A method and system for monitoring sewage treatment data in water conservancy projects

Abstract

The embodiment of the present invention provides a method and system for monitoring sewage treatment data of a water conservancy project, which belongs to the field of data monitoring technology. The method includes: determining first abnormal data of first data, second abnormal data of second data, third abnormal data of third data, and fourth abnormal data of fourth data; determining a first association relationship according to the first abnormal data and the second abnormal data; determining a second association relationship according to the third abnormal data and the fourth abnormal data; determining to obtain first target abnormal data according to the first association relationship, and determining to obtain second target abnormal data according to the second association relationship; performing data elimination according to the first target abnormal data to obtain first target data and second target data, and performing data elimination according to the second target abnormal data to obtain third target data and fourth target data; performing trend prediction on the first target data, the second target data, the third target data, and the fourth target data to obtain target monitoring results.

Description

Technical Field

[0001] The present invention relates to the technical field of data monitoring, and in particular to a method and system for monitoring sewage treatment data in a water conservancy project. Background Art

[0002] In the process of sewage treatment in water conservancy projects, sensors are used to collect real-time sewage data, but the existing technology in this field still has many shortcomings. The accuracy and stability of sensors are key factors that directly affect the monitoring effect. At present, some sensors are difficult to maintain long-term stable performance in complex and changeable sewage environments, and the accuracy of their data is interfered by many factors, such as changes in the chemical composition of sewage, temperature fluctuations, and physical wear of the sensors themselves. Therefore, the existing water conservancy project sewage treatment data monitoring technology still has certain limitations in accuracy. Summary of the invention

[0003] The main purpose of the embodiments of the present invention is to provide a method and system for monitoring sewage treatment data of a water conservancy project, aiming to solve the problem that the existing technology for monitoring sewage treatment data of a water conservancy project still has certain limitations in accuracy.

[0004] In a first aspect, an embodiment of the present invention provides a method for monitoring sewage treatment data of a water conservancy project, comprising:

[0005] Obtaining first data of a first sensor and second data of a second sensor at a first monitoring point, and obtaining third data of a third sensor of the same sensor type as the first sensor and fourth data of a fourth sensor of the same sensor type as the second sensor at the second monitoring point;

[0006] Performing anomaly detection on the first data to obtain first anomaly data and performing anomaly detection on the second data to obtain second anomaly data;

[0007] Performing anomaly detection on the third data to obtain third abnormal data and performing anomaly detection on the fourth data to obtain fourth abnormal data;

[0008] determining a first association relationship between the first sensor and the second sensor according to the first data, the second data, the first abnormal data, and the second abnormal data;

[0009] determining a second association relationship between the third sensor and the fourth sensor according to the third data, the fourth data, the third abnormal data and the fourth abnormal data;

[0010] According to the first association relationship, the first abnormal data and the second abnormal data are abnormally determined to obtain first target abnormal data, and according to the second association relationship, the third abnormal data and the fourth abnormal data are abnormally determined to obtain second target abnormal data;

[0011] Eliminating the first data and the second data according to the first target abnormal data to obtain first target data corresponding to the first data and second target data corresponding to the second data, and eliminating the third data and the fourth data according to the second target abnormal data to obtain third target data corresponding to the third data and fourth target data corresponding to the fourth data;

[0012] Data trend prediction is performed on the first target data, the second target data, the third target data and the fourth target data to obtain target monitoring results corresponding to sewage treatment in the water conservancy project.

[0013] In a second aspect, an embodiment of the present invention provides a water conservancy project sewage treatment data monitoring system, comprising:

[0014] a data acquisition module, configured to obtain first data of a first sensor and second data of a second sensor at a first monitoring point, and obtain third data of a third sensor of the same sensor type as the first sensor and fourth data of a fourth sensor of the same sensor type as the second sensor at the second monitoring point;

[0015] A first processing module, configured to perform anomaly detection on the first data to obtain first anomaly data and perform anomaly detection on the second data to obtain second anomaly data;

[0016] A second processing module, configured to perform anomaly detection on the third data to obtain third abnormal data and perform anomaly detection on the fourth data to obtain fourth abnormal data;

[0017] a first analysis module, configured to determine a first association relationship between the first sensor and the second sensor according to the first data, the second data, the first abnormal data, and the second abnormal data;

[0018] a second analysis module, configured to determine a second association relationship between the third sensor and the fourth sensor according to the third data, the fourth data, the third abnormal data and the fourth abnormal data;

[0019] an abnormality determination module, configured to perform abnormality determination on the first abnormal data and the second abnormal data according to the first association relationship to obtain first target abnormal data, and perform abnormality determination on the third abnormal data and the fourth abnormal data according to the second association relationship to obtain second target abnormal data;

[0020] an abnormality elimination module, configured to eliminate the first data and the second data according to the first target abnormal data to obtain first target data corresponding to the first data and second target data corresponding to the second data, and to eliminate the third data and the fourth data according to the second target abnormal data to obtain third target data corresponding to the third data and fourth target data corresponding to the fourth data;

[0021] The result determination module is used to perform data trend prediction on the first target data, the second target data, the third target data and the fourth target data to obtain target monitoring results corresponding to sewage treatment in water conservancy projects.

[0022] The embodiment of the present invention provides a method and system for monitoring sewage treatment data of a water conservancy project. The method comprises: obtaining first data of a first sensor and second data of a second sensor at a first monitoring point, and obtaining third data of a third sensor and fourth data of a fourth sensor at a second monitoring point; performing anomaly detection on the first data to obtain first abnormal data and performing anomaly detection on the second data to obtain second abnormal data; performing anomaly detection on the third data to obtain third abnormal data and performing anomaly detection on the fourth data to obtain fourth abnormal data; determining a first association relationship between the first sensor and the second sensor according to the first data, the second data, the first abnormal data and the second abnormal data; determining a first association relationship between the third sensor and the fourth sensor according to the third data, the fourth data, the third abnormal data and the fourth abnormal data. The second association relationship between the first abnormal data and the second abnormal data; according to the first association relationship, the first abnormal data and the second abnormal data are abnormally determined to obtain the first target abnormal data, and according to the second association relationship, the third abnormal data and the fourth abnormal data are abnormally determined to obtain the second target abnormal data; according to the first target abnormal data, the first data and the second data are data-removed to obtain the first target data corresponding to the first data and the second target data corresponding to the second data, and according to the second target abnormal data, the third data and the fourth data are data-removed to obtain the third target data corresponding to the third data and the fourth target data corresponding to the fourth data; the first target data, the second target data, the third target data and the fourth target data are data-trend predicted to obtain the target monitoring result corresponding to the sewage treatment of the water conservancy project. The method obtains multiple sets of data from different sensors at the same monitoring point, and by analyzing the association relationship between different sensors, it can better understand the internal connection of the sensor data, thereby performing anomaly detection and data removal according to the association relationship, and then effectively eliminating the abnormal data caused by sensor failure, environmental interference and other factors at the same monitoring point, providing a more solid foundation for subsequently improving the accuracy of the target monitoring results. Finally, by predicting the trend of the target data after removing the anomalies, the target monitoring results in the sewage treatment process can be evaluated more accurately, thereby providing support for optimizing the sewage treatment process, reducing pollution emissions, and improving the overall efficiency of sewage treatment. It also solves the problem that the existing water conservancy project sewage treatment data monitoring technology still has certain limitations in accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] 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.

[0024] Figure 1A schematic diagram of a flow chart of a method for monitoring sewage treatment data in a water conservancy project provided by an embodiment of the present invention;

[0025] Figure 2 A schematic diagram of the module structure of a water conservancy project sewage treatment data monitoring system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0026] 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.

[0027] 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.

[0028] 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.

[0029] The embodiment of the present invention provides a method and system for monitoring water conservancy project sewage treatment data. The method 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.

[0030] 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.

[0031] Please refer to Figure 1 , Figure 1 A schematic flow chart of a method for monitoring sewage treatment data in a water conservancy project provided in an embodiment of the present invention.

[0032] like Figure 1 As shown, the water conservancy project sewage treatment data monitoring method includes steps S101 to S108.

[0033] Step S101, obtaining first data of a first sensor and second data of a second sensor in a first monitoring point, and obtaining third data of a third sensor of the same sensor type as the first sensor and fourth data of a fourth sensor of the same sensor type as the second sensor in a second monitoring point.

[0034] For example, multiple monitoring points are arranged in the sewage treatment system of the water conservancy project, including but not limited to key locations such as water inlet, treatment pool, and water outlet, and various sensors such as water quality sensors, flow sensors, temperature sensors, and pressure sensors are installed at the monitoring points. Thus, the sensors at each monitoring point collect data such as water quality parameters (such as COD, BOD, ammonia nitrogen, total phosphorus, etc.), flow, temperature, and pressure in real time.

[0035] Exemplarily, the first monitoring point is the water inlet, the second monitoring point is the water outlet, the first sensor is a water quality sensor, the second sensor is a flow sensor, the third sensor is a water quality sensor, and the fourth sensor is a flow sensor, so as to obtain the first data of the first sensor and the second data of the second sensor in the first monitoring point, and obtain the third data of the third sensor of the same sensor type as the first sensor in the second monitoring point and the fourth data of the fourth sensor of the same sensor type as the second sensor.

