Sewage treatment supervision method and system based on artificial intelligence control

Abstract

The invention discloses a sewage treatment supervision method and system based on artificial intelligence control. The method comprises the steps of data acquisition, water quality prediction, water quality data clustering treatment and sewage treatment supervision. The invention belongs to the field of data processing, and particularly relates to a sewage treatment supervision method and system based on artificial intelligence control. According to the scheme, water quality deviation measurement and time decay factors are introduced for sampling point optimization, sampling points and time are finely adjusted, outdated data influences are avoided, then high-sensitivity sampling points are selected, and the accuracy of sewage treatment supervision is improved. The sewage treatment supervision accuracy is improved; according to the method, adaptive distance measurement is introduced, and by designing feature weights of environmental adaptability and timeliness response, the weights of different features can be adaptively adjusted according to environmental changes, so that an analysis result is more biased to real-time data, early warning and decision making in the water quality treatment process are more timely, the problem of relying on outdated data is avoided, and the water quality treatment efficiency is improved. And the sewage treatment supervision effect is improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and specifically to a sewage treatment supervision method and system based on artificial intelligence control. Background Art

[0002] Sewage treatment supervision methods refer to monitoring, evaluating and optimizing the effects of the sewage treatment process through a series of technical and management means to ensure the normal operation of sewage treatment facilities and the compliance of water quality. These methods usually involve data collection, data analysis, real-time monitoring, early warning mechanisms and decision support systems, with the aim of promptly discovering and correcting problems in the treatment process, thereby effectively preventing water pollution and protecting the environment and public health. However, general sewage treatment supervision methods have poor timeliness and are unable to promptly reflect equipment operating status or changes in the external environment, which in turn leads to insufficient water quality prediction accuracy and poor accuracy in sewage treatment supervision. General sewage treatment supervision methods are unable to quickly adapt to changes in the sewage treatment environment, cannot promptly discover water quality problems, and the impact of key characteristics is underestimated or overestimated, affecting water quality prediction and supervision effectiveness. Summary of the invention

[0003] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides a sewage treatment supervision method and system based on artificial intelligence control. The general sewage treatment supervision method has poor timeliness and cannot reflect the equipment operation status or external environment changes in time, which leads to insufficient water quality prediction accuracy and poor sewage treatment supervision accuracy. This scheme introduces time series covariance to capture the time change trend of water quality data; through the dynamic adjustment of the interpolation weight coefficient, it ensures that it can better adapt to water quality changes when the equipment is abnormal or the meteorological conditions change drastically; introduces water quality deviation measurement and time attenuation factor to optimize the sampling points, and makes detailed adjustments on the sampling points and time to avoid the influence of outdated data, while ensuring the weights of important sampling points. The weight is large, and then high-sensitivity sampling points are selected to improve the accuracy of sewage treatment supervision; in view of the problems that general sewage treatment supervision methods cannot quickly adapt to changes in the sewage treatment environment, cannot detect water quality problems in time, and the impact of key features is underestimated or overestimated, affecting water quality prediction and supervision effects, this scheme introduces adaptive distance measurement, and through the design of feature weights that respond to environmental adaptability and timeliness, the weights of different features can be adaptively adjusted according to environmental changes in different time periods, meteorological conditions and equipment operating conditions, so that the analysis results are more biased towards real-time data, making early warnings and decisions in the water quality treatment process more timely, avoiding the problem of relying on outdated data, and thus improving the sewage treatment supervision effect.

[0004] The technical solution adopted by the present invention is as follows: The sewage treatment supervision method based on artificial intelligence control provided by the present invention comprises the following steps:

[0005] Step S1: data acquisition;

[0006] Step S2: water quality prediction;

[0007] Step S3: water quality data clustering processing;

[0008] Step S4: Sewage treatment supervision.

[0009] Furthermore, in step S1, the data acquisition is to collect various data of the sewage treatment plant through sensors and monitoring equipment; and perform preliminary cleaning, conversion and standardization on the collected raw data.

