A method and system for collecting environmental hydrological monitoring data based on artificial intelligence

Abstract

The embodiment of the present invention provides an environmental protection hydrological monitoring data collection method and system based on artificial intelligence, which belongs to the field of data processing technology. The method includes: using a first sensor to collect data from a target object to obtain first collected data and using a second sensor to collect data from a target object to obtain second collected data; using the first collected data and the second collected data to perform correlation analysis on the first sensor and the second sensor to obtain a relationship analysis result; performing an abnormality analysis on the first collected data and the second collected data according to the relationship analysis result to obtain first abnormal data of the first collected data and second abnormal data of the second collected data; performing data prediction on the first abnormal data according to the first collected data to obtain first target data; performing data prediction on the second abnormal data according to the second collected data to obtain second target data; determining the target collection result of the target object according to the first target data and the second target data.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to an environmental protection hydrological monitoring data collection method and system based on artificial intelligence. Background Art

[0002] In the process of environmental protection hydrological detection data collection, due to the existence of collection errors, collection errors may come from many aspects, including equipment failure, environmental interference, human operation errors and other factors. These factors may act alone or in combination, resulting in deviations between the collected data and the actual environmental parameters. For example, insufficient accuracy or aging of sensor equipment may cause data distortion; electromagnetic interference or other force majeure factors in the environment may also have a negative impact on data collection; and the lack of technical proficiency or negligence of operators may also cause inaccurate data collection. Therefore, the existence of collection errors may lead to incorrect identification and judgment in the subsequent data analysis stage, thereby affecting the overall analysis accuracy and decision-making quality. Therefore, a more accurate data collection method is urgently needed. Summary of the invention

[0003] The main purpose of the embodiments of the present invention is to provide an environmental protection hydrological monitoring data collection method and system based on artificial intelligence, aiming to solve the problem of inaccurate data collection in related technologies, which leads to erroneous identification and judgment in the subsequent data analysis stage.

[0004] In a first aspect, an embodiment of the present invention provides an environmental protection hydrological monitoring data collection method based on artificial intelligence, comprising:

[0005] Using a first sensor to collect data from a target object to obtain first collected data and using a second sensor to collect data from the target object to obtain second collected data, wherein the first sensor and the second sensor are two different sensors;

[0006] Performing a correlation analysis on the first sensor and the second sensor using the first collected data and the second collected data to obtain a relationship analysis result, wherein the relationship analysis result is used to characterize the association relationship between the measurement data of the target object under the first sensor and the second sensor;

[0007] Performing an abnormality analysis on the first collected data and the second collected data according to the relationship analysis result to obtain first abnormal data corresponding to the first collected data and second abnormal data corresponding to the second collected data;

[0008] Performing data prediction on the first abnormal data according to the first collected data to obtain first target data corresponding to the first sensor;

[0009] Performing data prediction on the second abnormal data according to the second collected data to obtain second target data corresponding to the second sensor;

[0010] A target acquisition result corresponding to the target object is determined according to the first target data and the second target data.

[0011] In a second aspect, an embodiment of the present invention provides an environmental protection hydrological monitoring data acquisition system based on artificial intelligence, comprising:

[0012] A data acquisition module, used to acquire first acquired data by using a first sensor to acquire data from a target object and to acquire second acquired data by using a second sensor to acquire data from the target object, wherein the first sensor and the second sensor are two different sensors;

[0013] a data analysis module, configured to perform a correlation analysis on the first sensor and the second sensor using the first collected data and the second collected data to obtain a relationship analysis result, wherein the relationship analysis result is used to characterize the correlation relationship between the measurement data of the target object under the first sensor and the second sensor;

[0014] an abnormality classification module, used for performing abnormality analysis on the first collected data and the second collected data according to the relationship analysis result, and obtaining first abnormal data corresponding to the first collected data and second abnormal data corresponding to the second collected data;

[0015] A first prediction module, configured to perform data prediction on the first abnormal data according to the first collected data, and obtain first target data corresponding to the first sensor;

[0016] A second prediction module, configured to perform data prediction on the second abnormal data according to the second collected data, and obtain second target data corresponding to the second sensor;

[0017] A result determination module is used to determine a target acquisition result corresponding to the target object according to the first target data and the second target data.

[0018] The embodiment of the present invention provides an environmental protection hydrological monitoring data collection method and system based on artificial intelligence, the method comprising: using a first sensor to collect data on a target object to obtain first collected data and using a second sensor to collect data on the target object to obtain second collected data; using the first collected data and the second collected data to perform correlation analysis on the first sensor and the second sensor to obtain a relationship analysis result; performing abnormal analysis on the first collected data and the second collected data according to the relationship analysis result to obtain first abnormal data corresponding to the first collected data and second abnormal data corresponding to the second collected data; performing data prediction on the first abnormal data according to the first collected data to obtain first target data corresponding to the first sensor; performing data prediction on the second abnormal data according to the second collected data to obtain second target data corresponding to the second sensor; determining the target collection result corresponding to the target object according to the first target data and the second target data. The method collects data on the same target object by using the first sensor and the second sensor, which can effectively reduce the error and deviation of a single sensor. Performing correlation analysis on the first collected data and the second collected data provides support for the subsequent correction of the measurement error of the first sensor or the second sensor, further improving the accuracy of the collected data. Then, based on the pattern of normal data, the abnormal data is filled or corrected, so that the final target collection data is more in line with the actual situation, which can not only improve the overall reliability of the collected data, but also significantly improve the quality of data collection, making the final target collection results more accurate and reliable, providing support for solving the problems of incorrect identification and judgment in the subsequent collection data analysis stage, thereby improving the overall analysis accuracy and decision quality. It solves the problem of inaccurate data collection in related technologies, which leads to incorrect identification and judgment in the subsequent data analysis stage. 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 an environmental protection hydrological monitoring data collection method based on artificial intelligence provided by an embodiment of the present invention;

[0021] Figure 2 A schematic diagram of the module structure of an environmental protection hydrological monitoring data acquisition system based on artificial intelligence provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0022] 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.

[0023] 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 be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0024] 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.

[0025] The embodiment of the present invention provides an environmental protection hydrological monitoring data collection method and system based on artificial intelligence. The environmental protection hydrological monitoring data collection method based on artificial intelligence 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.

[0026] 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.

[0027] Please refer to Figure 1 , Figure 1 A flowchart of an environmental protection hydrological monitoring data collection method based on artificial intelligence is provided in an embodiment of the present invention.

[0028] like Figure 1 As shown, the environmental protection hydrological monitoring data collection method based on artificial intelligence includes steps S101 to S106.

[0029] Step S101: using a first sensor to collect data on a target object to obtain first collected data and using a second sensor to collect data on the target object to obtain second collected data, wherein the first sensor and the second sensor are two different sensors.

