CN117610578A - Intelligent sewage treatment system and method - Google Patents

Abstract

The application discloses an intelligent sewage treatment system and method, which are characterized in that various indexes of sewage are collected through real-time monitoring, a data processing and analyzing algorithm is introduced into the rear end to conduct time sequence change analysis on associated information between parameter indexes of two different time points, and meanwhile, time sequence change analysis is conducted on aeration quantity between the two time points, so that the association relation between the time sequence change semantics of the sewage indexes and the time sequence change semantics of the aeration quantity is excavated, and the aeration quantity at the current time point is controlled, so that the efficiency and quality of sewage treatment are improved. Therefore, the intelligent and automatic sewage treatment can be realized, the risks of manual intervention and misoperation are reduced, and meanwhile, the resource waste is reduced, so that the method has important significance for environmental protection and sustainable development.

Description

Intelligent sewage treatment system and method

Technical Field

The present application relates to the field of sewage treatment, and more particularly, to an intelligent sewage treatment system and method.

Background

The sewage treatment is an important environmental protection engineering, and aims to remove or reduce harmful substances in sewage so as to enable the harmful substances to reach the emission standard or the recycling requirement. In the sewage treatment process, aeration is a common biological treatment method, and air is filled into sewage to promote the growth and metabolism of microorganisms, so that organic matters, nitrogen, phosphorus and other nutrient salts in the sewage are degraded. The aeration rate directly influences the activity of microorganisms and the sewage treatment effect, so that the method is important for the precise control of the aeration rate. However, the current aeration quantity control mode is mostly manually set or is based on regulation, lacks sensitivity and self-adaptability to sewage quality change, and easily causes excessive or insufficient aeration quantity, so that energy consumption is wasted or the treatment effect is poor.

Therefore, an optimized intelligent sewage treatment system is desired to achieve automatic optimal control of aeration amount in the sewage treatment process, thereby improving efficiency and quality of sewage treatment.

Disclosure of Invention

The present application has been made in order to solve the above technical problems. The embodiment of the application provides an intelligent sewage treatment system and method, which are used for carrying out time sequence change analysis on the associated information between the parameter indexes of two different time points by monitoring and collecting various indexes of sewage in real time and introducing a data processing and analyzing algorithm at the rear end, and simultaneously carrying out time sequence change analysis on the aeration quantity between the two time points, so as to mine the association relation between the time sequence change semantics of the sewage indexes and the time sequence change semantics of the aeration quantity, and carrying out aeration quantity control at the current time point so as to improve the efficiency and quality of sewage treatment. Therefore, the intelligent and automatic sewage treatment can be realized, the risks of manual intervention and misoperation are reduced, and meanwhile, the resource waste is reduced, so that the method has important significance for environmental protection and sustainable development.

According to one aspect of the present application, there is provided an intelligent sewage treatment system comprising:

the parameter index data acquisition module is used for acquiring parameter indexes of the treated sewage at a first time point and parameter indexes of the treated sewage at a second time point, wherein the parameter indexes comprise pH value, dissolved oxygen, chemical oxygen demand, ammonia nitrogen, total phosphorus and total nitrogen;

an aeration amount acquisition module for acquiring aeration amounts at a plurality of predetermined time points between the first time point and the second time point;

the sewage index inter-correlation coding module is used for respectively carrying out the inter-sewage index correlation analysis on the parameter index of the treated sewage at the first time point and the parameter index at the second time point so as to obtain a first sewage index feature vector and a second sewage index feature vector;

the sewage index change semantic feature extraction module is used for calculating a difference feature vector between the first sewage index feature vector and the second sewage index feature vector to obtain a sewage index change semantic feature vector;

the aeration local time sequence feature extraction module is used for carrying out local time sequence feature analysis on the aeration time sequence input vectors after arranging the aeration at a plurality of preset time points into the aeration time sequence input vectors according to the time dimension so as to obtain a sequence of the aeration local time sequence feature vectors;

the feature embedded fusion updating module is used for carrying out feature embedded fusion expression optimization updating on the sequence of the aeration local time sequence feature vector and the sewage index change semantic feature vector so as to obtain updated sewage index change semantic features;

and the aeration control module is used for determining that the aeration amount at the current time point should be increased, decreased or maintained based on the updated sewage index change semantic features.