[0036] Step S102: performing anomaly detection on the first data to obtain first abnormal data and performing anomaly detection on the second data to obtain second abnormal data.

[0037] Exemplarily, the first data is subjected to anomaly detection using a machine learning algorithm such as an isolation forest, a local anomaly factor, or a deep learning method such as an autoencoder to obtain the first anomaly data. The second data is subjected to anomaly detection using a machine learning algorithm such as an isolation forest, a local anomaly factor, or a deep learning method such as an autoencoder to obtain the second anomaly data.

[0038] In some embodiments, the performing anomaly detection on the first data to obtain the first anomaly data and the performing anomaly detection on the second data to obtain the second anomaly data include: determining the first fitness corresponding to each data in the first data, and determining the first crossover individual and the second crossover individual from the first data according to the first fitness; obtaining the first target value corresponding to the first crossover individual and the second target value corresponding to the second crossover individual from the first fitness; calculating the average value corresponding to all the first fitnesses to obtain the first average fitness, and calculating the square of the first average fitness to obtain the first square value; calculating the square of the difference between the first target value and the second target value to obtain the second square value; determining the first crossover rate and the first mutation rate according to the first square value and the second square value; searching the first data according to the first crossover rate and the first mutation rate to obtain the first cluster center, and calculating the first distance between the first data and the first cluster center; Perform data clustering to obtain a first clustering result; determine the first abnormal data corresponding to the first data according to the first clustering result; determine the second fitness corresponding to each data in the second data, and determine the third crossover individual and the fourth crossover individual according to the second fitness; obtain the third target value corresponding to the third crossover individual and the fourth target value corresponding to the fourth crossover individual from the second fitness; calculate the average value corresponding to all the second fitnesses to obtain the second average fitness, and calculate the square of the second average fitness to obtain the third square value; calculate the square of the difference between the third target value and the fourth target value to obtain the fourth square value; determine the second crossover rate and the second mutation rate according to the third square value and the fourth square value; search the second data according to the second crossover rate and the second mutation rate to obtain the second cluster center, and calculate the second data and the second cluster center to perform data clustering on the second data to obtain the second clustering result; determine the second abnormal data corresponding to the second data according to the second clustering result.

[0039] Exemplarily, a first expected value of the first monitoring point under the first sensor is determined, and based on a first mean square error between each data in the first data and the first expected value, the inverse of the first mean square error is determined as the corresponding first fitness.

[0040] Exemplarily, the data corresponding to the first maximum value in the first fitness is determined as the first crossover individual and the data corresponding to the second maximum value in the first fitness is determined as the second crossover individual, and the first maximum value is determined as the first target value corresponding to the first crossover individual and the second maximum value is determined as the second target value corresponding to the second crossover individual.

[0041] Exemplarily, the average value of all first fitnesses is calculated to obtain the first average fitness, and the square of the first average fitness is calculated to obtain the first square value. The difference between the first target value and the second target value is calculated, and the difference is squared to obtain the second square value. If the second square is equal to 0, the first crossover rate is determined as a fixed value; if the second square is not equal to 0, the first square value and the second square value are divided to obtain the first result, and the first exponential result corresponding to the first result under the exponent of the natural constant is obtained, so as to obtain the inverse of the first exponential result, add 1, and then multiply by the fixed value to obtain the first crossover rate. If the second square value is equal to 0, the first mutation rate is determined as a fixed adjustment coefficient; if the second square value is not equal to 0, the first square value and the second square value are divided to obtain the first result, and the first exponential result corresponding to the first result under the exponent of the natural constant is obtained, so as to obtain the inverse of the first exponential result, add 1, and then multiply by the fixed adjustment coefficient to obtain the first mutation rate.

[0042] Exemplarily, the first data is searched for an optimal solution based on the first crossover rate and the first mutation rate, thereby obtaining the first cluster center, and then the first distance between each data in the first data and each cluster center in the first cluster center is calculated according to common distance measurement methods such as Euclidean distance and Manhattan distance, so as to classify the first data into the first cluster center corresponding to the smaller distance according to the minimum value in the first distance, thereby obtaining the first clustering result, and finally determining the cluster cluster corresponding to the smaller number of data in each cluster cluster in the first clustering result as the first abnormal data.

[0043] Exemplarily, a second expected value of the first monitoring point under the second sensor is determined, and based on a second mean square error between each data in the second data and the second expected value, the inverse of the second mean square error is determined as the corresponding second fitness.

[0044] Exemplarily, the data corresponding to the first maximum value in the second fitness is determined as the third crossover individual and the data corresponding to the second maximum value in the second fitness is determined as the fourth crossover individual, and the first maximum value is determined as the third target value corresponding to the third crossover individual and the second maximum value is determined as the fourth target value corresponding to the fourth crossover individual.

[0045] Exemplarily, the average value of all second fitnesses is calculated to obtain the second average fitness, and the square of the second average fitness is calculated to obtain the third square value. The third target value and the fourth target value are calculated by difference, and the difference is squared to obtain the fourth square value. If the fourth square value is equal to 0, the second crossover rate is determined as a fixed value; if the fourth square value is not equal to 0, the third square value and the fourth square value are divided to obtain the second result, and the second exponential result corresponding to the second result under the exponent of the natural constant is obtained, so as to obtain the reciprocal of the second exponential result, add 1, and then multiply by the fixed value to obtain the second crossover rate. If the fourth square value is equal to 0, the second mutation rate is determined as a fixed adjustment coefficient; if the fourth square value is not equal to 0, the third square value and the fourth square value are divided to obtain the first result, and the first exponential result corresponding to the first result under the exponent of the natural constant is obtained, so as to obtain the reciprocal of the first exponential result, add 1, and then multiply by the fixed adjustment coefficient to obtain the second mutation rate.

[0046] Exemplarily, the second data is searched for an optimal solution based on the second crossover rate and the second mutation rate, thereby obtaining the second cluster center, and then the second distance between each data in the second data and each cluster center in the second cluster center is calculated according to common distance measurement methods such as Euclidean distance and Manhattan distance, so that the second data is classified into the second cluster center corresponding to the smaller distance according to the minimum value of the second distance, thereby obtaining the second clustering result, and finally the cluster cluster corresponding to the smaller data number in each cluster cluster in the second clustering result is determined as the second abnormal data.

[0047] Step S103: performing anomaly detection on the third data to obtain third abnormal data and performing anomaly detection on the fourth data to obtain fourth abnormal data.

[0048] Exemplarily, the third data is subjected to anomaly detection using a machine learning algorithm such as an isolation forest, a local anomaly factor, or a deep learning method such as an autoencoder to obtain the third anomaly data. The fourth data is subjected to anomaly detection using a machine learning algorithm such as an isolation forest, a local anomaly factor, or a deep learning method such as an autoencoder to obtain the fourth anomaly data.

[0049] In some embodiments, the performing anomaly detection on the third data to obtain the third anomaly data and the performing anomaly detection on the fourth data to obtain the fourth anomaly data include: determining the third fitness corresponding to each data in the third data, and determining the fifth crossover individual and the sixth crossover individual according to the third fitness; obtaining the fifth target value corresponding to the fifth crossover individual and the sixth target value corresponding to the sixth crossover individual from the third fitness; calculating the average value corresponding to all the third fitnesses to obtain the third average fitness, and calculating the square of the third average fitness to obtain the fifth square value; calculating the square of the difference corresponding to the fifth target value and the sixth target value to obtain the sixth square value; determining the third crossover rate and the third mutation rate according to the fifth square value and the sixth square value; searching the third data according to the third crossover rate and the third mutation rate to obtain the third cluster center, and calculating the third distance between the third data and the third cluster center; performing data clustering on the third data according to the third distance and the third cluster center to obtain the third clustering result; Determine the third abnormal data corresponding to the third data according to the third clustering result; determine the fourth fitness corresponding to each data in the fourth data, and determine the seventh crossover individual and the eighth crossover individual according to the fourth fitness; obtain the seventh target value corresponding to the seventh crossover individual and the eighth target value corresponding to the eighth crossover individual from the fourth fitness; calculate the average value corresponding to the fourth fitness to obtain the fourth average fitness, and calculate the square of the fourth average fitness to obtain the seventh square value; calculate the square of the difference corresponding to the seventh target value and the eighth target value to obtain the eighth square value; determine the fourth crossover rate and the fourth mutation rate according to the seventh square value and the eighth square value; search the fourth data according to the fourth crossover rate and the fourth mutation rate to obtain the fourth cluster center, and calculate the fourth distance between the fourth data and the fourth cluster center; cluster the fourth data according to the fourth distance and the fourth cluster center to obtain the fourth clustering result; determine the fourth abnormal data corresponding to the fourth data according to the fourth clustering result.

[0050] Exemplarily, a third expected value of the second monitoring point under the third sensor is determined, and based on a third mean square error between each data in the third data and the third expected value, the inverse of the third mean square error is determined as the corresponding third fitness.