[0030] Exemplarily, the target object is an environmental protection hydrological monitoring object, and the first sensor and the second sensor are selected according to the characteristics of the target object and the parameters to be measured. Ensure that the first sensor and the second sensor can effectively sense the relevant data of the target object. For example, the first sensor and the second sensor can be any one of sensors such as a temperature sensor, a humidity sensor, a conductivity sensor, a pH sensor, and a pressure sensor.

[0031] Exemplarily, the first sensor is started, and acquisition parameters (such as sampling frequency, data format, etc.) are set to collect data from the target object. The sensor is ensured to work stably during the acquisition period, and the first acquisition data from time t1 to t2 is recorded. The second sensor is started at the same time as the first sensor, and the same acquisition parameters are set to collect data from the same target object. The second acquisition data from time t1 to t2 is recorded.

[0032] Step S102: performing correlation analysis on the first sensor and the second sensor using the first collected data and the second collected data to obtain a relationship analysis result, wherein the relationship analysis result is used to characterize the correlation relationship between the measurement data of the target object under the first sensor and the second sensor.

[0033] Exemplarily, the first collected data and the second collected data are ensured to be aligned at the same time point. If the data collection time is slightly offset, the timestamp can be adjusted by interpolation method so that the first collected data and the second collected data have corresponding data points at the same time point.

[0034] Exemplarily, the analysis method corresponding to the first collected data and the second collected data is determined from correlation analysis methods such as Pearson correlation coefficient, Spearman correlation coefficient, mutual information, etc., so as to calculate the correlation coefficient between the first collected data and the second collected data according to the selected correlation analysis method, and then analyze the relationship analysis result between the first sensor and the second sensor data according to the calculated correlation coefficient. For example, the value range of the correlation coefficient is from -1 to 1, and a value close to 1 indicates that the relationship analysis result between the first sensor and the second sensor data is a strong positive correlation, a value close to -1 indicates that the relationship analysis result between the first sensor and the second sensor data is a strong negative correlation, and a value close to 0 indicates that the relationship analysis result between the first sensor and the second sensor data is uncorrelated.

[0035] Exemplarily, the first collected data and the second collected data come from two different sensors. By performing correlation analysis on the first collected data and the second collected data, the correlation relationship between the measurement data of the target object under the first sensor and the second sensor is obtained, thereby providing support for subsequent abnormality identification of the first collected data and the second collected data through the correlation relationship.

[0036] In some embodiments, the use of the first collected data and the second collected data to perform correlation analysis on the first sensor and the second sensor to obtain a relationship analysis result includes: using a feature representation layer of a correlation analysis model to perform feature conversion on the first collected data to obtain a corresponding first feature vector and performing feature conversion on the second collected data to obtain a corresponding second feature vector; using the association analysis layer of the correlation analysis model to calculate the degree of difference between the first sensor and the second sensor according to the first feature vector and the second feature vector; using the distance calculation layer of the correlation analysis model to calculate the distance information between the first sensor and the second sensor according to the first collected data and the second collected data; using the correlation calculation layer of the correlation analysis model to determine the correlation value between the first sensor and the second sensor according to the degree of difference, the distance information, the first collected data and the second collected data; using the relationship determination layer of the correlation analysis model to determine the relationship analysis result between the first sensor and the second sensor according to the correlation value; wherein the correlation value is obtained according to the following formula:

[0037]

[0038] in, Indicates the first sensor and the second sensor The correlation value between them, diff represents the degree of difference, dis represents the distance information, represents the first collected data, represents the second collected data, n represents the data volume corresponding to the first collected data or the data volume corresponding to the second collected data, and ln represents the natural logarithm.

[0039] Exemplarily, the correlation analysis model includes a feature representation layer, an association analysis layer, a distance calculation layer, a correlation calculation layer, and a relationship determination layer.

[0040] Exemplarily, a feature representation layer of a correlation analysis model is used to define a mechanism for feature conversion of the first collected data and the second collected data. The feature representation layer can use various methods, such as principal component analysis (PCA), linear discriminant analysis (LDA), or deep learning methods (such as convolutional neural networks, for image data), to process the first collected data, extract key features, and generate a first feature vector. The first feature vector can fully represent the main features and information of the first collected data. Similarly, feature extraction is performed on the second collected data, and a corresponding second feature vector is generated. This process is consistent with the processing process of the first collected data to ensure feature consistency.

[0041] Exemplarily, in the association analysis layer of the correlation analysis model, vector difference, cosine similarity or other distance metrics (such as Euclidean distance) are used to quantify the difference between the first feature vector and the second feature vector.

[0042] Exemplarily, in the distance calculation layer, the distance information between the first collected data and the second collected data is calculated according to the Euclidean distance, the Manhattan distance or the Mahalanobis distance.

[0043] Exemplarily, in the correlation calculation layer, the difference degree, the distance information, the first collected data, and the second collected data are integrated to calculate the correlation value between the first sensor and the second sensor using the following formula, wherein the correlation value is obtained according to the following formula:

[0044]

[0045] in, Indicates the first sensor and the second sensor The correlation value between them, diff indicates the degree of difference, and dis indicates the distance information. represents the first collected data, represents the second collected data, n represents the data volume corresponding to the first collected data or the data volume corresponding to the second collected data, and ln represents the natural logarithm. is the matrix composed of all the first collected data, is a matrix composed of all the second acquisition data.

[0046] Exemplarily, the above formula comprehensively considers the difference degree, distance information, the first collected data and the second collected data, thereby achieving a comprehensive and accurate evaluation of the correlation value between the first sensor and the second sensor.

[0047] Exemplarily, the relationship determination layer of the correlation analysis model is used to compare the correlation value with the preset value. When the correlation value is greater than the preset value, the relationship analysis result between the first sensor and the second sensor is determined to be a strong correlation; when the correlation value is less than the preset value, the relationship analysis result between the first sensor and the second sensor is determined to be a weak correlation.

[0048] Step S103: performing an abnormality analysis on the first collected data and the second collected data according to the relationship analysis result to obtain first abnormal data corresponding to the first collected data and second abnormal data corresponding to the second collected data.

[0049] Exemplarily, the first collected data is identified using a first abnormality recognition model corresponding to the first sensor to obtain first recognition data corresponding to the first collected data, and the second collected data is identified using a second abnormality recognition model corresponding to the second sensor to obtain second recognition data corresponding to the second collected data.

[0050] Exemplarily, a first collection time corresponding to the first identification data and a second collection time corresponding to the second identification data are obtained, and then when the relationship analysis result is a strong correlation, the first associated data corresponding to the second collection data at the first collection time is obtained, and then data relationship analysis is performed based on the first associated data and the first identification data. When the relationship between the first associated data and the first identification data does not satisfy the strong correlation, the corresponding first identification data is determined as the first abnormal data; and the second associated data corresponding to the first collection data at the second collection time is obtained, and then data relationship analysis is performed based on the second associated data and the second identification data. When the relationship between the second associated data and the second identification data does not satisfy the strong correlation, the corresponding second identification data is determined as the second abnormal data.