According to another aspect of the present application, there is provided an intelligent sewage treatment method, comprising:

acquiring a parameter index of the treated sewage at a first time point and a parameter index of the treated sewage at a second time point, wherein the parameter indexes comprise pH value, dissolved oxygen, chemical oxygen demand, ammonia nitrogen, total phosphorus and total nitrogen;

acquiring aeration amounts at a plurality of predetermined time points between the first time point and the second time point;

respectively carrying out correlation analysis on the parameter indexes of the treated sewage at a first time point and the parameter indexes at a second time point to obtain a first sewage index feature vector and a second sewage index feature vector;

calculating a differential feature vector between the first sewage index feature vector and the second sewage index feature vector to obtain a sewage index change semantic feature vector;

after arranging the aeration quantity at a plurality of preset time points into aeration quantity time sequence input vectors according to a time dimension, carrying out local time sequence feature analysis on the aeration quantity time sequence input vectors to obtain a sequence of aeration quantity local time sequence feature vectors;

performing feature embedded fusion expression optimization updating on the sequence of the aeration local time sequence feature vector and the sewage index change semantic feature vector to obtain updated sewage index change semantic features;

based on the updated sewage index change semantic features, it is determined that the aeration amount at the current time point should be increased, decreased or maintained.

Compared with the prior art, the intelligent sewage treatment system and the intelligent sewage treatment method provided by the application have the advantages that various indexes of the collected sewage are monitored in real time, the data processing and analyzing algorithm is introduced into the rear end to conduct time sequence change analysis on the associated information between the parameter indexes of two different time points, meanwhile, the time sequence change analysis is conducted on the aeration quantity between the two time points, so that the association relation between the time sequence change semantics of the sewage indexes and the time sequence change semantics of the aeration quantity is excavated, and the aeration quantity at the current time point is controlled, so that the efficiency and the quality of sewage treatment are improved. Therefore, the intelligent and automatic sewage treatment can be realized, the risks of manual intervention and misoperation are reduced, and meanwhile, the resource waste is reduced, so that the method has important significance for environmental protection and sustainable development.

Drawings

The foregoing and other objects, features and advantages of the present application will become more apparent from the following more particular description of embodiments of the present application, as illustrated in the accompanying drawings. The accompanying drawings are included to provide a further understanding of embodiments of the application and are incorporated in and constitute a part of this specification, illustrate the application and not constitute a limitation to the application. In the drawings, like reference numerals generally refer to like parts or steps.

FIG. 1 is a block diagram of an intelligent wastewater treatment system according to an embodiment of the present application;

FIG. 2 is a system architecture diagram of an intelligent wastewater treatment system according to an embodiment of the present application;

FIG. 3 is a block diagram of an aeration local time sequence feature extraction module in an intelligent wastewater treatment system according to an embodiment of the present application;

FIG. 4 is a block diagram of an aeration rate control module in an intelligent wastewater treatment system according to an embodiment of the present application;

FIG. 5 is a block diagram of a feature correction unit in an intelligent wastewater treatment system according to an embodiment of the present application;

fig. 6 is a flow chart of an intelligent wastewater treatment method according to an embodiment of the present application.

Detailed Description

Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application and not all of the embodiments of the present application, and it should be understood that the present application is not limited by the example embodiments described herein.

As used in this application and in the claims, the terms "a," "an," "the," and/or "the" are not specific to the singular, but may include the plural, unless the context clearly dictates otherwise. In general, the terms "comprises" and "comprising" merely indicate that the steps and elements are explicitly identified, and they do not constitute an exclusive list, as other steps or elements may be included in a method or apparatus.

Although the present application makes various references to certain modules in a system according to embodiments of the present application, any number of different modules may be used and run on a user terminal and/or server. The modules are merely illustrative, and different aspects of the systems and methods may use different modules.

Flowcharts are used in this application to describe the operations performed by systems according to embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in order precisely. Rather, the various steps may be processed in reverse order or simultaneously, as desired. Also, other operations may be added to or removed from these processes.

Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application and not all of the embodiments of the present application, and it should be understood that the present application is not limited by the example embodiments described herein.

The existing aeration quantity control modes are mostly manually set or are based on regulation, lack of sensitivity and self-adaptability to sewage quality change, and easily cause excessive or insufficient aeration quantity, so that energy consumption is wasted or treatment effect is poor. Therefore, an optimized intelligent sewage treatment system is desired to achieve automatic optimal control of aeration amount in the sewage treatment process, thereby improving efficiency and quality of sewage treatment.