[0010] Further, in step S2, the water quality prediction uses the data set obtained in step S1 to predict the future water quality data set; specifically, the following steps are included:

[0011] Step S21: Calculate the time series covariance of the input data; used to capture the trend of water quality changes over time, expressed as: ; where C(·) is the time series covariance; cov(·) is the covariance; x i and x j is the sampling location; t is the sampling time; Z(·) is the water quality data value at the sampling location; then the water quality prediction is performed, and the prediction point The predicted water quality value is: ;in, is the predicted water quality data value; n is the total number of sampling points; i is the sampling point index; is the interpolation weight coefficient;

[0012] Step S22: Update the interpolation weight coefficient, and dynamically adjust the interpolation weight coefficient by introducing the device state S and the meteorological parameter Q; expressed as: ; Update the interpolation weight coefficient, recalculate the time series covariance of the data and continue to update the interpolation weight coefficient; when the weight coefficient converges, the iteration is completed and go to step S23;

[0013] Step S23: Optimize sampling point selection and calculate the expected position correspondence; pre-set a sensitivity threshold, select the predicted water quality data value of the sampling point expected to be higher than the sensitivity threshold, and obtain the final water quality data set; introduce adaptive sampling weights, and the expectation is expressed as: ; ; Where E(x) is the expectation at position x; is the minimum water quality target value; is the predicted water quality value at sampling point x; is the standard deviation of the predicted value at the sampling point x; is the cumulative distribution function, which is used to measure the relative position between the target water quality value and the current predicted value; is the probability density function; is the adaptive sampling weight; D(·) is the water quality deviation measure of the sampling point; is the current time; and is the adjustment coefficient.

[0014] Further, in step S3, the water quality data clustering process is to perform k-means clustering analysis on the final water quality data set; adaptive distance is used in clustering, feature weights are introduced, and feature weights are designed based on environmental adaptability and timeliness response; the adaptive distance is expressed as: ; ;in, is the adaptive distance between water quality data X and water quality data Y; N is the feature dimension; u is the feature index; and is the eigenvalue of the u-th dimension of water quality data; is the feature weight at time t; is the impact factor; It is the water quality deviation measure of the u-th dimension of the water quality data X at time t; is the time-effective attenuation factor; T is the sampling time of the original cluster center of water quality data; the objective function M of cluster analysis is expressed as: ; Where K is the total number of clustering result categories, and k is the category index; is the k-th clustering result; is the cluster center; W is the cluster weight.

[0015] Furthermore, in step S4, the sewage treatment supervision is carried out by clustering the water quality data to obtain clustering results, each cluster represents a similar water quality state, each water quality data will be assigned to a cluster, different clusters represent different water quality types, and the water quality category is defined according to the clustering results; when a new predicted water quality data set arrives, the distance between the new predicted water quality data and the known cluster center is calculated, and the data is assigned to the nearest cluster centroid, and the corresponding classification result is obtained according to the cluster to which the new data point belongs; sewage treatment supervision is realized based on the classification result of the water quality data.

[0016] The sewage treatment supervision system based on artificial intelligence control provided by the present invention includes a data acquisition module, a water quality prediction module, a water quality data clustering processing module and a sewage treatment supervision module;

[0017] The data acquisition module collects various data of the sewage treatment plant through sensors and monitoring equipment;

[0018] The water quality prediction module combines time series covariance and dynamic interpolation weights to predict future water quality data;

[0019] The water quality data clustering processing module uses adaptive distance and feature weight to perform k-means clustering analysis on the predicted water quality data;

[0020] The sewage treatment supervision module realizes sewage treatment supervision based on the water quality data clustering results.

[0021] The beneficial effects achieved by the present invention using the above scheme are as follows:

[0022] (1) In view of the fact that general sewage treatment supervision methods have poor timeliness and cannot reflect equipment operating status or external environmental changes in a timely manner, which leads to insufficient water quality prediction accuracy and poor sewage treatment supervision accuracy, this scheme introduces time series covariance to capture the temporal trend of water quality data. By dynamically adjusting the interpolation weight coefficient, it ensures that it can better adapt to water quality changes when equipment is abnormal or meteorological conditions change drastically. It introduces water quality deviation measurement and time attenuation factor to optimize sampling points, and makes detailed adjustments on sampling points and time to avoid the influence of outdated data. At the same time, it ensures that the weights of important sampling points are large, and then selects highly sensitive sampling points to improve the accuracy of sewage treatment supervision.