[0051] Exemplarily, when the relationship analysis result is a weak correlation, the first identification data is directly determined as the first abnormal data, and the second identification data is directly determined as the second abnormal data.

[0052] In some embodiments, performing an abnormality analysis on the first collected data and the second collected data according to the relationship analysis result to obtain first abnormal data corresponding to the first collected data and second abnormal data corresponding to the second collected data includes: performing data clustering on the first collected data to obtain a first target clustering result, and performing data clustering on the second collected data to obtain a second target clustering result; analyzing the first quantity corresponding to each first class cluster in the first target clustering result to obtain a first judgment result corresponding to the first collected data; analyzing the second quantity corresponding to each second class cluster in the second target clustering result to obtain a second judgment result corresponding to the second collected data; performing data alignment on the first judgment result and the second judgment result to obtain a target alignment result; performing an abnormality determination on the first judgment result according to the relationship analysis result combined with the target alignment result to obtain the first abnormal data corresponding to the first collected data; performing an abnormality determination on the second judgment result according to the relationship analysis result combined with the target alignment result to obtain the second abnormal data corresponding to the second collected data.

[0053] Exemplarily, clustering algorithms including K-means, Hierarchical Clustering, DBSCAN, etc. are used to cluster the first collected data to generate a first target clustering result. Similarly, the second collected data are clustered using the selected clustering algorithm to generate a second target clustering result.

[0054] Exemplarily, the number of data points corresponding to each first cluster in the first target clustering result, that is, the first quantity, is counted, and an analysis is performed to obtain a first difference between the first quantities between any two first clusters. When the first difference is greater than a preset value, the first cluster corresponding to the smaller first quantity of the two first clusters is determined as the first judgment result.

[0055] Exemplarily, the number of data points corresponding to each second cluster in the second target clustering result, that is, the second quantity, is counted, and an analysis is performed to obtain a second difference between the second quantities between any two second clusters. When the second difference is greater than a preset value, the second cluster corresponding to the smaller second quantity of the two second clusters is determined as the second judgment result.

[0056] Exemplarily, a first collection time corresponding to each data in the first judgment result and a second collection time corresponding to each data in the second judgment result are obtained, and then when the relationship analysis result is a strong correlation, the first associated data corresponding to the second collection data at the first collection time is obtained, and then based on the correlation coefficient, regression analysis or other statistical methods, a data relationship analysis is performed on the first associated data and the first judgment result; when the relationship between the first associated data and the first judgment result does not satisfy the strong correlation, the corresponding first judgment result is determined as the first abnormal data; and the second associated data corresponding to the first collection data at the second collection time is obtained, and then based on the correlation coefficient, regression analysis or other statistical methods, a data relationship analysis is performed on the second associated data and the second judgment result; when the relationship between the second associated data and the second judgment result does not satisfy the strong correlation, the corresponding second judgment result is determined as the second abnormal data.

[0057] Exemplarily, when the relationship analysis result is a weak correlation, the first judgment result is directly determined as the first abnormal data, and the second judgment result is directly determined as the second abnormal data.

[0058] In some embodiments, clustering the first collected data to obtain a first target clustering result, and clustering the second collected data to obtain a second target clustering result, include: initializing the cluster center corresponding to the first collected data to obtain a first cluster center, and calculating a first distance from the first collected data to the first cluster center; clustering the first collected data according to the first distance and the first cluster center to obtain a first clustering result; optimizing the first cluster center according to the first clustering result to obtain a second cluster center, and clustering the first collected data according to the second cluster center to obtain a second clustering result; optimizing the first cluster center according to the second cluster result to obtain a second clustering result; and optimizing the first cluster center according to the second cluster center to obtain a second clustering result. The first target clustering result is obtained by optimizing the second clustering center based on the clustering result to obtain a third clustering center, and clustering the first collected data according to the third clustering center to obtain a third clustering result; when the second clustering result is the same as the third clustering result, the first target clustering result is obtained from the first clustering result and the second clustering result; when the second clustering result is different from the third clustering result, the third clustering center is continuously optimized using the third clustering result to obtain a fourth clustering center, and clustering the first collected data according to the fourth clustering center to obtain a fourth clustering result, until the third clustering result is the same as the fourth clustering result; at least from the first clustering result, the first target clustering result is obtained. The first target clustering result is obtained from the first target clustering result, the second clustering result and the third clustering result; the cluster center corresponding to the second collected data is initialized to obtain the fifth cluster center, and the second distance corresponding to the first collected data to the fifth cluster center is calculated; the second collected data is clustered according to the second distance and the fifth cluster center to obtain the fifth clustering result; the fifth cluster center is optimized according to the fifth clustering result to obtain the sixth cluster center, and the second collected data is clustered according to the sixth cluster center to obtain the sixth clustering result; the sixth cluster center is optimized according to the sixth clustering result to obtain the seventh cluster center, and the seventh cluster center is optimized according to the seventh cluster center. The center clusters the second collected data to obtain a seventh clustering result; when the sixth clustering result is the same as the seventh clustering result, the second target clustering result is obtained from the fifth clustering result and the sixth clustering result; when the sixth clustering result is different from the seventh clustering result, the seventh clustering center is continuously optimized using the seventh clustering result to obtain an eighth clustering center, and the second collected data is clustered according to the eighth clustering center to obtain an eighth clustering result, until the seventh clustering result is the same as the eighth clustering result; the second target clustering result is obtained from at least the fifth clustering result, the sixth clustering result and the seventh clustering result.

[0059] Exemplarily, one data is arbitrarily selected from the first collected data as the first cluster center, and then the first distance from each data in the first collected data to the first cluster center is calculated; when the first distance is less than the preset distance, the data corresponding to the first distance is determined to belong to the first cluster center, and when the first distance is greater than or equal to the preset distance, it is determined that the data corresponding to the first distance does not belong to the first cluster center, thereby obtaining a first clustering result, and the first clustering result contains at most two clusters, one cluster contains all the data belonging to the first cluster center, and the other cluster contains all the data not belonging to the first cluster center.

[0060] Exemplarily, average values ​​are calculated respectively according to the two clusters in the first clustering result, and then the average values ​​are determined as the second clustering center respectively, so as to calculate the distance from each data in the first collected data to the second clustering center to obtain the second clustering result, and so on, according to the second clustering result, the second clustering center is optimized to obtain the third clustering center, and the first collected data is clustered according to the third clustering center to obtain the third clustering result.