[0051] Exemplarily, the data corresponding to the first maximum value in the third fitness is determined as the fifth crossover individual and the data corresponding to the second maximum value in the third fitness is determined as the sixth crossover individual, and the first maximum value is determined as the fifth target value corresponding to the fifth crossover individual and the second maximum value is determined as the sixth target value corresponding to the sixth crossover individual.

[0052] Exemplarily, the average value of all third fitnesses is calculated to obtain the third average fitness, and the square of the third average fitness is calculated to obtain the fifth square value. The fifth target value and the sixth target value are calculated by difference, and the difference is squared to obtain the sixth square value. If the sixth square value is equal to 0, the third crossover rate is determined as a fixed value; if the sixth square value is not equal to 0, the fifth square value and the sixth square value are divided to obtain the third result, and the third exponential result corresponding to the third result under the exponent of the natural constant is obtained, so as to obtain the inverse of the third exponential result, add 1, and then multiply by the fixed value to obtain the third crossover rate. If the sixth square value is equal to 0, the third mutation rate is determined as a fixed adjustment coefficient; if the sixth square value is not equal to 0, the fifth square value and the sixth square value are divided to obtain the third result, and the third exponential result corresponding to the third result under the exponent of the natural constant is obtained, so as to obtain the inverse of the third exponential result, add 1, and then multiply by the fixed adjustment coefficient to obtain the third mutation rate.

[0053] Exemplarily, the third data is searched for an optimal solution based on the third crossover rate and the third mutation rate, thereby obtaining the third cluster center, and then the third distance between each data in the third data and each cluster center in the third cluster center is calculated according to common distance measurement methods such as Euclidean distance and Manhattan distance, so that the third data is classified into the third cluster center corresponding to the smaller distance according to the minimum value of the third distance, thereby obtaining the third clustering result, and finally the cluster cluster corresponding to the smaller data number in each cluster cluster in the third clustering result is determined as the third abnormal data.

[0054] Exemplarily, a fourth expected value of the second monitoring point under the fourth sensor is determined, and based on a fourth mean square error between each data in the fourth data and the fourth expected value, the inverse of the fourth mean square error is determined as the corresponding fourth fitness.

[0055] Exemplarily, the data corresponding to the first maximum value in the fourth fitness is determined as the seventh crossover individual and the data corresponding to the second maximum value in the fourth fitness is determined as the eighth crossover individual, and the first maximum value is determined as the seventh target value corresponding to the seventh crossover individual and the second maximum value is determined as the eighth target value corresponding to the eighth crossover individual.

[0056] Exemplarily, the average value of all fourth fitnesses is calculated to obtain the fourth average fitness, and the square of the fourth average fitness is calculated to obtain the seventh square value. The seventh target value and the eighth target value are calculated by difference, and the difference is squared to obtain the eighth square value. If the eighth square value is equal to 0, the fourth crossover rate is determined as a fixed value; if the eighth square value is not equal to 0, the seventh square value and the eighth square value are divided to obtain the fourth result, and the fourth exponential result corresponding to the fourth result under the exponent of the natural constant is obtained, so as to obtain the inverse of the fourth exponential result, add 1, and then multiply by the fixed value to obtain the fourth crossover rate. If the eighth square value is equal to 0, the eighth mutation rate is determined as a fixed adjustment coefficient; if the eighth square value is not equal to 0, the seventh square value and the eighth square value are divided to obtain the fourth result, and the fourth exponential result corresponding to the fourth result under the exponent of the natural constant is obtained, so as to obtain the inverse of the fourth exponential result, add 1, and then multiply by the fixed adjustment coefficient to obtain the fourth mutation rate.

[0057] Exemplarily, the second data is searched for the optimal solution according to the fourth crossover rate and the fourth mutation rate, so as to obtain the fourth cluster center, and then the fourth distance between each data in the fourth data and each cluster center in the fourth cluster center is calculated according to common distance measurement methods such as Euclidean distance and Manhattan distance, so as to classify the fourth data into the fourth cluster center corresponding to the smaller distance according to the minimum value of the fourth distance, so as to obtain the fourth clustering result, and finally the cluster cluster corresponding to the smaller data number in each cluster cluster in the fourth clustering result is determined as the fourth abnormal data.

[0058] Step S104: determining a first association relationship between the first sensor and the second sensor according to the first data, the second data, the first abnormal data, and the second abnormal data.

[0059] Exemplarily, the first data and the second data are time-aligned to ensure that the first data and the second data at the same time point can correspond to each other, and then the first data is subjected to statistical features such as mean, variance, maximum value, minimum value, and time series features such as autocorrelation coefficient and cross-correlation coefficient, and the second data is subjected to statistical features such as mean, variance, maximum value, minimum value, and time series features such as autocorrelation coefficient and cross-correlation coefficient, so as to determine the first relationship between the first sensor and the second sensor under the first data and the second data based on the statistical features and time series features using the Pearson correlation coefficient or the Spearman correlation coefficient.

[0060] Exemplarily, the first abnormal data is removed from the first data to obtain the first normal data, and the second abnormal data is removed from the second data to obtain the second normal data, so that the first normal data and the second normal data are time-aligned to ensure that the first normal data and the second normal data at the same time point can correspond to each other, and then the first normal data is subjected to statistical features such as mean, variance, maximum value, minimum value, and time series features such as autocorrelation coefficient and cross-correlation coefficient, and the second normal data is subjected to statistical features such as mean, variance, maximum value, minimum value, and time series features such as autocorrelation coefficient and cross-correlation coefficient, so as to determine the second relationship between the first sensor and the second sensor under the first data and the second data based on the statistical features and time series features using the Pearson correlation coefficient or the Spearman correlation coefficient.

[0061] Exemplarily, the consistency of the first relationship and the second relationship is compared. If the first relationship and the second relationship are the same, it indicates that after the first abnormal data and the second abnormal data are eliminated, the association pattern between the first sensor and the second sensor has not changed significantly. This further illustrates that the first abnormal data and the second abnormal data may not be true abnormal data, but normal fluctuations or noise in the data. At this time, the first association relationship can be simply taken from either the first relationship or the second relationship. If the first relationship and the second relationship are not the same, it indicates that the first abnormal data and the second abnormal data have a significant impact on the association pattern between the first sensor and the second sensor. This means that the first abnormal data and the second abnormal data are determined as abnormal information in the first data and the second data. The first association relationship should be based on the first relationship, that is, after considering the abnormal data, the relationship between the first sensor and the second sensor has changed. This new relationship reflects the change pattern between the first sensor and the second sensor after the presence of abnormal data.

[0062] In some embodiments, determining the first association relationship between the first sensor and the second sensor based on the first data, the second data, the first abnormal data, and the second abnormal data includes: aligning the first data and the second data one by one according to the acquisition time to obtain first aligned data corresponding to the first data and second aligned data corresponding to the second data at the same time; obtaining any first analysis data in the first aligned data and second analysis data corresponding to the first analysis data in the second aligned data; obtaining first association data corresponding to the first analysis data under a first difference from the first aligned data; obtaining first corresponding data corresponding to the first association data from the second aligned data, and calculating whether the first corresponding data and the second analysis data satisfy a second difference to obtain second association data; obtaining a first quantity corresponding to the first association data and a second quantity corresponding to the second association data, and determining a first association value between the first analysis data and the second analysis data according to the first quantity and the second quantity; determining a first average association value between the first data and the second data according to the first association value; determining a first abnormal position according to the first abnormal data and the second abnormal data, and calculating the first abnormal position according to the first abnormal position. The invention provides a method for obtaining a first processed data by eliminating data from the first processed data, and obtaining a second processed data by eliminating data from the second data according to the first abnormal position; aligning the first processed data and the second processed data one by one according to the acquisition time, and obtaining third aligned data corresponding to the first processed data and fourth aligned data corresponding to the second processed data at the same time; obtaining any third analysis data in the third aligned data and fourth analysis data corresponding to the third analysis data in the fourth aligned data; obtaining third associated data corresponding to the third analysis data under the first difference value satisfied from the third aligned data; obtaining second corresponding data corresponding to the third associated data from the fourth aligned data, and calculating whether the second corresponding data and the fourth analysis data satisfy the second difference value to obtain fourth associated data; obtaining a third quantity corresponding to the third associated data and a fourth quantity corresponding to the fourth associated data, and determining a second associated value between the third analysis data and the fourth analysis data according to the third quantity and the fourth quantity; determining a second average associated value between the first processed data and the second processed data according to the second associated value; and determining the first associated relationship between the first sensor and the second sensor according to the first average associated value and the second average associated value.

[0063] Exemplarily, a timestamp or other time identifier is used to match the first data and the second data at the same time point to obtain first alignment data corresponding to the first data and second alignment data corresponding to the second data at the same time point. The first analysis data at any time point is selected from the first alignment data, and the second analysis data corresponding to the time point is found in the second alignment data.

