality prediction model, and obtains a target prediction result corresponding to the target water environment; The water quality monitoring module is used to receive the target prediction result sent by the cloud server and determine the target monitoring result corresponding to the target water environment according to the target prediction result.

9. A terminal device, characterized in that: The terminal device includes a processor and a memory; The memory is used to store computer programs; The processor is used to execute the computer program and implement the water quality monitoring method based on edge computing as described in any one of claims 1 to 7 when executing the computer program.

10. A computer storage medium for computer storage, characterized in that: The computer storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of the water quality monitoring method based on edge computing as described in any one of claims 1 to 7.

Images