[0061] Exemplarily, when the second clustering result is the same as the third clustering result, it means that the maximum clustering cluster corresponding to the first collected data has been determined, and then for each cluster in the first clustering result, the number of data points corresponding to it is counted to form the frequency distribution of each cluster. This can be done by calculating the ratio of the number of data points in each cluster to the total number of data points. Thus, according to the frequency distribution in the first clustering result, the probability of each cluster is calculated, and then the entropy value calculation formula is substituted to calculate the first entropy value. The first entropy value quantifies the data uncertainty or confusion of the first clustering result. The higher the value, the greater the amount of information and the wider the data distribution. And for each cluster in the second clustering result, the number of data points corresponding to it is counted to form the frequency distribution of each cluster. This can be done by calculating the ratio of the number of data points in each cluster to the total number of data points. Thus, according to the frequency distribution in the second clustering result, the probability of each cluster is calculated, and then the entropy value calculation formula is substituted to calculate the second entropy value. The second entropy value quantifies the data uncertainty or confusion of the second clustering result. The higher the value, the greater the amount of information and the wider the data distribution. Therefore, the clustering result corresponding to the maximum of the first entropy value and the second entropy value is determined as the first target clustering result.

[0062] Exemplarily, when the second clustering result and the third clustering result are different, it means that the maximum clustering class corresponding to the first collected data has not been determined, and then the third clustering result is continued to be used to optimize the third clustering center to obtain the fourth clustering center, and the first collected data is clustered according to the fourth clustering center to obtain the fourth clustering result, until the third clustering result and the fourth clustering result are the same, that is, until the maximum clustering class corresponding to the first collected data is obtained. Therefore, when the third clustering result and the fourth clustering result are the same, the first target clustering result is obtained from the first clustering result, the second clustering result and the third clustering result; when the third clustering result and the fourth clustering result are different, it is necessary to continue to optimize the fourth clustering center according to the fourth clustering result to obtain a new clustering result. When the new clustering result and the fourth clustering result are the same, the first target clustering result is obtained from the first clustering result, the second clustering result, the third clustering result and the fourth clustering result.

[0063] It should be noted that the method of obtaining the first target clustering result from the first clustering result, the second clustering result, the third clustering result and the fourth clustering result, and the method of obtaining the first target clustering result from the first clustering result, the second clustering result and the third clustering result is the same as the method of obtaining the first target clustering result from the first clustering result, the second clustering result, and so will not be repeated here.

[0064] It should be noted that the cluster center corresponding to the second collected data is initialized to obtain the fifth cluster center, and the second distance corresponding to the first collected data to the fifth cluster center is calculated; the second collected data is clustered according to the second distance and the fifth cluster center to obtain the fifth cluster result; the fifth cluster center is optimized according to the fifth cluster result to obtain the sixth cluster center, and the second collected data is clustered according to the sixth cluster center to obtain the sixth cluster result; the sixth cluster center is optimized according to the sixth cluster result to obtain the seventh cluster center, and the second collected data is clustered according to the seventh cluster center. The data is clustered to obtain the seventh clustering result; when the sixth clustering result and the seventh clustering result are the same, the second target clustering result is obtained from the fifth clustering result and the sixth clustering result; when the sixth clustering result and the seventh clustering result are different, the seventh clustering center is optimized using the seventh clustering result to obtain the eighth clustering center, and the second collected data is clustered according to the eighth clustering center to obtain the eighth clustering result, until the seventh clustering result and the eighth clustering result are the same; the second target clustering result is obtained from at least the fifth clustering result, the sixth clustering result and the seventh clustering result. The process is the same as that of obtaining the first target clustering result, which will not be repeated here.

[0065] In some embodiments, when the second clustering result and the third clustering result are the same, the first target clustering result is obtained from the first clustering result and the second clustering result, including: calculating the third distance between each first data in the third cluster in the first clustering result and the first cluster center corresponding to the third cluster; calculating the first center distance between the first cluster center and the second cluster center corresponding to the remaining clusters in the first clustering result; calculating the fourth distance between the second data in the fourth cluster in the second cluster result and the third cluster center corresponding to the fourth cluster; calculating the second center distance between the third cluster center and the fourth cluster center corresponding to the remaining clusters in the second clustering result; obtaining the first clustering quality corresponding to the first clustering result according to the third distance and the first center distance; obtaining the second clustering quality corresponding to the second clustering result according to the fourth distance and the second center distance; obtaining the first target clustering result according to the first clustering quality and the second clustering quality; wherein the first clustering quality and the second clustering quality are obtained according to the following formula:

[0066]

[0067]

[0068] in, represents the first clustering quality, max represents the maximum value obtained, represents the i-th first data in the third cluster and the first cluster center corresponding to the i-th first data The third distance between represents the i-th first data in the third cluster The corresponding first type of cluster center and the tth cluster center of the second type the first center distance between; represents the second clustering quality, represents the jth second data in the fourth cluster and the third cluster center corresponding to the jth second data The fourth distance between represents the jth second data in the fourth cluster The corresponding third cluster center and the yth fourth cluster center The second center distance between.

[0069] Exemplarily, the third distance between each first data in the third cluster in the first clustering result and the center of the first cluster corresponding to the third cluster is calculated using Euclidean distance or Manhattan distance; and the first center distance between the center of the first cluster and the center of the second cluster corresponding to the remaining clusters in the first clustering result is calculated using Euclidean distance or Manhattan distance.

[0070] Exemplarily, the fourth distance between the second data in the fourth cluster in the second clustering result and the center of the third cluster corresponding to the fourth cluster is calculated using Euclidean distance or Manhattan distance; and the second center distance between the center of the third cluster and the center of the fourth cluster corresponding to the remaining clusters in the second clustering result is calculated using Euclidean distance or Manhattan distance.

[0071] Exemplarily, the first clustering quality corresponding to the first clustering result is obtained by using the third distance and the first center distance according to the following formula:

[0072]

[0073] in, Indicates the first clustering quality, max indicates the maximum value, Indicates the first data of the i-th cluster in the third class The first cluster center corresponding to the i-th first data The third distance between Indicates the first data of the third cluster i The corresponding first cluster center and the tth second-class cluster center The first center distance between .

[0074] Exemplarily, the maximum value of the ratio between the third distance and the first center distance is obtained, so that the maximum value of the intra-class and inter-class distances in the first clustering result is indicated according to the maximum value, so that the first clustering quality is determined according to the first clustering quality. Determine the reliability of the first clustering result.

[0075] Exemplarily, the second clustering quality corresponding to the second clustering result is obtained by using the fourth distance and the second center distance according to the following formula:

[0076]

[0077] Among them, max means obtaining the maximum value; represents the second clustering quality, Indicates the jth second data in the fourth cluster The third cluster center corresponding to the jth second data The fourth distance between Indicates the jth second data in the fourth cluster The corresponding third cluster center and the yth fourth cluster center The second center distance between .

[0078] Exemplarily, the maximum value of the ratio between the fourth distance and the second center distance is obtained, so that the maximum value of the intra-class and inter-class distances in the second clustering result is indicated according to the maximum value, so that the second clustering quality is determined according to the second clustering quality. Determine the reliability of the second clustering result.

[0079] Exemplarily, the maximum value of the first clustering quality and the second clustering quality is taken, and the clustering result corresponding to the maximum value is determined as the first target clustering result.