[0064] Exemplarily, the first associated data corresponding to the time when the absolute value of the difference between any data and the first analysis data is less than the first difference is extracted from the first aligned data. The first corresponding data corresponding to the first associated data is obtained from the second aligned data according to the timestamp corresponding to the first associated data, and then it is calculated whether the absolute value of the difference between the first corresponding data and the second analysis data is less than the second difference. If the absolute value of the difference between the first corresponding data and the second analysis data is less than the second difference, the first corresponding data is determined as the second associated data; if the absolute value of the difference between the first corresponding data and the second analysis data is greater than or equal to the second difference, the first corresponding data is not determined as the second associated data.

[0065] Exemplarily, a first number corresponding to the first associated data and a second number corresponding to the second associated data are counted, then a target number of the first aligned data is counted, and the first number is divided by the target number to obtain a first probability, and the reciprocal of the first probability is calculated to obtain a first reciprocal, and the logarithm of the first reciprocal with the number of the first aligned data as the base is taken to obtain a first logarithm value, and then the second number is divided by the target number to obtain a second probability, and then the first logarithm value and the second probability are multiplied to obtain a first association value between the first analysis data and the second analysis data.

[0066] Exemplarily, after obtaining the first correlation value corresponding to each first analysis data, summing and averaging are performed to obtain a first average correlation value.

[0067] Exemplarily, a first timestamp corresponding to the first data is obtained based on the first abnormal data, and a second timestamp corresponding to the second data is obtained based on the second abnormal data, and then the first timestamp and the second timestamp are merged to obtain a first abnormal position, and then the data at the corresponding position in the first data is eliminated based on the first abnormal position to obtain first processed data; and the data at the corresponding position in the second data is eliminated based on the first abnormal position to obtain second processed data.

[0068] Exemplarily, the first processed data and the second processed data are matched at the same time point to obtain the third aligned data corresponding to the first processed data and the fourth aligned data corresponding to the second processed data at the same time point. After the data are aligned, data matching is performed to ensure that the processed data at each time point can correspond to each other.

[0069] Exemplarily, the third analysis data of any time point is selected from the third alignment data, and the fourth analysis data corresponding to the time point is found in the fourth alignment data.

[0070] Exemplarily, the third associated data corresponding to the case where the absolute value of the difference between any data and the third analysis data is less than the first difference is extracted from the third aligned data. The second corresponding data corresponding to the third associated data is obtained from the fourth aligned data according to the timestamp corresponding to the third associated data, and then it is calculated whether the absolute value of the difference between the second corresponding data and the fourth analysis data is less than the second difference. If the absolute value of the difference between the second corresponding data and the fourth analysis data is less than the second difference, the second corresponding data is determined as the fourth associated data; if the absolute value of the difference between the second corresponding data and the fourth analysis data is greater than or equal to the second difference, the second corresponding data is not determined as the fourth associated data.

[0071] Exemplarily, the third quantity corresponding to the third associated data and the fourth quantity corresponding to the fourth associated data are counted, then the target quantity of the third aligned data is counted, and the third quantity is divided by the target quantity to obtain a third probability, and the reciprocal of the third probability is calculated to obtain a second reciprocal, and the logarithm of the second reciprocal with the target quantity as the base is taken to obtain a second logarithm value, and then the fourth quantity is divided by the target quantity to obtain a fourth probability, and then the second logarithm value and the fourth probability are multiplied to obtain a second association value between the third analysis data and the fourth analysis data.

[0072] Exemplarily, after obtaining the second correlation value corresponding to each third analysis data, sum and average them to obtain the second average correlation value.

[0073] Exemplarily, the absolute value of the difference between the first average correlation value and the second average correlation value is calculated to obtain the first target value, and the first target value is compared with the preset value. When the first target value is less than or equal to the preset value, it indicates that after the first abnormal data and the second abnormal data are eliminated, the correlation pattern between the first sensor and the second sensor has not changed significantly. This further illustrates that the first abnormal data and the second abnormal data may not be true abnormal data, but normal fluctuations or noise in the data. At this time, the first correlation relationship can be determined as the first relationship corresponding to the first average correlation value or the second relationship corresponding to the second average correlation value. If the first target value is greater than the preset value, it indicates that the first abnormal data and the second abnormal data have a significant impact on the correlation pattern between the first sensor and the second sensor. This means that the first abnormal data and the second abnormal data are determined as abnormal information in the first data and the second data. The first correlation relationship should be based on the first relationship determined by the first average correlation value, that is, after considering the abnormal data, the relationship between the first sensor and the second sensor has changed. This new relationship reflects the change pattern between the first sensor and the second sensor after the presence of abnormal data.

[0074] Step S105: determining a second association relationship between the third sensor and the fourth sensor according to the third data, the fourth data, the third abnormal data and the fourth abnormal data.

[0075] Exemplarily, the third data and the fourth data are time-aligned to ensure that the third data and the fourth data at the same time point can correspond to each other, and then the statistical features such as mean, variance, maximum value, minimum value, etc. of the third data, and the time series features such as autocorrelation coefficient and cross-correlation coefficient are calculated, and the statistical features such as mean, variance, maximum value, minimum value, etc. of the fourth data, and the time series features such as autocorrelation coefficient and cross-correlation coefficient are calculated, so as to determine the third relationship between the third sensor and the fourth sensor under the third data and the fourth data according to the statistical features and the time series features using the Pearson correlation coefficient or the Spearman correlation coefficient.

[0076] Exemplarily, the third abnormal data is removed from the third data to obtain the third normal data, and the fourth abnormal data is removed from the fourth data to obtain the fourth normal data, so that the third normal data and the fourth normal data are time-aligned to ensure that the third normal data and the fourth normal data at the same time point can correspond to each other, and then the third normal data is subjected to statistical features such as mean, variance, maximum value, minimum value, and time series features such as autocorrelation coefficient and cross-correlation coefficient, and the fourth normal data is subjected to statistical features such as mean, variance, maximum value, minimum value, and time series features such as autocorrelation coefficient and cross-correlation coefficient, so that the fourth relationship between the third sensor and the fourth sensor under the third data and the fourth data is determined according to the statistical features and time series features using the Pearson correlation coefficient or the Spearman correlation coefficient.

[0077] Exemplarily, the consistency of the third relationship and the fourth relationship is compared. If the third relationship and the fourth relationship are the same, it indicates that after the third abnormal data and the fourth abnormal data are eliminated, the association pattern between the third sensor and the fourth sensor has not changed significantly. This further illustrates that the third abnormal data and the fourth abnormal data may not be true abnormal data, but normal fluctuations or noise in the data. At this time, the third association relationship can be simply taken from either the third relationship or the fourth relationship. If the third relationship and the fourth relationship are not the same, it indicates that the third abnormal data and the fourth abnormal data have a significant impact on the association pattern between the third sensor and the fourth sensor. This means that the third abnormal data and the fourth abnormal data are determined as abnormal information in the third data and the fourth data. The third association relationship should be based on the third relationship, that is, after considering the abnormal data, the relationship between the third sensor and the fourth sensor has changed. This new relationship reflects the change pattern between the third sensor and the fourth sensor after the presence of abnormal data.

[0078] In some embodiments, determining the second association relationship between the third sensor and the fourth sensor according to the third data, the fourth data, the third abnormal data, and the fourth abnormal data includes: aligning the third data and the fourth data one by one according to the acquisition time, and obtaining fifth alignment data corresponding to the third data and sixth alignment data corresponding to the fourth data at the same time; obtaining any fifth analysis data in the fifth alignment data and sixth analysis data corresponding to the fifth analysis data in the sixth alignment data; obtaining fifth association data corresponding to the fifth analysis data under the third difference from the fifth alignment data; obtaining third corresponding data corresponding to the fifth association data from the sixth alignment data, and calculating whether the third corresponding data and the sixth analysis data satisfy the fourth difference to obtain sixth association data; obtaining a fifth quantity corresponding to the fifth association data and a sixth quantity corresponding to the sixth association data, and determining a third association value between the fifth analysis data and the sixth analysis data according to the fifth quantity and the sixth quantity; determining a third average association value between the third data and the fourth data according to the third association value; determining a second abnormal position according to the third abnormal data and the fourth abnormal data, and calculating the third abnormal position according to the second abnormal position. The method comprises the steps of: performing data elimination on the third processed data and the fourth processed data according to the second abnormal position to obtain the third processed data; and performing data elimination on the fourth data according to the second abnormal position to obtain the fourth processed data; aligning the third processed data and the fourth processed data one by one according to the acquisition time to obtain the seventh aligned data corresponding to the third processed data and the eighth aligned data corresponding to the fourth processed data at the same time; obtaining any seventh analysis data in the seventh aligned data and the eighth analysis data corresponding to the seventh analysis data in the eighth aligned data; obtaining the seventh associated data corresponding to the seventh analysis data satisfying the third difference from the seventh aligned data; obtaining the fourth corresponding data corresponding to the seventh associated data from the eighth aligned data, and calculating whether the fourth corresponding data and the eighth analysis data satisfy the fourth difference to obtain the eighth associated data; obtaining the seventh quantity corresponding to the seventh associated data and the eighth quantity corresponding to the eighth associated data, and determining the fourth associated value between the seventh analysis data and the eighth analysis data according to the seventh quantity and the eighth quantity; determining the fourth average associated value between the third processed data and the fourth processed data according to the fourth associated value; and determining the second associated relationship between the third sensor and the fourth sensor according to the third average associated value and the fourth average associated value.