[0023] (2) In view of the problems that general sewage treatment supervision methods cannot quickly adapt to changes in the sewage treatment environment, cannot detect water quality problems in a timely manner, and the impact of key features is underestimated or overestimated, which affects the water quality prediction and supervision effect, this scheme introduces an adaptive distance metric. By designing feature weights that respond to environmental adaptability and timeliness, the weights of different features can be adaptively adjusted according to environmental changes in different time periods, meteorological conditions and equipment operating conditions, making the analysis results more biased towards real-time data, making early warnings and decisions in the water quality treatment process more timely, avoiding the problem of relying on outdated data, and thus improving the sewage treatment supervision effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A schematic diagram of the process of the sewage treatment supervision method based on artificial intelligence control provided by the present invention;

[0025] Figure 2 A schematic diagram of a sewage treatment monitoring system based on artificial intelligence control provided by the present invention;

[0026] Figure 3 It is a schematic diagram of the process of step S2.

[0027] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention. DETAILED DESCRIPTION

[0028] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0029] In the description of the present invention, it is necessary to understand that terms such as “upper”, “lower”, “front”, “back”, “left”, “right”, “top”, “bottom”, “inside” and “outside” indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the referred system or element must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as a limitation on the present invention.

[0030] Example 1, see Figure 1 The sewage treatment supervision method based on artificial intelligence control provided by the present invention comprises the following steps:

[0031] Step S1: Data acquisition: collect various data of the sewage treatment plant through sensors and monitoring equipment;

[0032] Step S2: Water quality prediction: combining time series covariance and dynamic interpolation weights to predict future water quality data;

[0033] Step S3: clustering of water quality data; using adaptive distance and feature weights to perform k-means clustering analysis on the predicted water quality data;

[0034] Step S4: sewage treatment supervision; sewage treatment supervision is implemented based on the water quality data clustering results.

[0035] Example 2, see Figure 1 This embodiment is based on the above embodiment. In step S1, data acquisition is to collect various data of the sewage treatment plant through sensors and monitoring equipment, including water quality indicators, flow, temperature, meteorological data, sampling time, sampling location and equipment status data; the collected raw data is preliminarily cleaned, converted and standardized to ensure data quality, remove outliers, and fill in missing data; the water quality indicators include pH, COD, BOD, ammonia nitrogen and total phosphorus; the meteorological data include temperature and humidity; the equipment includes pump stations, aerators and sedimentation tanks.

[0036] Example 3, see Figure 1 and Figure 3 This embodiment is based on the above embodiment. In step S2, water quality prediction uses the data set obtained in step S1 to predict the future water quality data set. Specifically, the following steps are included:

[0037] Step S21: Calculate the time series covariance of the input data to capture the changing trend of water quality over time and reduce the prediction error caused by time series problems, especially under the influence of meteorological changes and equipment operation status, expressed as: ; where C(·) is the time series covariance; cov(·) is the covariance; x i and x j is the sampling location; t is the sampling time; Z(·) is the water quality data value at the sampling location; then the water quality prediction is performed, and the prediction point The predicted water quality value is: ;in, is the predicted water quality data value; n is the total number of sampling points; i is the sampling point index; is the interpolation weight coefficient. The initial interpolation weight coefficient is obtained by solving the equation system based on the covariance matrix between the sampling points and the constraint that the sum of all weight coefficients is 1, and then solving the equation system by Gaussian elimination method.

[0038] Step S22: Update the interpolation weight coefficient. By introducing the device status S and the meteorological parameter Q, the interpolation weight coefficient is dynamically adjusted so that the weight coefficient can change in real time when the device is operating abnormally or the meteorological conditions change drastically; it is expressed as: ; Update the interpolation weight coefficient, recalculate the time series covariance of the data and continue to update the interpolation weight coefficient; when the weight coefficient converges, the iteration is completed and go to step S23;

[0039] Step S23: Optimize the sampling point selection and calculate the corresponding expectation of the position. The larger the expectation, the greater the influence of position x on the prediction result, so the sensitivity of position x is higher. A sensitivity threshold is set in advance, and the predicted water quality data value of the sampling point with an expectation higher than the sensitivity threshold is selected to obtain the final water quality data set. Adaptive sampling weights are introduced, and the expectation is expressed as: ; ; Where E(x) is the expectation at position x; is the minimum water quality target value; is the predicted water quality value at sampling point x; is the standard deviation of the predicted value at the sampling point x; is the cumulative distribution function, which is used to measure the relative position between the target water quality value and the current predicted value; It is a probability density function, which reflects the density of the target water quality value near the current predicted water quality value; is the adaptive sampling weight; D(·) is the water quality deviation measure of the sampling point, which represents the difference between the historical water quality data of the sampling point and the target water quality value; is the current time; and is the adjustment coefficient.