[0080] It should be noted that in the present application, "at least the first target clustering result is obtained from the first clustering result, the second clustering result and the third clustering result", "when the sixth clustering result and the seventh clustering result are the same, the second target clustering result is obtained from the fifth clustering result and the sixth clustering result;", "at least the second target clustering result is obtained from the fifth clustering result, the sixth clustering result and the seventh clustering result", all of which can adopt the above-mentioned scheme of determining the target clustering result by clustering quality, and the present application will not repeat them here.

[0081] In some embodiments, the first judgment result is determined to be abnormal based on the relationship analysis result in combination with the target alignment result to obtain the first abnormal data corresponding to the first collected data, including: when the relationship analysis result is a strong correlation, the data belonging to the first collected data in the target alignment result is removed from the first judgment result to obtain the first abnormal data corresponding to the first collected data; when the relationship analysis result is a weak correlation, the first judgment result is directly determined as the first abnormal data corresponding to the first collected data.

[0082] Exemplarily, the target alignment result is used to indicate that the collected data corresponding to the first sensor and the second sensor at the same time are both abnormal. If the relationship analysis result is a strong correlation, the data that does not belong to the target alignment result in the first judgment result needs to be determined as the first abnormal data corresponding to the first collected data, that is, the data belonging to the first sensor in the target alignment result is possible real abnormal data, rather than collection error data. Therefore, the data belonging to the first collected data in the target alignment result is removed from the first judgment result, thereby obtaining the first abnormal data corresponding to the first collected data. That is, the first abnormal data is the abnormal data collected by the first sensor on the target object.

[0083] Exemplarily, when the relationship analysis result is a weak correlation, the first judgment result is directly determined as the first abnormal data corresponding to the first collected data.

[0084] Specifically, based on the relationship analysis results, the first judgment result of the first sensor is judged as abnormal, which can effectively identify and filter the erroneous data that appears in the data collection process, significantly improve the quality of the data, ensure the accuracy of subsequent analysis, and avoid misleading due to erroneous data. This process not only helps maintain the reliability of data, but also provides a more solid foundation for decision-making.

[0085] In some embodiments, the second judgment result is determined to be abnormal based on the relationship analysis result in combination with the target alignment result to obtain the second abnormal data corresponding to the second collected data, including: when the relationship analysis result is the strong correlation, the data belonging to the second collected data in the target alignment result is removed from the second judgment result to obtain the second abnormal data corresponding to the second collected data; when the relationship analysis result is the weak correlation, the second judgment result is directly determined as the second abnormal data corresponding to the second collected data.

[0086] Exemplarily, the target alignment result is used to indicate that the collected data corresponding to the first sensor and the second sensor at the same time are both abnormal. When the relationship analysis result is a strong correlation, the data that does not belong to the target alignment result in the second judgment result needs to be determined as the second abnormal data corresponding to the second collected data, that is, the data belonging to the second sensor in the target alignment result is possible real abnormal data, rather than collection error data. Therefore, the data belonging to the second collected data in the target alignment result is removed from the second judgment result, thereby obtaining the second abnormal data corresponding to the second collected data. That is, the second abnormal data is the collection abnormal data of the target object by the second sensor.

[0087] Exemplarily, when the relationship analysis result is a weak correlation, the second judgment result is directly determined as the second abnormal data corresponding to the second collected data.

[0088] Specifically, according to the relationship analysis results, the second judgment result of the second sensor is judged as abnormal, which can effectively identify and filter the erroneous data that appears in the data collection process, significantly improve the quality of the data, ensure the accuracy of subsequent analysis, and avoid misleading due to erroneous data. This process not only helps to maintain the reliability of data, but also provides a more solid foundation for decision-making.

[0089] Step S104: performing data prediction on the first abnormal data according to the first collected data to obtain first target data corresponding to the first sensor.

[0090] Exemplarily, the first position information corresponding to the first abnormal data is obtained, and then the first associated data corresponding to the time period before the first position information is obtained from the first collected data, and then data fitting is performed on the first associated data to obtain a fitting curve, and then data prediction is performed on the first abnormal data based on the first position information and the fitting curve, so as to obtain the first predicted data corresponding to the first position information, and then the first predicted data is replaced with the first position information corresponding to the first abnormal data, so as to obtain the first target data corresponding to the first sensor.

[0091] In some embodiments, the performing data prediction on the first abnormal data based on the first collected data to obtain first target data corresponding to the first sensor includes: obtaining first position information corresponding to the first abnormal data in the first collected data, and cutting the first collected data according to the first position information to obtain first cut data; performing data prediction on the first position information using a neural network algorithm based on the first cut data to obtain first predicted data corresponding to the first position information; and determining the first target data corresponding to the first sensor based on the first predicted data and the first cut data.

[0092] Exemplarily, the first position information corresponding to the first abnormal data in the first collected data is confirmed and marked. The first collected data is cut according to the extracted first position information to obtain first cut data, and the first cut data generally includes several data points before and after the first abnormal data to capture the context of the data. The first cut data includes normal data around the first abnormal data to facilitate subsequent analysis thereof.

[0093] Exemplarily, a neural network model is designed, and the first cutting data is input into the neural network for training, so that the first position information is predicted according to the trained neural network model to obtain the first predicted data corresponding to the first position information. The first predicted data is then replaced with the first position information corresponding to the first abnormal data, so as to obtain the first target data corresponding to the first sensor.

[0094] Step S105: performing data prediction on the second abnormal data according to the second collected data to obtain second target data corresponding to the second sensor.

[0095] Exemplarily, the second position information corresponding to the second abnormal data is obtained, and then the second associated data corresponding to the time period before the second position information is obtained from the second collected data, and then data fitting is performed on the second associated data to obtain a fitting curve, and then data prediction is performed on the second abnormal data based on the second position information and the fitting curve, so as to obtain second predicted data corresponding to the second position information, and then the second predicted data is replaced with the second position information corresponding to the second abnormal data, so as to obtain second target data corresponding to the second sensor.

[0096] In some embodiments, the performing data prediction on the second abnormal data based on the second collected data to obtain second target data corresponding to the second sensor includes: obtaining second position information corresponding to the second abnormal data in the second collected data, and cutting the second collected data according to the second position information to obtain second cut data; performing data prediction on the second position information using the neural network algorithm based on the second cut data to obtain second predicted data corresponding to the second position information; and determining the second target data corresponding to the second sensor based on the second predicted data and the second cut data.

[0097] Exemplarily, the second position information corresponding to the second abnormal data in the second collected data is confirmed and marked. The second collected data is cut according to the extracted second position information to obtain second cut data, and the second cut data generally includes several data points before and after the second abnormal data to capture the context of the data. The first cut data includes normal data around the second abnormal data to facilitate subsequent analysis thereof.

[0098] Exemplarily, a neural network model is designed, and the second cutting data is input into the neural network for training, so that the second position information is predicted according to the trained neural network model to obtain the second predicted data corresponding to the second position information. Then, the second predicted data is replaced with the second position information corresponding to the second abnormal data, so as to obtain the second target data corresponding to the second sensor.