[0079] Exemplarily, the third data and the fourth data are matched at the same time point using a timestamp or other time identifier to obtain fifth alignment data corresponding to the third data and sixth alignment data corresponding to the fourth data at the same time. The fifth analysis data at any time point is selected from the fifth alignment data, and the sixth analysis data corresponding to the time point is found in the sixth alignment data.

[0080] Exemplarily, the fifth associated data corresponding to the case where the absolute value of the difference between any data and the fifth analysis data is less than the third difference is extracted from the fifth alignment data. The third corresponding data corresponding to the fifth association data is obtained from the sixth alignment data according to the timestamp corresponding to the fifth association data, and then it is calculated whether the absolute value of the difference between the third corresponding data and the sixth analysis data is less than the third difference. If the absolute value of the difference between the third corresponding data and the sixth analysis data is less than the third difference, the third corresponding data is determined as the sixth associated data; if the absolute value of the difference between the third corresponding data and the sixth analysis data is greater than or equal to the second difference, the third corresponding data is not determined as the sixth associated data.

[0081] Exemplarily, the fifth quantity corresponding to the fifth associated data and the sixth quantity corresponding to the sixth associated data are counted, then the target quantity of the fifth aligned data is counted, and the fifth quantity is divided by the target quantity to obtain the fifth probability, and the reciprocal of the fifth probability is calculated to obtain the third reciprocal, and the logarithm of the third reciprocal with the quantity of the fifth aligned data as the base is taken to obtain the third logarithm value, and then the sixth quantity is divided by the target quantity to obtain the sixth probability, and then the third logarithm value and the sixth probability are multiplied to obtain the third association value between the fifth analysis data and the sixth analysis data.

[0082] Exemplarily, after obtaining the third correlation value corresponding to each fifth analysis data, sum and average are performed to obtain a third average correlation value.

[0083] Exemplarily, a corresponding third timestamp in the third data is obtained based on the third abnormal data, and a corresponding fourth timestamp in the fourth data is obtained based on the fourth abnormal data, and then the third timestamp and the fourth timestamp are merged to obtain a second abnormal position, and then the data at the corresponding position in the third data is eliminated based on the second abnormal position to obtain third processed data; and the data at the corresponding position in the fourth data is eliminated based on the second abnormal position to obtain fourth processed data.

[0084] Exemplarily, the third processed data and the fourth processed data are matched at the same time point to obtain the seventh alignment data corresponding to the third processed data and the eighth alignment data corresponding to the fourth processed data at the same time. After data alignment, data matching is performed to ensure that the processed data at each time point can correspond to each other.

[0085] Exemplarily, the seventh analysis data of any time point is selected from the seventh alignment data, and the eighth analysis data corresponding to the time point is found in the eighth alignment data.

[0086] Exemplarily, the seventh associated data corresponding to the time when the absolute value of the difference between any data and the seventh analysis data is less than the third difference is extracted from the seventh alignment data. And the fourth corresponding data corresponding to the seventh associated data is obtained from the eighth alignment data according to the timestamp corresponding to the seventh associated data, and then it is calculated whether the absolute value of the difference between the fourth corresponding data and the eighth analysis data is less than the fourth difference. If the absolute value of the difference between the fourth corresponding data and the eighth analysis data is less than the fourth difference, the fourth corresponding data is determined as the eighth associated data; if the absolute value of the difference between the fourth corresponding data and the eighth analysis data is greater than or equal to the fourth difference, the fourth corresponding data is not determined as the eighth associated data.

[0087] Exemplarily, the seventh quantity corresponding to the seventh associated data and the eighth quantity corresponding to the eighth associated data are counted, then the target quantity of the seventh aligned data is counted, and the seventh quantity is divided by the target quantity to obtain the seventh probability, and the reciprocal of the seventh probability is calculated to obtain the fourth reciprocal, and the logarithm of the fourth reciprocal with the target quantity as the base is taken to obtain the fourth logarithm value, and then the eighth quantity is divided by the target quantity to obtain the eighth probability, and then the fourth logarithm value and the eighth probability are multiplied to obtain the fourth association value between the seventh analysis data and the eighth analysis data.

[0088] Exemplarily, after obtaining the fourth correlation value corresponding to each seventh analysis data, sum and average them to obtain the fourth average correlation value.

[0089] Exemplarily, the absolute value of the difference between the third average correlation value and the fourth average correlation value is calculated to obtain the second target value, and the second target value is compared with the preset value. When the second target value is less than or equal to the preset value, it indicates that after the third abnormal data and the fourth abnormal data are eliminated, the correlation pattern between the third sensor and the fourth sensor has not changed significantly. This further illustrates that the third abnormal data and the fourth abnormal data may not be true abnormal data, but normal fluctuations or noise in the data. At this time, the second correlation relationship can be determined as the third relationship corresponding to the third average correlation value or the fourth relationship corresponding to the fourth average correlation value. If the second target value is greater than the preset value, it indicates that the third abnormal data and the fourth abnormal data have a significant impact on the correlation pattern between the third sensor and the fourth sensor. This means that the third abnormal data and the fourth abnormal data are determined as abnormal information in the third data and the fourth data. The second correlation relationship should be based on the third relationship determined by the third average correlation value, that is, after considering the abnormal data, the relationship between the third sensor and the fourth sensor has changed. This new relationship reflects the change pattern between the third sensor and the fourth sensor after the presence of abnormal data.

[0090] Step S106: performing abnormality determination on the first abnormal data and the second abnormal data according to the first association relationship to obtain first target abnormal data, and performing abnormality determination on the third abnormal data and the fourth abnormal data according to the second association relationship to obtain second target abnormal data.

[0091] Exemplarily, when the first association relationship is the same as the first preset relationship between the first sensor and the second sensor at the first monitoring point, the first target abnormal data is empty; when the first association relationship is different from the first preset relationship between the first sensor and the second sensor at the first monitoring point, a first timestamp corresponding to the first abnormal data and a second timestamp corresponding to the second abnormal data are obtained, and the first timestamp and the second timestamp are merged to determine the first target abnormal data.

[0092] Exemplarily, when the second association relationship is the same as the second preset relationship between the third sensor and the fourth sensor at the second monitoring point, the second target abnormal data is empty; when the second association relationship is different from the second preset relationship between the third sensor and the fourth sensor at the second monitoring point, the third timestamp corresponding to the third abnormal data and the fourth timestamp corresponding to the fourth abnormal data are obtained, and the third timestamp and the fourth timestamp are merged to determine as the first target abnormal data.

[0093] Step S107, performing data elimination on the first data and the second data according to the first target abnormal data to obtain first target data corresponding to the first data and second target data corresponding to the second data, and performing data elimination on the third data and the fourth data according to the second target abnormal data to obtain third target data corresponding to the third data and fourth target data corresponding to the fourth data.

[0094] Exemplarily, the data corresponding to the timestamp in the first data is eliminated when it is the same as the first target abnormal data to obtain the first target data corresponding to the first data, and the data corresponding to the timestamp in the second data is eliminated when it is the same as the first target abnormal data to obtain the second target data corresponding to the second data.

[0095] Exemplarily, the data corresponding to the timestamp in the third data and the second target abnormal data are eliminated to obtain the third target data corresponding to the third data, and the data corresponding to the timestamp in the fourth data and the second target abnormal data are eliminated to obtain the fourth target data corresponding to the fourth data.

[0096] Step S108: perform data trend prediction on the first target data, the second target data, the third target data and the fourth target data to obtain target monitoring results corresponding to sewage treatment in the water conservancy project.

[0097] Exemplarily, the first target data and the second target data are integrated to determine the first pollution level corresponding to the water conservancy project at the first monitoring point, and the third target data and the fourth target data are integrated to determine the second pollution level corresponding to the water conservancy project at the second monitoring point, and then the sewage treatment level between the first monitoring point and the second monitoring point is determined according to the first pollution level and the second pollution level, and the sewage treatment level is compared with the preset level. When the sewage treatment level is less than the preset level, it is determined that the target monitoring result corresponding to the sewage treatment of the water conservancy project is abnormal sewage treatment; when the sewage treatment level is greater than or equal to the preset level, it is determined that the target monitoring result corresponding to the sewage treatment of the water conservancy project is normal sewage treatment.

[0098] In some embodiments, the data trend prediction of the first target data, the second target data, the third target data and the fourth target data to obtain the target monitoring result corresponding to the sewage treatment of the water conservancy project includes: determining a first weight corresponding to the first target data and a second weight corresponding to the second target data; fusing the first target data and the second target data according to the first weight and the second weight to obtain the first abnormal information corresponding to the first monitoring point; determining the third weight corresponding to the third target data and the fourth weight corresponding to the fourth target data; fusing the third target data and the fourth target data according to the third weight and the fourth weight to obtain the second abnormal information corresponding to the second monitoring point; determining a trend characterization value corresponding to the sewage treatment trend of the water conservancy project according to the first abnormal information and the second abnormal information; and obtaining the target monitoring result corresponding to the sewage treatment of the water conservancy project according to the trend characterization value.