[0040] By performing the above operations, the general sewage treatment supervision method has poor timeliness and cannot reflect the equipment operation status or external environment changes in time, which leads to insufficient water quality prediction accuracy and poor accuracy of sewage treatment supervision. This scheme introduces time series covariance to capture the temporal trend of water quality data. Through dynamic adjustment of the interpolation weight coefficient, it ensures that it can better adapt to water quality changes when equipment is abnormal or meteorological conditions change drastically. It introduces water quality deviation measurement and time attenuation factor to optimize sampling points, and makes detailed adjustments on sampling points and time to avoid the influence of outdated data. At the same time, it ensures that the weights of important sampling points are large, and then selects highly sensitive sampling points to improve the accuracy of sewage treatment supervision.

[0041] Example 4, see Figure 1 This embodiment is based on the above embodiment. In step S3, the water quality data clustering process is to perform k-means clustering analysis on the final water quality data set; adaptive distance is used during clustering, feature weights are introduced, and feature weights are designed based on environmental adaptability and timeliness response. During the sewage treatment process, different time periods, equipment operating conditions and meteorological conditions may have different effects on different features. Based on environmental adaptability, the feature weights change with the changes in the environment, thereby improving the accuracy of the clustering results; based on timeliness response, the water quality data is compared with the sampling data at the same time as much as possible; the adaptive distance is expressed as: ; ;in, is the adaptive distance between water quality data X and water quality data Y; N is the feature dimension; u is the feature index; and is the eigenvalue of the u-th dimension of water quality data; is the feature weight at time t; is the impact factor; It is the water quality deviation measure of the u-th dimension of the water quality data X at time t; is the time-effective attenuation factor; T is the sampling time of the original cluster center of water quality data; the objective function M of cluster analysis is expressed as: ; Where K is the total number of clustering result categories, and k is the category index; is the k-th clustering result; is the cluster center; W is the cluster weight.

[0042] By performing the above operations, this solution introduces adaptive distance measurement to address the problems that general sewage treatment supervision methods cannot quickly adapt to changes in the sewage treatment environment, cannot detect water quality problems in a timely manner, and the impact of key features is underestimated or overestimated, affecting water quality prediction and supervision effects. By designing feature weights that respond to environmental adaptability and timeliness, the weights of different features can be adaptively adjusted according to environmental changes in different time periods, meteorological conditions, and equipment operating conditions, so that the analysis results are more biased towards real-time data, making early warnings and decisions in the water quality treatment process more timely, avoiding the problem of relying on outdated data, and thereby improving the sewage treatment supervision effect.

[0043] Example 5, see Figure 1 This embodiment is based on the above embodiment. In step S4, sewage treatment supervision is performed by clustering water quality data to obtain clustering results. Each cluster represents a similar water quality state. Each water quality data will be assigned to a cluster. Different clusters represent different water quality types. Water quality categories are defined according to the clustering results. The water quality categories are normal water quality, lightly polluted water quality, heavily polluted water quality and extremely polluted water quality. When a new predicted water quality data set arrives, the distance between the new predicted water quality data and the known cluster center is calculated and assigned to the nearest cluster centroid. According to the cluster to which the new data point belongs, the corresponding classification result is obtained. Sewage treatment supervision is implemented based on the classification result of water quality data. When the water quality result is heavily polluted water quality or extremely polluted water quality, it indicates that the sewage treatment is improper, and the corresponding management personnel of the sampling point are warned.

[0044] Example 6, see Figure 2 , this embodiment is based on the above embodiment. The sewage treatment supervision system based on artificial intelligence control provided by the present invention includes a data acquisition module, a water quality prediction module, a water quality data clustering processing module and a sewage treatment supervision module;

[0045] The data acquisition module collects various data of the sewage treatment plant through sensors and monitoring equipment; and sends the data to the water quality prediction module;

[0046] The water quality prediction module combines time series covariance and dynamic interpolation weights to predict future water quality data; and sends the data to the water quality data clustering processing module;

[0047] The water quality data clustering processing module uses adaptive distance and feature weight to perform k-means clustering analysis on the predicted water quality data; and sends the data to the sewage treatment supervision module;

[0048] The sewage treatment supervision module realizes sewage treatment supervision based on the water quality data clustering results.

[0049] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0050] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.