[0099] Step S106: Determine a target acquisition result corresponding to the target object according to the first target data and the second target data.

[0100] Exemplarily, first target data corresponding to the first sensor and second target data corresponding to the second sensor are determined as target acquisition results corresponding to the target object.

[0101] See also Figure 2 , Figure 2An artificial intelligence-based environmental protection hydrological monitoring data acquisition system 200 is provided in an embodiment of the present application. The artificial intelligence-based environmental protection hydrological monitoring data acquisition system 200 includes a data acquisition module 201, a data analysis module 202, an abnormal classification module 203, a first prediction module 204, a second prediction module 205, and a result determination module 206, wherein the data acquisition module 201 is used to use a first sensor to perform data acquisition on a target object to obtain first acquisition data and to use a second sensor to perform data acquisition on the target object to obtain second acquisition data, wherein the first sensor and the second sensor are two different sensors; the data analysis module 202 is used to use the first acquisition data and the second acquisition data to perform correlation analysis on the first sensor and the second sensor to obtain a relationship analysis result, and the relationship analysis The result is used to characterize the correlation relationship between the measurement data of the target object under the first sensor and the second sensor; the abnormality classification module 203 is used to perform abnormality analysis on the first collected data and the second collected data according to the relationship analysis result, and obtain first abnormal data corresponding to the first collected data and second abnormal data corresponding to the second collected data; the first prediction module 204 is used to perform data prediction on the first abnormal data according to the first collected data, and obtain first target data corresponding to the first sensor; the second prediction module 205 is used to perform data prediction on the second abnormal data according to the second collected data, and obtain second target data corresponding to the second sensor; the result determination module 206 is used to determine the target collection result corresponding to the target object according to the first target data and the second target data.

[0102] In some implementations, the data analysis module 202, in the process of performing correlation analysis on the first sensor and the second sensor using the first collected data and the second collected data to obtain a relationship analysis result, executes:

[0103] Using a feature representation layer of a correlation analysis model, performing feature conversion on the first collected data to obtain a corresponding first feature vector and performing feature conversion on the second collected data to obtain a corresponding second feature vector;

[0104] Calculating the degree of difference between the first sensor and the second sensor according to the first feature vector and the second feature vector using the association analysis layer of the correlation analysis model;

[0105] Calculate the distance information between the first sensor and the second sensor according to the first collected data and the second collected data using the distance calculation layer of the correlation analysis model;

[0106] Determine a correlation value between the first sensor and the second sensor according to the difference degree, the distance information, the first collected data, and the second collected data using a correlation calculation layer of the correlation analysis model;

[0107] Determine the relationship analysis result between the first sensor and the second sensor according to the association value using the relationship determination layer of the correlation analysis model;

[0108] The correlation value is obtained according to the following formula:

[0109]

[0110] in, Indicates the first sensor and the second sensor The correlation value between them, diff represents the degree of difference, dis represents the distance information, represents the first collected data, represents the second collected data, n represents the data volume corresponding to the first collected data or the data volume corresponding to the second collected data, and ln represents the natural logarithm.

[0111] In some implementations, the abnormality classification module 203, in the process of performing abnormality analysis on the first collected data and the second collected data according to the relationship analysis result to obtain first abnormal data corresponding to the first collected data and second abnormal data corresponding to the second collected data, executes:

[0112] Performing data clustering on the first collected data to obtain a first target clustering result, and performing data clustering on the second collected data to obtain a second target clustering result;

[0113] Analyze the first quantity corresponding to each first cluster in the first target clustering result to obtain a first judgment result corresponding to the first collected data;

[0114] Analyze the second quantity corresponding to each second cluster in the second target clustering result to obtain a second judgment result corresponding to the second collected data;

[0115] Performing data alignment on the first judgment result and the second judgment result to obtain a target alignment result;

[0116] Determine an abnormality of the first judgment result according to the relationship analysis result combined with the target alignment result, and obtain the first abnormal data corresponding to the first collected data;

[0117] An abnormality determination is performed on the second judgment result according to the relationship analysis result in combination with the target alignment result to obtain the second abnormal data corresponding to the second collected data.

[0118] In some implementations, the abnormal classification module 203 performs, during the process of clustering the first collected data to obtain a first target clustering result and clustering the second collected data to obtain a second target clustering result:

[0119] Initializing the cluster center corresponding to the first collected data to obtain a first cluster center, and calculating a first distance from the first collected data to the first cluster center;

[0120] Clustering the first collected data according to the first distance and the first cluster center to obtain a first clustering result;

[0121] According to the first clustering result, the first clustering center is optimized to obtain a second clustering center, and according to the second clustering center, the first collected data is clustered to obtain a second clustering result;

[0122] According to the second clustering result, the second cluster center is optimized to obtain a third cluster center, and according to the third cluster center, the first collected data is clustered to obtain a third clustering result;

[0123] When the second clustering result and the third clustering result are the same, obtaining the first target clustering result from the first clustering result and the second clustering result;

[0124] When the second clustering result is different from the third clustering result, the third clustering center is optimized by using the third clustering result to obtain a fourth clustering center, and the first collected data is clustered according to the fourth clustering center to obtain a fourth clustering result, until the third clustering result is the same as the fourth clustering result;

[0125] Obtaining the first target clustering result at least from the first clustering result, the second clustering result, and the third clustering result;

[0126] Initializing the cluster center corresponding to the second collected data to obtain a fifth cluster center, and calculating a second distance from the first collected data to the fifth cluster center;

[0127] Clustering the second collected data according to the second distance and the fifth cluster center to obtain a fifth clustering result;

[0128] According to the fifth clustering result, the fifth cluster center is optimized to obtain a sixth clustering center, and according to the sixth clustering center, data clustering is performed on the second collected data to obtain a sixth clustering result;

[0129] According to the sixth clustering result, the sixth cluster center is optimized to obtain a seventh cluster center, and according to the seventh cluster center, the second collected data is clustered to obtain a seventh clustering result;

[0130] When the sixth clustering result and the seventh clustering result are the same, obtaining the second target clustering result from the fifth clustering result and the sixth clustering result;

[0131] When the sixth clustering result is different from the seventh clustering result, the seventh clustering center is continuously optimized using the seventh clustering result to obtain an eighth clustering center, and the second collected data is clustered according to the eighth clustering center to obtain an eighth clustering result, until the seventh clustering result is the same as the eighth clustering result;

[0132] The second target clustering result is obtained from at least the fifth clustering result, the sixth clustering result, and the seventh clustering result.