[0099] Exemplarily, a first weight corresponding to the first target data and a second weight corresponding to the second target data are determined based on expert experience, and a third weight corresponding to the third target data and a fourth weight corresponding to the fourth target data are determined.

[0100] Exemplarily, the first target data and the second target data are fused according to the first weight and the second weight to obtain a first fusion result, and then the first fusion result is analyzed using the abnormal type recognition model to determine the corresponding abnormal type, and then the first abnormal type is summarized to obtain the first abnormal information corresponding to the first monitoring point. The first abnormal information is used to characterize the distribution information of the water pollution type at the first monitoring point. The first abnormal information includes the water pollution type that may exist at the first monitoring point and the corresponding occurrence frequency of the water pollution type.

[0101] Exemplarily, the third target data and the fourth target data are fused according to the third weight and the fourth weight to obtain a second fusion result, and then the second fusion result is analyzed using the anomaly type recognition model to determine the corresponding anomaly type, and then the second anomaly type is summarized to obtain the second anomaly information corresponding to the second monitoring point. The second anomaly information is used to characterize the distribution information of the water pollution type at the second monitoring point. The second anomaly information includes the water pollution type that may exist at the second monitoring point and the corresponding occurrence frequency of the water pollution type.

[0102] Exemplarily, the KL divergence or cross entropy is used to calculate the distribution difference between the first abnormal information and the second abnormal information to obtain the trend characterization value corresponding to the sewage treatment trend of the water conservancy project at the first monitoring point and the second monitoring point. The greater the distribution difference between the first abnormal information and the second abnormal information, the greater the trend characterization value, which means that the water pollution treatment effect is more obvious, and the target monitoring result is normal sewage treatment; the smaller the distribution difference between the first abnormal information and the second abnormal information, the smaller the trend characterization value, which means that the water pollution treatment effect is less obvious, and the target monitoring result is abnormal sewage treatment.

[0103] In some embodiments, determining the first weight corresponding to the first target data and the second weight corresponding to the second target data includes: obtaining first historical data corresponding to the first sensor and second historical data corresponding to the second sensor; establishing a first prediction model corresponding to the first sensor based on the first historical data and establishing a second prediction model corresponding to the second sensor based on the second historical data; determining the first prediction data corresponding to the first target data based on the first prediction model and obtaining the second prediction data corresponding to the second target data based on the second prediction model; obtaining first error information corresponding to the first sensor based on the first target data and the first prediction data; obtaining second error information corresponding to the second sensor based on the second target data and the second prediction data; and determining the first weight corresponding to the first target data and the second weight corresponding to the second target data based on the first error information and the second error information.

[0104] Exemplarily, first historical data corresponding to the first sensor and second historical data corresponding to the second sensor are obtained from the database, so that a first prediction model corresponding to the first sensor is obtained by using the first historical data according to a machine learning model (such as linear regression, decision tree, neural network, etc.), and a second prediction model corresponding to the second sensor is obtained by using the second historical data according to a machine learning model (such as linear regression, decision tree, neural network, etc.).

[0105] Exemplarily, first prediction data corresponding to the first target data is obtained according to the first prediction model, and second prediction data corresponding to the second target data is obtained according to the second prediction model. Thus, first error information between the first target data and the first prediction data is calculated using an indicator such as mean square error or mean absolute error, and second error information between the second target data and the second prediction data is calculated using an indicator such as mean square error or mean absolute error.

[0106] Exemplarily, a first weight corresponding to the first sensor and a second weight corresponding to the second sensor are determined according to the first error information and the second error information, and the weights are assigned according to the size of the error information. The smaller the error information, the greater the weight; the larger the error information, the smaller the weight.

[0107] In some embodiments, the determining of the trend characterization value corresponding to the sewage treatment trend of the water conservancy project according to the first abnormal information and the second abnormal information includes: obtaining the first abnormal number corresponding to the first abnormal information, and obtaining the first occurrence frequency corresponding to the first abnormal type from the first abnormal information; determining the first average frequency of occurrence of the first abnormal type in the first monitoring point and the first target weight corresponding to the first abnormal type; obtaining the second abnormal number corresponding to the second abnormal information, and obtaining the second occurrence frequency corresponding to the second abnormal type from the second abnormal information; determining the second average frequency of occurrence of the second abnormal type in the second monitoring point and the second target weight corresponding to the second abnormal type; fusing the first abnormal number, the first occurrence frequency, the first average frequency, the first target weight, and the second abnormal number, the second occurrence frequency, the second average frequency, and the second target weight to determine the trend characterization value corresponding to the sewage treatment trend of the water conservancy project; wherein the trend characterization value is obtained according to the following formula:

[0108]

[0109] in, represents the trend characterization value corresponding to the sewage treatment trend of the water conservancy project at the second monitoring point relative to the first monitoring point, M represents the second abnormality number, m represents the number of types corresponding to the second abnormality type, represents the second target weight corresponding to the i-th second abnormality type, represents the second occurrence frequency of the i-th occurrence of the second abnormal type, represents the second average frequency of the i-th occurrence of the second abnormal type; N represents the first abnormal number, and n represents the number of types corresponding to the first abnormal type. represents the first target weight corresponding to the j-th first abnormality type, represents the first occurrence frequency of the j-th occurrence of the first abnormal type, represents the first average frequency of the j-th occurrence of the first abnormal type.

[0110] Exemplarily, the first abnormal number corresponding to the first abnormal information is obtained, that is, the number of times the pollution abnormality is monitored at the first monitoring point, and the first occurrence frequency corresponding to the first abnormal type is obtained from the first abnormal information; the first average frequency of the first abnormal type in the first monitoring point is determined, and the first target weight corresponding to the first abnormal type is determined according to the pollution degree corresponding to the first abnormal type. The greater the pollution degree, the greater the first target weight.

[0111] Exemplarily, the second abnormal number corresponding to the second abnormal information is obtained, that is, the number of times the pollution abnormality is monitored at the second monitoring point, and the second occurrence frequency corresponding to the second abnormal type is obtained from the second abnormal information; the second average frequency of occurrence of the second abnormal type at the second monitoring point is determined, and the second target weight corresponding to the second abnormal type is determined according to the pollution degree corresponding to the second abnormal type.

[0112] Exemplarily, the trend characterization value corresponding to the sewage treatment trend of the water conservancy project is determined by fusing the first abnormal number, the first occurrence frequency, the first average frequency, the first target weight, and the second abnormal number, the second occurrence frequency, the second average frequency, and the second target weight according to the following formula;

[0113]

[0114] in, It represents the trend characterization value corresponding to the sewage treatment trend of the water conservancy project at the second monitoring point relative to the first monitoring point, M represents the second abnormal number, m represents the number of types corresponding to the second abnormal type, represents the second target weight corresponding to the i-th second abnormal type, represents the second occurrence frequency of the i-th second abnormal type, represents the second average frequency of the occurrence of the i-th second abnormal type; N represents the number of first abnormal times, and n represents the number of types corresponding to the first abnormal type. represents the first target weight corresponding to the jth first abnormal type, represents the first occurrence frequency of the jth first abnormal type, Represents the first average frequency of occurrence of the jth first abnormal type.

[0115] For example, the calculation of the trend characterization value not only considers the frequency and number of occurrences of the abnormal type in the abnormal information, but also considers the weight of the abnormal type. This helps to identify the abnormal type that has the greatest impact on sewage treatment, thereby effectively quantifying the sewage treatment effect. Thus, the sewage treatment effect can be identified in time by calculating the trend characterization value.

[0116] Exemplarily, the trend characterization value is the trend characterization value corresponding to the sewage treatment trend at the second monitoring point relative to the first monitoring point. If the second monitoring point is the water outlet and the first monitoring point is the water inlet, the smaller the trend characterization value data is, the better the sewage treatment effect is, and the target monitoring result is normal sewage treatment. The larger the trend characterization value data is, the worse the sewage treatment effect is, and the target monitoring result is abnormal sewage treatment.

[0117] In some implementations, the method further includes: generating target alarm information according to the target monitoring result, and sending the target alarm information to a user terminal associated with a target user, so that the target user obtains the target monitoring result according to the user terminal.

[0118] For example, when the target monitoring result is that the sewage treatment is abnormal, a corresponding target alarm message is generated. The target alarm message should contain necessary data and description so that the target user can understand the specific situation.

[0119] Exemplarily, the target user and the user terminal associated with the target user are determined. The user terminal can be a mobile phone, computer, tablet or other device. Thus, the contact information of the target user, such as mobile phone number, email address, client ID, etc., is obtained. The generated target alarm information is sent to the user terminal associated with the target user. This usually involves the use of push notifications, emails, text messages and other communication methods. Thus, the target user receives the target alarm information through its associated user terminal and takes corresponding actions according to the received information.