In the technical scheme of this application, provided an intelligent sewage treatment system. Fig. 1 is a block diagram of an intelligent wastewater treatment system according to an embodiment of the present application. Fig. 2 is a system architecture diagram of an intelligent sewage treatment system according to an embodiment of the present application. As shown in fig. 1 and 2, an intelligent sewage treatment system 300 according to an embodiment of the present application includes: a parameter index data acquisition module 310, configured to acquire a parameter index of the treated sewage at a first time point and a parameter index of the treated sewage at a second time point, where the parameter indexes include a pH value, dissolved oxygen, chemical oxygen demand, ammonia nitrogen, total phosphorus, and total nitrogen; an aeration amount acquisition module 320 for acquiring aeration amounts at a plurality of predetermined time points between the first time point and the second time point; the sewage index inter-correlation encoding module 330 is configured to perform a sewage index inter-correlation analysis on the parameter index of the treated sewage at the first time point and the parameter index at the second time point to obtain a first sewage index feature vector and a second sewage index feature vector; a sewage index change semantic feature extraction module 340, configured to calculate a differential feature vector between the first sewage index feature vector and the second sewage index feature vector to obtain a sewage index change semantic feature vector; an aeration local time sequence feature extraction module 350, configured to arrange aeration amounts at the plurality of predetermined time points into aeration amount time sequence input vectors according to a time dimension, and then perform local time sequence feature analysis on the aeration amount time sequence input vectors to obtain a sequence of aeration amount local time sequence feature vectors; the feature embedded fusion updating module 360 is configured to perform feature embedded fusion expression optimization updating on the sequence of the aeration local time sequence feature vector and the sewage index change semantic feature vector to obtain updated sewage index change semantic features; the aeration quantity control module 370 is used for determining that the aeration quantity at the current time point should be increased, decreased or maintained based on the updated sewage index change semantic features.

In particular, the parameter index data collection module 310 is configured to obtain a parameter index of the treated sewage at a first time point and a parameter index of the treated sewage at a second time point, where the parameter indexes include pH, dissolved oxygen, chemical oxygen demand, ammonia nitrogen, total phosphorus and total nitrogen. It should be understood that the pollution degree and the treatment effect of the sewage can be reflected by considering various indexes of the sewage, such as the association relationship among the pH value, the dissolved oxygen, the chemical oxygen demand, the ammonia nitrogen, the total phosphorus and the total nitrogen. Therefore, by analyzing the correlation change trend among various indexes of the sewage, parameters in the sewage treatment process, such as aeration rate, can be automatically adjusted, so that the sewage treatment effect is optimized. Thus, first, a parameter index of the treated sewage at a first time point and a parameter index at a second time point are acquired.

In particular, the aeration amount acquisition module 320 is configured to acquire aeration amounts at a plurality of predetermined time points between the first time point and the second time point. Wherein, the aeration quantity refers to the gas quantity supplied to the water body through the aeration device in the water treatment process. In water treatment, aeration is commonly used to increase the dissolved oxygen content of water to promote microbial growth and metabolic activity during biological treatment. The aeration rate has important influence on the water treatment effect and the running cost. If the aeration is too small, the dissolved oxygen content in the water is insufficient, and the growth and metabolic activity of microorganisms are limited, which may result in poor wastewater treatment effect. If the aeration amount is too large, not only energy waste is caused, but also the problems of excessive bubbles, stirring of suspended matters, foam overflow and the like can be caused. Thus, first, the aeration amounts at a plurality of predetermined time points between the first time point and the second time point are acquired.

In particular, the inter-sewage-index association encoding module 330 is configured to perform inter-sewage-index association analysis on the parameter index of the treated sewage at the first time point and the parameter index at the second time point to obtain a first sewage-index feature vector and a second sewage-index feature vector. Considering that in the intelligent sewage treatment system, various index parameters of sewage can reflect the pollution degree and treatment effect of the sewage. And the parameter indexes of the treated sewage have association relations based on sample dimensions at different time points. Therefore, in order to capture the sample association characteristics of the sewage indexes at each time point, in the technical scheme of the application, after the parameter indexes of the treated sewage at the first time point and the parameter indexes at the second time point are further arranged as input vectors, feature mining is performed in the sewage index feature extractor based on the full connection layer, so that the association feature information between the parameter indexes of the treated sewage at the first time point and the parameter index sample data at the second time point is extracted, and the first sewage index feature vector and the second sewage index feature vector are obtained.