[0133] In some implementations, the abnormal classification module 203 performs, in the process of obtaining the first target clustering result from the first clustering result and the second clustering result when the second clustering result and the third clustering result are the same:

[0134] Calculating a third distance between each first data in the third cluster in the first clustering result and the first cluster center corresponding to the third cluster;

[0135] Calculating a first center distance between the center of the first cluster and the center of the second cluster corresponding to the remaining clusters in the first clustering result;

[0136] Calculating a fourth distance between the second data in the fourth cluster in the second clustering result and the center of the third cluster corresponding to the fourth cluster;

[0137] Calculating a second center distance between the center of the third cluster and the center of the fourth cluster corresponding to the remaining clusters in the second clustering result;

[0138] Obtaining a first clustering quality corresponding to the first clustering result according to the third distance and the first center distance;

[0139] Obtaining a second clustering quality corresponding to the second clustering result according to the fourth distance and the second center distance;

[0140] Obtaining the first target clustering result according to the first clustering quality and the second clustering quality;

[0141] The first clustering quality and the second clustering quality are obtained according to the following formula:

[0142]

[0143]

[0144] in, represents the first clustering quality, max represents the maximum value obtained, represents the i-th first data in the third cluster and the first cluster center corresponding to the i-th first data The third distance between represents the i-th first data in the third cluster The corresponding first type of cluster center and the tth cluster center of the second type the first center distance between; represents the second clustering quality, represents the jth second data in the fourth cluster and the third cluster center corresponding to the jth second data The fourth distance between represents the jth second data in the fourth cluster The corresponding third cluster center and the yth fourth cluster center The second center distance between.

[0145] In some implementations, the abnormality classification module 203, in the process of determining the abnormality of the first judgment result according to the relationship analysis result combined with the target alignment result to obtain the first abnormal data corresponding to the first collected data, performs:

[0146] When the relationship analysis result is a strong correlation, the data belonging to the first collected data in the target alignment result is removed from the first judgment result to obtain the first abnormal data corresponding to the first collected data;

[0147] When the relationship analysis result is a weak correlation, the first judgment result is directly determined as the first abnormal data corresponding to the first collected data.

[0148] In some implementations, the abnormality classification module 203, in the process of determining the abnormality of the second judgment result according to the relationship analysis result in combination with the target alignment result to obtain the second abnormal data corresponding to the second collected data, performs:

[0149] When the relationship analysis result is the strong correlation, the data belonging to the second collected data in the target alignment result is removed from the second judgment result to obtain the second abnormal data corresponding to the second collected data;

[0150] When the relationship analysis result is the weak correlation, the second judgment result is directly determined as the second abnormal data corresponding to the second collected data.

[0151] In some implementations, the first prediction module 204 performs, during the process of performing data prediction on the first abnormal data according to the first collected data to obtain the first target data corresponding to the first sensor:

[0152] Obtaining first position information corresponding to the first abnormal data in the first collected data, and cutting the first collected data according to the first position information to obtain first cut data;

[0153] Performing data prediction on the first position information using a neural network algorithm according to the first cutting data to obtain first prediction data corresponding to the first position information;

[0154] The first target data corresponding to the first sensor is determined according to the first prediction data and the first cutting data.

[0155] In some implementations, the second prediction module 205 performs, during the process of performing data prediction on the second abnormal data according to the second collected data to obtain the second target data corresponding to the second sensor:

[0156] Obtaining second position information corresponding to the second abnormal data in the second collected data, and cutting the second collected data according to the second position information to obtain second cut data;

[0157] Performing data prediction on the second position information using the neural network algorithm according to the second cutting data to obtain second prediction data corresponding to the second position information;

[0158] The second target data corresponding to the second sensor is determined according to the second prediction data and the second cutting data.

[0159] In some embodiments, the artificial intelligence-based environmental protection hydrological monitoring data acquisition system 200 can be applied to terminal devices.

[0160] 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 environmental protection hydrological monitoring data acquisition system 200 based on artificial intelligence described above can refer to the corresponding process in the aforementioned embodiment of the environmental protection hydrological monitoring data acquisition method based on artificial intelligence, and will not be repeated here.

[0161] 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 one of the environmental protection hydrological monitoring data collection methods based on artificial intelligence provided in the description of the embodiment of the present invention.

[0162] 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.

[0163] 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.

[0164] 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.

[0165] 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 description is only a specific embodiment 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. An environmental protection hydrological monitoring data collection method based on artificial intelligence, characterized in that: The method comprises: Using a first sensor to collect data from a target object to obtain first collected data and using a second sensor to collect data from the target object to obtain second collected data, wherein the first sensor and the second sensor are two different sensors; Performing a correlation analysis on the first sensor and the second sensor using the first collected data and the second collected data to obtain a relationship analysis result, wherein the relationship analysis result is used to characterize the association relationship between the measurement data of the target object under the first sensor and the second sensor; Performing an abnormality analysis on the first collected data and the second collected data according to the relationship analysis result to obtain first abnormal data corresponding to the first collected data and second abnormal data corresponding to the second collected data; Performing data prediction on the first abnormal data according to the first collected data to obtain first target data corresponding to the first sensor; Performing data prediction on the second abnormal data according to the second collected data to obtain second target data corresponding to the second sensor; Determine a target acquisition result corresponding to the target object according to the first target data and the second target data; The performing correlation analysis on the first sensor and the second sensor using the first collected data and the second collected data to obtain a relationship analysis result includes: Using a feature representation layer of a correlation analysis model, performing feature conversion on the first collected data to obtain a corresponding first feature vector and performing feature conversion on the second collected data to obtain a corresponding second feature vector; Calculating the degree of difference between the first sensor and the second sensor according to the first feature vector and the second feature vector using the association analysis layer of the correlation analysis model; Calculate the distance information between the first sensor and the second sensor according to the first collected data and the second collected data using the distance calculation layer of the correlation analysis model; Determine a correlation value between the first sensor and the second sensor according to the difference degree, the distance information, the first collected data, and the second collected data using a correlation calculation layer of the correlation analysis model; Determine the relationship analysis result between the first sensor and the second sensor according to the association value using the relationship determination layer of the correlation analysis model; The correlation value is obtained according to the following formula: ； in, Indicates the first sensor and the second sensor The correlation value between them, diff represents the degree of difference, dis represents the distance information, represents the first collected data, represents the second collected data, n represents the data volume corresponding to the first collected data or the data volume corresponding to the second collected data, and ln represents the natural logarithm; The step of performing data prediction on the first abnormal data according to the first collected data to obtain first target data corresponding to the first sensor includes: Obtaining first position information corresponding to the first abnormal data in the first collected data, and cutting the first collected data according to the first position information to obtain first cut data; Performing data prediction on the first position information using a neural network algorithm according to the first cutting data to obtain first prediction data corresponding to the first position information; The first target data corresponding to the first sensor is determined according to the first prediction data and the first cutting data.