[0120] See also Figure 2 , Figure 2A water conservancy project sewage treatment data monitoring system 200 provided in an embodiment of the present application includes a data acquisition module 201, a first processing module 202, a second processing module 203, a first analysis module 204, a second analysis module 205, an abnormality determination module 206, an abnormality elimination module 207, and a result determination module 208, wherein the data acquisition module 201 is used to obtain first data of a first sensor and second data of a second sensor in a first monitoring point, and obtain third data of a third sensor of the same sensor type as the first sensor and fourth data of a fourth sensor of the same sensor type as the second sensor in a second monitoring point; the first processing module 202 is used to perform abnormality detection on the first data to obtain first abnormality data and perform abnormality detection on the second data to obtain second abnormality data; the second processing module 203 is used to perform abnormality detection on the third data to obtain third abnormality data and perform abnormality detection on the fourth data to obtain fourth abnormality data; the first analysis module 204 is used to determine the first sensor and the second sensor according to the first data, the second data, the first abnormal data, and the second abnormal data. a first correlation relationship between the sensors; a second analysis module 205, for determining a second correlation relationship between the third sensor and the fourth sensor according to the third data, the fourth data, the third abnormal data and the fourth abnormal data; an abnormality determination module 206, for performing abnormality determination on the first abnormal data and the second abnormal data according to the first correlation relationship to obtain first target abnormal data, and performing abnormality determination on the third abnormal data and the fourth abnormal data according to the second correlation relationship to obtain second target abnormal data; an abnormality elimination module 207, for performing data elimination on the first data and the second data according to the first target abnormal data to obtain first target data corresponding to the first data and second target data corresponding to the second data, and performing data elimination on the third data and the fourth data according to the second target abnormal data to obtain third target data corresponding to the third data and fourth target data corresponding to the fourth data; a result determination module 208, for performing data trend prediction on the first target data, the second target data, the third target data and the fourth target data to obtain target monitoring results corresponding to sewage treatment in water conservancy projects.

[0121] In some embodiments, the water conservancy project sewage treatment data monitoring system 200 can be applied to terminal devices.

[0122] 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 conservancy project sewage treatment data monitoring system 200 described above can refer to the corresponding process in the aforementioned water conservancy project sewage treatment data monitoring method embodiment, and will not be repeated here.

[0123] 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 conservancy project sewage treatment data monitoring method provided in the description of the embodiment of the present invention.

[0124] The storage medium may be an internal storage unit of the terminal device in the aforementioned 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 (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the terminal device.

[0125] 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.

[0126] 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.

[0127] 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 method for monitoring sewage treatment data in a water conservancy project, characterized in that: The method comprises: Obtaining first data of a first sensor and second data of a second sensor at a first monitoring point, and obtaining third data of a third sensor of the same sensor type as the first sensor and fourth data of a fourth sensor of the same sensor type as the second sensor at a second monitoring point; Performing anomaly detection on the first data to obtain first anomaly data and performing anomaly detection on the second data to obtain second anomaly data; Performing anomaly detection on the third data to obtain third abnormal data and performing anomaly detection on the fourth data to obtain fourth abnormal data; determining a first association relationship between the first sensor and the second sensor according to the first data, the second data, the first abnormal data, and the second abnormal data; determining a second association relationship between the third sensor and the fourth sensor according to the third data, the fourth data, the third abnormal data and the fourth abnormal data; According to the first association relationship, the first abnormal data and the second abnormal data are abnormally determined to obtain first target abnormal data, and according to the second association relationship, the third abnormal data and the fourth abnormal data are abnormally determined to obtain second target abnormal data; Eliminating the first data and the second data according to the first target abnormal data to obtain first target data corresponding to the first data and second target data corresponding to the second data, and eliminating the third data and the fourth data according to the second target abnormal data to obtain third target data corresponding to the third data and fourth target data corresponding to the fourth data; Determine a first weight corresponding to the first target data and a second weight corresponding to the second target data; fusing the first target data and the second target data according to the first weight and the second weight to obtain first abnormal information corresponding to the first monitoring point; Determine a third weight corresponding to the third target data and a fourth weight corresponding to the fourth target data; fusing the third target data and the fourth target data according to the third weight and the fourth weight to obtain second abnormal information corresponding to the second monitoring point; Obtaining a first abnormal number corresponding to the first abnormal information, and obtaining a first occurrence frequency corresponding to a first abnormal type from the first abnormal information; Determine a first average frequency of occurrence of the first abnormality type in the first monitoring point and a first target weight corresponding to the first abnormality type; Obtaining a second abnormal number corresponding to the second abnormal information, and obtaining a second occurrence frequency corresponding to a second abnormal type from the second abnormal information; Determine a second average frequency of occurrence of the second abnormal type in the second monitoring point and a second target weight corresponding to the second abnormal type; Determine a trend characterization value corresponding to the sewage treatment trend of the water conservancy project by fusing the first abnormal number, the first occurrence frequency, the first average frequency, the first target weight, and the second abnormal number, the second occurrence frequency, the second average frequency, and the second target weight; Obtaining target monitoring results corresponding to sewage treatment by the water conservancy project according to the trend characterization value; The trend characterization value is obtained according to the following formula: ; in, represents the trend characterization value corresponding to the sewage treatment trend of the water conservancy project at the second monitoring point relative to the first monitoring point, M represents the second abnormality number, m represents the number of types corresponding to the second abnormality type, represents the second target weight corresponding to the i-th second abnormality type, represents the second occurrence frequency of the i-th occurrence of the second abnormal type, represents the second average frequency of the i-th occurrence of the second abnormal type; N represents the first abnormal number, and n represents the number of types corresponding to the first abnormal type. represents the first target weight corresponding to the j-th first abnormality type, represents the first occurrence frequency of the j-th occurrence of the first abnormal type, represents the first average frequency of the j-th occurrence of the first abnormal type.

2. The method according to claim 1, characterized in that: The performing anomaly detection on the first data to obtain first abnormal data and performing anomaly detection on the second data to obtain second abnormal data include: Determine a first fitness corresponding to each data in the first data, and determine a first crossover individual and a second crossover individual from the first data according to the first fitness; Obtaining a first target value corresponding to the first crossover individual and a second target value corresponding to the second crossover individual from the first fitness; Calculating an average value corresponding to all the first fitnesses to obtain a first average fitness, and calculating the square of the first average fitness to obtain a first square value; Calculate the square of the difference between the first target value and the second target value to obtain a second square value; Determine a first crossover rate and a first mutation rate according to the first square value and the second square value; Searching the first data according to the first crossover rate and the first mutation rate to obtain a first cluster center, and calculating a first distance between the first data and the first cluster center; Performing data clustering on the first data according to the first distance and the first cluster center to obtain a first clustering result; Determine the first abnormal data corresponding to the first data according to the first clustering result; Determine a second fitness corresponding to each data in the second data, and determine a third crossover individual and a fourth crossover individual according to the second fitness; Obtaining a third target value corresponding to the third crossover individual and a fourth target value corresponding to the fourth crossover individual from the second fitness; Calculate the average value corresponding to all the second fitness values ​​to obtain a second average fitness value, and calculate the square of the second average fitness value to obtain a third square value; Calculate the square of the difference between the third target value and the fourth target value to obtain a fourth square value; Determine a second crossover rate and a second mutation rate according to the third square value and the fourth square value; Searching the second data according to the second crossover rate and the second mutation rate to obtain a second cluster center, and calculating a second distance between the second data and the second cluster center; Performing data clustering on the second data according to the second distance and the second cluster center to obtain a second clustering result; The second abnormal data corresponding to the second data is determined according to the second clustering result.

3. The method according to claim 1, characterized in that The performing anomaly detection on the third data to obtain third abnormal data and performing anomaly detection on the fourth data to obtain fourth abnormal data include: Determine a third fitness corresponding to each data in the third data, and determine a fifth crossover individual and a sixth crossover individual according to the third fitness; Obtaining a fifth target value corresponding to the fifth crossover individual and a sixth target value corresponding to the sixth crossover individual from the third fitness; Calculate the average value corresponding to all the third fitnesses to obtain a third average fitness, and calculate the square of the third average fitness to obtain a fifth square value; Calculating the square of the difference between the fifth target value and the sixth target value to obtain a sixth square value; Determine a third crossover rate and a third mutation rate according to the fifth square value and the sixth square value; Searching the third data according to the third crossover rate and the third mutation rate to obtain a third cluster center, and calculating a third distance between the third data and the third cluster center; Performing data clustering on the third data according to the third distance and the third cluster center to obtain a third clustering result; Determine the third abnormal data corresponding to the third data according to the third clustering result; Determine a fourth fitness corresponding to each data in the fourth data, and determine a seventh crossover individual and an eighth crossover individual according to the fourth fitness; Obtaining a seventh target value corresponding to the seventh crossover individual and an eighth target value corresponding to the eighth crossover individual from the fourth fitness; Calculating an average value corresponding to the fourth fitness to obtain a fourth average fitness, and calculating the square of the fourth average fitness to obtain a seventh square value; Calculate the square of the difference between the seventh target value and the eighth target value to obtain an eighth square value; Determine a fourth crossover rate and a fourth mutation rate according to the seventh square value and the eighth square value; Searching the fourth data according to the fourth crossover rate and the fourth mutation rate to obtain a fourth cluster center, and calculating a fourth distance between the fourth data and the fourth cluster center; Performing data clustering on the fourth data according to the fourth distance and the fourth cluster center to obtain a fourth clustering result; The fourth abnormal data corresponding to the fourth data is determined according to the fourth clustering result.