In particular, the sewage indicator change semantic feature extraction module 340 is configured to calculate a differential feature vector between the first sewage indicator feature vector and the second sewage indicator feature vector to obtain a sewage indicator change semantic feature vector. In the technical scheme of the application, in order to compare the difference and the change condition between the parameter index association characteristic of the treated sewage at the first time point and the parameter index association characteristic at the second time point, the differential feature vector between the first sewage index feature vector and the second sewage index feature vector is further calculated to obtain a sewage index change semantic feature vector. It should be appreciated that the sewage indicator change semantic feature vector may represent a change rate or trend of the sewage indicator, such as an increase, decrease, or remain stable, so as to more intuitively reflect a dynamic change in the sewage treatment process. The dynamic change characteristic information has important significance for judging the effect of the sewage treatment process and adjusting the treatment parameters to optimize the sewage treatment process. Specifically, calculating a differential feature vector between the first sewage index feature vector and the second sewage index feature vector to obtain a sewage index change semantic feature vector includes: calculating a differential feature vector between the first sewage index feature vector and the second sewage index feature vector by using the following differential formula to obtain a sewage index change semantic feature vector; wherein, the formula is:wherein->Representing the first sewage index feature vector, < >>Representing the second sewage index feature vector, < >>Representing the sewage index change semantic feature vector, < >>Representing the difference by location.

Specifically, the aeration local time sequence feature extraction module 350 is configured to perform local time sequence feature analysis on the aeration time sequence input vectors to obtain a sequence of aeration local time sequence feature vectors after the aeration at the plurality of predetermined time points are arranged into the aeration time sequence input vectors according to a time dimension. In particular, in one specific example of the present application, as shown in fig. 3, the aeration amount local time series feature extraction module 350 includes: an aeration local time sequence vector segmentation unit 351, configured to arrange aeration amounts at the plurality of predetermined time points into aeration amount time sequence input vectors according to a time dimension, and then vector-segment the aeration amount time sequence input vectors to obtain a sequence of aeration amount local time sequence input vectors; and an aeration local time sequence feature analysis unit 352 for passing the sequence of aeration local time sequence input vectors through an aeration local time sequence feature extractor based on a one-dimensional convolution layer to obtain the sequence of aeration local time sequence feature vectors.

Specifically, the aeration local time sequence vector segmentation unit 351 is configured to, after arranging aeration amounts at the plurality of predetermined time points according to a time dimension into aeration time sequence input vectors, perform vector segmentation on the aeration time sequence input vectors to obtain a sequence of aeration local time sequence input vectors. It will be appreciated that aeration is a common treatment method in sewage treatment by injecting gas into the body of water to provide dissolved oxygen for the degradation reaction by microorganisms. The aeration rate directly influences the supply amount of dissolved oxygen, thereby influencing the activity of microorganisms and the sewage treatment effect. And, consider that the aeration quantity has a time sequence dynamic change rule between a first time point and a second time point, that is, the aeration quantities at a plurality of preset time points have a time sequence dynamic association relation. Therefore, in order to analyze and characterize the time sequence change condition and trend of the aeration quantity more fully and accurately so as to better understand and analyze the influence of the aeration quantity on the sewage treatment effect, in the technical scheme of the application, after the aeration quantity at a plurality of preset time points is further arranged into the aeration quantity time sequence input vector according to the time dimension, vector segmentation is performed on the aeration quantity time sequence input vector so as to obtain a sequence of the aeration quantity local time sequence input vector.

Specifically, the aeration local time sequence feature analysis unit 352 is configured to pass the sequence of aeration local time sequence input vectors through an aeration local time sequence feature extractor based on a one-dimensional convolution layer to obtain the sequence of aeration local time sequence feature vectors. In other words, in the technical scheme of the application, the sequence of the aeration local time sequence input vector is subjected to feature mining in the aeration local time sequence feature extractor based on the one-dimensional convolution layer so as to extract the local time sequence dynamic feature information of the aeration in the time dimension between the first time point and the second time point, thereby obtaining the sequence of the aeration local time sequence feature vector. Therefore, the method is beneficial to better capturing the change characteristics of the aeration quantity in different time periods, thereby being beneficial to understanding the relationship between the aeration quantity and the sewage treatment effect and providing more accurate decision basis for an intelligent sewage treatment system. More specifically, passing the sequence of aeration rate local time sequence input vectors through an aeration rate time sequence feature extractor based on a one-dimensional convolution layer to obtain the sequence of aeration rate local time sequence feature vectors, comprising: each layer of the aeration amount time sequence characteristic extractor based on the one-dimensional convolution layer is used for respectively carrying out input data in forward transfer of the layer: carrying out convolution processing on input data to obtain a convolution characteristic diagram; pooling the convolution feature images based on a feature matrix to obtain pooled feature images; performing nonlinear activation on the pooled feature map to obtain an activated feature map; the output of the last layer of the aeration quantity time sequence feature extractor based on the one-dimensional convolution layer is a sequence of the aeration quantity local time sequence feature vectors, and the input of the first layer of the aeration quantity time sequence feature extractor based on the one-dimensional convolution layer is a sequence of the aeration quantity local time sequence input vectors.