2. The method according to claim 1, characterized in that The performing anomaly analysis on the first collected data and the second collected data according to the relationship analysis result to obtain first abnormal data corresponding to the first collected data and second abnormal data corresponding to the second collected data includes: Performing data clustering on the first collected data to obtain a first target clustering result, and performing data clustering on the second collected data to obtain a second target clustering result; Analyze the first quantity corresponding to each first cluster in the first target clustering result to obtain a first judgment result corresponding to the first collected data; Analyze the second quantity corresponding to each second cluster in the second target clustering result to obtain a second judgment result corresponding to the second collected data; Performing data alignment on the first judgment result and the second judgment result to obtain a target alignment result; Determine an abnormality of the first judgment result according to the relationship analysis result combined with the target alignment result, and obtain the first abnormal data corresponding to the first collected data; An abnormality determination is performed on the second judgment result according to the relationship analysis result in combination with the target alignment result to obtain the second abnormal data corresponding to the second collected data.

3. The method according to claim 2, characterized in that The performing data clustering on the first collected data to obtain a first target clustering result, and the performing data clustering on the second collected data to obtain a second target clustering result, include: Initializing the cluster center corresponding to the first collected data to obtain a first cluster center, and calculating a first distance from the first collected data to the first cluster center; Clustering the first collected data according to the first distance and the first cluster center to obtain a first clustering result; According to the first clustering result, the first clustering center is optimized to obtain a second clustering center, and according to the second clustering center, the first collected data is clustered to obtain a second clustering result; According to the second clustering result, the second cluster center is optimized to obtain a third cluster center, and according to the third cluster center, the first collected data is clustered to obtain a third clustering result; When the second clustering result and the third clustering result are the same, obtaining the first target clustering result from the first clustering result and the second clustering result; When the second clustering result is different from the third clustering result, the third clustering center is optimized by using the third clustering result to obtain a fourth clustering center, and the first collected data is clustered according to the fourth clustering center to obtain a fourth clustering result, until the third clustering result is the same as the fourth clustering result; Obtaining the first target clustering result at least from the first clustering result, the second clustering result, and the third clustering result; Initializing the cluster center corresponding to the second collected data to obtain a fifth cluster center, and calculating a second distance from the first collected data to the fifth cluster center; Clustering the second collected data according to the second distance and the fifth cluster center to obtain a fifth clustering result; According to the fifth clustering result, the fifth cluster center is optimized to obtain a sixth clustering center, and according to the sixth clustering center, data clustering is performed on the second collected data to obtain a sixth clustering result; According to the sixth clustering result, the sixth cluster center is optimized to obtain a seventh cluster center, and according to the seventh cluster center, the second collected data is clustered to obtain a seventh clustering result; When the sixth clustering result and the seventh clustering result are the same, obtaining the second target clustering result from the fifth clustering result and the sixth clustering result; When the sixth clustering result is different from the seventh clustering result, the seventh clustering center is continuously optimized using the seventh clustering result to obtain an eighth clustering center, and the second collected data is clustered according to the eighth clustering center to obtain an eighth clustering result, until the seventh clustering result is the same as the eighth clustering result; The second target clustering result is obtained from at least the fifth clustering result, the sixth clustering result, and the seventh clustering result.

4. The method according to claim 3, characterized in that When the second clustering result and the third clustering result are the same, obtaining the first target clustering result from the first clustering result and the second clustering result includes: Calculating a third distance between each first data in the third cluster in the first clustering result and the first cluster center corresponding to the third cluster; Calculating a first center distance between the center of the first cluster and the center of the second cluster corresponding to the remaining clusters in the first clustering result; Calculating a fourth distance between the second data in the fourth cluster in the second clustering result and the center of the third cluster corresponding to the fourth cluster; Calculating a second center distance between the center of the third cluster and the center of the fourth cluster corresponding to the remaining clusters in the second clustering result; Obtaining a first clustering quality corresponding to the first clustering result according to the third distance and the first center distance; Obtaining a second clustering quality corresponding to the second clustering result according to the fourth distance and the second center distance; Obtaining the first target clustering result according to the first clustering quality and the second clustering quality; The first clustering quality and the second clustering quality are obtained according to the following formula: ； ； in, represents the first clustering quality, max represents the maximum value obtained, represents the i-th first data in the third cluster and the first cluster center corresponding to the i-th first data The third distance between represents the i-th first data in the third cluster The corresponding first type of cluster center and the tth cluster center of the second type the first center distance between; represents the second clustering quality, represents the jth second data in the fourth cluster and the third cluster center corresponding to the jth second data The fourth distance between represents the jth second data in the fourth cluster The corresponding third cluster center and the yth fourth cluster center The second center distance between.

5. The method according to claim 2, characterized in that: The step of determining an abnormality of the first judgment result according to the relationship analysis result in combination with the target alignment result to obtain the first abnormal data corresponding to the first collected data includes: When the relationship analysis result is a strong correlation, the data belonging to the first collected data in the target alignment result is removed from the first judgment result to obtain the first abnormal data corresponding to the first collected data; When the relationship analysis result is a weak correlation, the first judgment result is directly determined as the first abnormal data corresponding to the first collected data.

6. The method according to claim 5, characterized in that The step of determining an abnormality of the second judgment result according to the relationship analysis result in combination with the target alignment result to obtain the second abnormal data corresponding to the second collected data includes: When the relationship analysis result is the strong correlation, the data belonging to the second collected data in the target alignment result is removed from the second judgment result to obtain the second abnormal data corresponding to the second collected data; When the relationship analysis result is the weak correlation, the second judgment result is directly determined as the second abnormal data corresponding to the second collected data.

7. The method according to claim 1, characterized in that The performing data prediction on the second abnormal data according to the second collected data to obtain second target data corresponding to the second sensor includes: Obtaining second position information corresponding to the second abnormal data in the second collected data, and cutting the second collected data according to the second position information to obtain second cut data; Performing data prediction on the second position information using the neural network algorithm according to the second cutting data to obtain second prediction data corresponding to the second position information; The second target data corresponding to the second sensor is determined according to the second prediction data and the second cutting data.

8. An artificial intelligence-based environmental protection hydrological monitoring data acquisition system, used to implement the artificial intelligence-based environmental protection hydrological monitoring data acquisition method as described in any one of claims 1 to 7, characterized in that: include: A data acquisition module, used to acquire first acquired data by using a first sensor to acquire data from a target object and to acquire second acquired data by using a second sensor to acquire data from the target object, wherein the first sensor and the second sensor are two different sensors; a data analysis module, configured to perform a correlation analysis on the first sensor and the second sensor using the first collected data and the second collected data to obtain a relationship analysis result, wherein the relationship analysis result is used to characterize the correlation relationship between the measurement data of the target object under the first sensor and the second sensor; an abnormality classification module, used for performing abnormality analysis on the first collected data and the second collected data according to the relationship analysis result, and obtaining first abnormal data corresponding to the first collected data and second abnormal data corresponding to the second collected data; A first prediction module, configured to perform data prediction on the first abnormal data according to the first collected data, and obtain first target data corresponding to the first sensor; A second prediction module, configured to perform data prediction on the second abnormal data according to the second collected data, and obtain second target data corresponding to the second sensor; A result determination module is used to determine a target acquisition result corresponding to the target object according to the first target data and the second target data.

Images