4. The method according to claim 1, characterized in that: The determining a first association relationship between the first sensor and the second sensor according to the first data, the second data, the first abnormal data, and the second abnormal data includes: Aligning the first data and the second data one by one according to the acquisition time to obtain first aligned data corresponding to the first data and second aligned data corresponding to the second data at the same time; Obtaining any first analysis data in the first aligned data and second analysis data corresponding to the first analysis data in the second aligned data; Obtaining first associated data from the first aligned data corresponding to the first analyzed data satisfying a first difference; Obtaining first corresponding data corresponding to the first associated data from the second aligned data, and calculating whether the first corresponding data and the second analyzed data satisfy a second difference, to obtain second associated data; obtaining a first quantity corresponding to the first associated data and a second quantity corresponding to the second associated data, and determining a first association value between the first analysis data and the second analysis data according to the first quantity and the second quantity; determining a first average correlation value between the first data and the second data according to the first correlation value; Determine a first abnormal position according to the first abnormal data and the second abnormal data, and remove data from the first data according to the first abnormal position to obtain first processed data, and remove data from the second data according to the first abnormal position to obtain second processed data; Aligning the first processed data and the second processed data one by one according to the acquisition time, to obtain third aligned data corresponding to the first processed data and fourth aligned data corresponding to the second processed data at the same time; Obtaining any third analysis data in the third aligned data and fourth analysis data corresponding to the third analysis data in the fourth aligned data; Obtaining, from the third aligned data, third associated data corresponding to the third analyzed data satisfying the first difference; Obtaining second corresponding data corresponding to the third associated data from the fourth aligned data, and calculating whether the second corresponding data and the fourth analyzed data satisfy the second difference, to obtain fourth associated data; obtaining a third quantity corresponding to the third association data and a fourth quantity corresponding to the fourth association data, and determining a second association value between the third analysis data and the fourth analysis data according to the third quantity and the fourth quantity; determining a second average correlation value between the first processed data and the second processed data based on the second correlation value; The first association relationship between the first sensor and the second sensor is determined according to the first average association value and the second average association value.

5. The method according to claim 1, characterized in that: The determining the second association relationship between the third sensor and the fourth sensor according to the third data, the fourth data, the third abnormal data and the fourth abnormal data includes: Aligning the third data and the fourth data one by one according to the acquisition time, to obtain fifth aligned data corresponding to the third data and sixth aligned data corresponding to the fourth data at the same time; Obtain any fifth analysis data in the fifth aligned data and sixth analysis data corresponding to the fifth analysis data in the sixth aligned data; Obtaining, from the fifth aligned data, fifth associated data corresponding to the fifth analyzed data satisfying a third difference; Obtaining third corresponding data corresponding to the fifth associated data from the sixth aligned data, and calculating whether the third corresponding data and the sixth analyzed data satisfy a fourth difference, to obtain sixth associated data; obtaining a fifth quantity corresponding to the fifth association data and a sixth quantity corresponding to the sixth association data, and determining a third association value between the fifth analysis data and the sixth analysis data according to the fifth quantity and the sixth quantity; determining a third average correlation value between the third data and the fourth data according to the third correlation value; Determine a second abnormal position according to the third abnormal data and the fourth abnormal data, and remove data from the third data according to the second abnormal position to obtain third processed data, and remove data from the fourth data according to the second abnormal position to obtain fourth processed data; Aligning the third processed data and the fourth processed data one by one according to the acquisition time, to obtain seventh aligned data corresponding to the third processed data and eighth aligned data corresponding to the fourth processed data at the same time; Obtain any seventh analysis data in the seventh alignment data and eighth analysis data corresponding to the seventh analysis data in the eighth alignment data; Obtaining, from the seventh alignment data, seventh associated data corresponding to the seventh analysis data satisfying the third difference; Obtaining fourth corresponding data corresponding to the seventh associated data from the eighth aligned data, and calculating whether the fourth corresponding data and the eighth analyzed data satisfy the fourth difference, to obtain eighth associated data; obtaining a seventh quantity corresponding to the seventh association data and an eighth quantity corresponding to the eighth association data, and determining a fourth association value between the seventh analysis data and the eighth analysis data according to the seventh quantity and the eighth quantity; determining a fourth average correlation value between the third processed data and the fourth processed data based on the fourth correlation value; The second correlation relationship between the third sensor and the fourth sensor is determined according to the third average correlation value and the fourth average correlation value.

6. The method according to claim 1, characterized in that The determining a first weight corresponding to the first target data and a second weight corresponding to the second target data includes: Obtaining first historical data corresponding to the first sensor and second historical data corresponding to the second sensor; Establishing a first prediction model corresponding to the first sensor according to the first historical data and establishing a second prediction model corresponding to the second sensor according to the second historical data; Determine first prediction data corresponding to the first target data according to the first prediction model and obtain second prediction data corresponding to the second target data according to the second prediction model; Obtaining first error information corresponding to the first sensor according to the first target data and the first prediction data; Obtaining second error information corresponding to the second sensor according to the second target data and the second prediction data; The first weight corresponding to the first target data and the second weight corresponding to the second target data are determined according to the first error information and the second error information.

7. The method according to any one of claims 1 to 6, characterized in that The method further comprises: Target alarm information is generated according to the target monitoring result, and the target alarm information is sent to a user terminal associated with a target user, so that the target user obtains the target monitoring result according to the user terminal.

8. A water conservancy project sewage treatment data monitoring system, characterized in that: include: a data acquisition module, configured to obtain first data of a first sensor and second data of a second sensor at a first monitoring point, and obtain third data of a third sensor of the same sensor type as the first sensor and fourth data of a fourth sensor of the same sensor type as the second sensor at a second monitoring point; A first processing module, configured to perform anomaly detection on the first data to obtain first anomaly data and perform anomaly detection on the second data to obtain second anomaly data; A second processing module, configured to perform anomaly detection on the third data to obtain third abnormal data and perform anomaly detection on the fourth data to obtain fourth abnormal data; a first analysis module, configured to determine a first association relationship between the first sensor and the second sensor according to the first data, the second data, the first abnormal data, and the second abnormal data; a second analysis module, configured to determine a second association relationship between the third sensor and the fourth sensor according to the third data, the fourth data, the third abnormal data and the fourth abnormal data; an abnormality determination module, configured to perform abnormality determination on the first abnormal data and the second abnormal data according to the first association relationship to obtain first target abnormal data, and perform abnormality determination on the third abnormal data and the fourth abnormal data according to the second association relationship to obtain second target abnormal data; an abnormality elimination module, configured to eliminate the first data and the second data according to the first target abnormal data to obtain first target data corresponding to the first data and second target data corresponding to the second data, and to eliminate the third data and the fourth data according to the second target abnormal data to obtain third target data corresponding to the third data and fourth target data corresponding to the fourth data; A result determination module is used to determine a first weight corresponding to the first target data and a second weight corresponding to the second target data; and to fuse the first target data and the second target data according to the first weight and the second weight to obtain first abnormal information corresponding to the first monitoring point; Determine a third weight corresponding to the third target data and a fourth weight corresponding to the fourth target data; obtain second abnormality information corresponding to the second monitoring point by fusing the third target data and the fourth target data according to the third weight and the fourth weight; obtain a first abnormal number corresponding to the first abnormality information, and obtain a first occurrence frequency corresponding to a first abnormality type from the first abnormality information; Determine a first average frequency of occurrence of the first abnormality type in the first monitoring point and a first target weight corresponding to the first abnormality type; Obtaining a second abnormal number corresponding to the second abnormal information, and obtaining a second occurrence frequency corresponding to a second abnormal type from the second abnormal information; Determine a second average frequency of occurrence of the second abnormal type in the second monitoring point and a second target weight corresponding to the second abnormal type; The trend characterization value corresponding to the sewage treatment trend of the water conservancy project is determined by integrating the first abnormal number, the first occurrence frequency, the first average frequency, the first target weight, the second abnormal number, the second occurrence frequency, the second average frequency, and the second target weight; the target monitoring result corresponding to the sewage treatment of the water conservancy project is obtained according to the trend characterization value; wherein the trend characterization value is obtained according to the following formula: ; in, represents the trend characterization value corresponding to the sewage treatment trend of the water conservancy project at the second monitoring point relative to the first monitoring point, M represents the second abnormality number, m represents the number of types corresponding to the second abnormality type, represents the second target weight corresponding to the i-th second abnormality type, represents the second occurrence frequency of the i-th occurrence of the second abnormal type, represents the second average frequency of the i-th occurrence of the second abnormal type; N represents the first abnormal number, and n represents the number of types corresponding to the first abnormal type. represents the first target weight corresponding to the j-th first abnormality type, represents the first occurrence frequency of the j-th occurrence of the first abnormal type, represents the first average frequency of the j-th occurrence of the first abnormal type.

Images