Notably, one-dimensional convolutional layers are one type of layer commonly used in Convolutional Neural Networks (CNNs) for processing one-dimensional time series data. It can effectively extract local patterns and features in the time series data. The input to the one-dimensional convolution layer is a one-dimensional signature sequence, which may be a time sequence, a signal sequence, or other one-dimensional data. This layer generates an output signature by convolving the input data with a set of learnable filters (also called convolution kernels).

It should be noted that, in other specific examples of the present application, after the aeration amounts at the plurality of predetermined time points are arranged according to the time dimension into the aeration amount time sequence input vector in other manners, the aeration amount time sequence input vector is subjected to local time sequence feature analysis to obtain a sequence of aeration amount local time sequence feature vectors, for example: collecting aeration amount data at a plurality of predetermined time points; arranging the collected aeration quantity data according to the time dimension, and constructing an aeration quantity time sequence input vector; and carrying out local time sequence characteristic analysis on the aeration quantity time sequence input vector. The local timing feature analysis may employ various methods, such as sliding window, convolutional Neural Network (CNN), etc.; and moving a window with a fixed size on the aeration quantity time sequence input vector by a sliding window method. The aeration magnitude within the window can be regarded as a local timing feature vector. Different window sizes and sliding steps may be selected to obtain local features on different time scales; and taking the aeration time sequence input vector as input, and designing a proper CNN structure to extract local time sequence characteristics. CNNs can capture local patterns and features in input vectors through convolution and pooling layers; and processing the aeration quantity time sequence input vector according to the local time sequence characteristic analysis method to obtain a sequence of aeration quantity local time sequence characteristic vectors.

In particular, the feature embedded fusion updating module 360 is configured to perform feature embedded fusion expression optimization updating on the sequence of the aeration local time sequence feature vector and the sewage index change semantic feature vector to obtain updated sewage index changeSemantic features. Considering that the aeration amount is an important parameter in the sewage treatment process, the aeration amount has a direct influence on the supply of dissolved oxygen and the activity of microorganisms, thereby influencing the sewage treatment effect. The time sequence change characteristics of the associated information among the parameter indexes of the sewage reflect the treatment effect of the pollution degree of the sewage. And, it is also considered that the aeration amount may have an influence on the change of the sewage index. Therefore, in order to perform updating optimization of the sewage index time sequence change semantic feature by utilizing the influence of the aeration quantity on the sewage index time sequence change, thereby improving the accuracy of aeration quantity real-time control, in the technical scheme of the application, the sewage index change semantic feature vector is further subjected to feature expression optimization based on aeration information so as to obtain the updated sewage index change semantic feature vector based on the sequence of the aeration quantity local time sequence feature vector. By carrying out embedded feature expression optimization updating based on each local time sequence feature of the aeration quantity and the time sequence change semantic feature of the sewage index, the time sequence change condition of the sewage index can be expressed more accurately, and therefore a more accurate basis is provided for real-time control of the aeration quantity. Specifically, the method is used for optimizing the feature expression based on aeration information on the sewage index change semantic feature vector based on the sequence of the aeration quantity local time sequence feature vector to obtain the updated sewage index change semantic feature vector according to the following formula:wherein (1)>Representing the change semantic feature vector of the sewage index,representation->Matrix of->A scale equal to the sewage index change semantic feature vector，/>Is->Matrix of->Equal to the number of aeration local time sequence eigenvectors in the sequence of aeration local time sequence eigenvectors,/->Is a Sigmoid function->Is a weight coefficient>And->Convolution operation representing a 1 x 1 convolution kernel, < >>Each aeration amount local time sequence characteristic vector in the sequence representing the aeration amount local time sequence characteristic vector,/for each aeration amount local time sequence characteristic vector>And representing the scale of each aeration local time sequence characteristic vector in the sequence of the aeration local time sequence characteristic vectors.