[0051] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.

Claims

1. A sewage treatment supervision method based on artificial intelligence control, characterized in that: The method comprises the following steps: Step S1: data acquisition; Step S2: water quality prediction; Step S3: water quality data clustering processing; Step S4: Sewage treatment supervision.

2. The sewage treatment supervision method based on artificial intelligence control according to claim 1 is characterized in that: In step S2, the water quality prediction uses the data set obtained in step S1 to predict the future water quality data set; The specific steps include: Step S21: Calculate the time series covariance of the input data; used to capture the trend of water quality changes over time, expressed as: ; where C(·) is the time series covariance; cov(·) is the covariance; x i and x j is the sampling location; t is the sampling time; Z(·) is the water quality data value at the sampling location; then the water quality prediction is performed, and the prediction point The predicted water quality value is: ;in, is the predicted water quality data value; n is the total number of sampling points; i is the sampling point index; is the interpolation weight coefficient; Step S22: Update the interpolation weight coefficient, and dynamically adjust the interpolation weight coefficient by introducing the device state S and the meteorological parameter Q; expressed as: ; Update the interpolation weight coefficient, recalculate the time series covariance of the data and continue to update the interpolation weight coefficient; when the weight coefficient converges, the iteration is completed and go to step S23; Step S23: Optimize sampling point selection and calculate the expected position correspondence; pre-set a sensitivity threshold, select the predicted water quality data value of the sampling point expected to be higher than the sensitivity threshold, and obtain the final water quality data set; introduce adaptive sampling weights, and the expectation is expressed as: ; ; Where E(x) is the expectation at position x; is the minimum water quality target value; is the predicted water quality value at sampling point x; is the standard deviation of the predicted value at the sampling point x; is the cumulative distribution function, which is used to measure the relative position between the target water quality value and the current predicted value; is the probability density function; is the adaptive sampling weight; D(·) is the water quality deviation measure of the sampling point; is the current time; and is the adjustment coefficient.

3. The sewage treatment supervision method based on artificial intelligence control according to claim 2 is characterized in that: In step S3, the water quality data clustering process is to perform k-means clustering analysis on the final water quality data set; adaptive distance is used in clustering, feature weights are introduced, and feature weights are designed based on environmental adaptability and timeliness response; The adaptive distance is expressed as: ; ;in, is the adaptive distance between water quality data X and water quality data Y; N is the feature dimension; u is the feature index; and is the eigenvalue of the u-th dimension of water quality data; is the feature weight at time t; is the impact factor; It is the water quality deviation measure of the u-th dimension of the water quality data X at time t; is the time-effective attenuation factor; T is the sampling time of the original cluster center of water quality data; the objective function M of cluster analysis is expressed as: ; Where K is the total number of clustering result categories, and k is the category index; is the k-th clustering result; is the cluster center; W is the cluster weight.

4. The sewage treatment supervision method based on artificial intelligence control according to claim 3 is characterized in that: In step S1, the data acquisition is to collect various data of the sewage treatment plant through sensors and monitoring equipment; and perform preliminary cleaning, conversion and standardization on the collected raw data.

5. The sewage treatment supervision method based on artificial intelligence control according to claim 4 is characterized in that: In step S4, the sewage treatment supervision is carried out by clustering the water quality data to obtain clustering results. Each cluster represents a similar water quality state. Each water quality data will be assigned to a cluster. Different clusters represent different water quality types. The water quality category is defined according to the clustering results. When a new predicted water quality data set arrives, the distance between the new predicted water quality data and the known cluster center is calculated and assigned to the nearest cluster centroid. According to the cluster to which the new data point belongs, the corresponding classification result is obtained. Sewage treatment supervision is realized based on the classification results of the water quality data.

6. A sewage treatment supervision system based on artificial intelligence control, used to implement the sewage treatment supervision method based on artificial intelligence control as described in any one of claims 1 to 5, characterized in that: It includes data acquisition module, water quality prediction module, water quality data clustering processing module and sewage treatment supervision module; The data acquisition module collects various data of the sewage treatment plant through sensors and monitoring equipment; The water quality prediction module combines time series covariance and dynamic interpolation weights to predict future water quality data; The water quality data clustering processing module uses adaptive distance and feature weight to perform k-means clustering analysis on the predicted water quality data; The sewage treatment supervision module realizes sewage treatment supervision based on the water quality data clustering results.

Images