In particular, the aeration amount control module 370 is configured to determine that the aeration amount at the current time point should be increased, decreased, or maintained based on the updated sewage index change semantic feature. In particular, in one specific example of the present application, as shown in fig. 4, the aeration amount control module 370 includes: a feature correction unit 371, configured to perform feature correction on the updated sewage index change semantic feature vector to obtain a corrected updated sewage index change semantic feature vector; and the aeration quantity real-time control unit 372 is used for passing the corrected updated sewage index change semantic feature vector through a classifier to obtain a classification result, wherein the classification result is used for indicating that the aeration quantity at the current time point should be increased, decreased or maintained.

Specifically, the feature correction unit 371 is configured to perform feature correction on the updated sewage index change semantic feature vector to obtain a corrected updated sewage index change semantic feature vector. In particular, in one specific example of the present application, as shown in fig. 5, the feature correction unit 371 includes: a correspondence corrector unit 3711, configured to perform correspondence correction on the sequence of the aeration local time sequence feature vector and the sewage index change semantic feature vector to obtain a correspondence correction feature vector; and the correction feature fusion subunit 3712 is configured to fuse the corresponding correction feature vector with the updated sewage index change semantic feature vector to obtain the corrected updated sewage index change semantic feature vector.

More specifically, the correspondence correction subunit 3711 is configured to perform correspondence correction on the sequence of the aeration local time sequence feature vectors and the sewage index change semantic feature vector to obtain a correspondence correction feature vector. In particular, in the above technical solution, the sequence of aeration local time-series feature vectors expresses the local time-domain time-series related feature of the aeration in the global time domain under the local time domain determined by vector segmentation, and the sewage index change semantic feature vector expresses the global time-domain differential feature of the parameter index related feature of the treated sewage, whereby, when the sewage index change semantic feature vector is subjected to feature expression optimization based on aeration information based on the sequence of aeration local time-series feature vectors, the feature correspondence sparsity caused by the difference between the aeration time-series related feature and the parameter index related time-domain differential feature in the source data sample space domain and the feature time-series related domain is considered, so that it is desirable to perform the correspondence feature vector optimization based on the feature expression significance and the criticality of each of the sequence of aeration local time-series feature vectors and the sewage index change semantic feature vector, thereby improving the updated sewage index change semantic feature vectorExpression effect. Based on the above, the applicant of the present application performs correspondence correction on the sequence of the aeration amount local time sequence feature vector and the sewage index change semantic feature vector, specifically expressed as:wherein->Is a cascade feature vector obtained by cascading the sequence of aeration local time sequence feature vectors, and +.>Is the sewage index change semantic feature vector,representing the position-wise evolution of the feature vector, < >>And->Feature vector +.>And->Reciprocal of maximum eigenvalue, ++>And->Is a weight superparameter,/->Representing multiplication by location +.>Representing difference in position->Is the correspondence correction feature vector. Here, a pre-divided partial group of feature value sets is obtained by the sequence of the aeration amount partial time sequence feature vector and the evolution value of each feature value of the sewage index variation semantic feature vector, and the sequence of the aeration amount partial time sequence feature vector and the key maximum value feature of the sewage index variation semantic feature vector are regressed therefrom, so that the per-position significance distribution of feature values can be promoted based on the concept of furthest point sampling, thereby performing sparse correspondence control between feature vectors by key features with significance distribution to realize correspondence correction feature vector>And restoring the sequence of the aeration local time sequence characteristic vector and the original manifold geometry of the sewage index change semantic characteristic vector. Thus, the correspondence correction feature vector +.>And the expression effect of the updated sewage index change semantic feature vector can be improved by fusing the updated sewage index change semantic feature vector, so that the accuracy of a classification result obtained by the classifier is improved. Therefore, the aeration rate can be automatically controlled in real time based on the time sequence change condition of the parameter index of the sewage so as to improve the efficiency and quality of sewage treatment.

More specifically, the correction feature fusion subunit 3712 is configured to fuse the corresponding correction feature vector with the updated sewage indicator change semantic feature vector to obtain the corrected updated sewage indicator change semantic feature vector. It should be understood that by fusing the correspondence correction feature vector with the updated sewage index change semantic feature vector, the updated sewage index change semantic feature vector can be better supplemented and corrected, and the integrity and accuracy of the features can be improved.

It should be noted that, in other specific examples of the present application, the updated sewage indicator change semantic feature vector may be further subjected to feature correction by other manners to obtain a corrected updated sewage indicator change semantic feature vector, for example: collecting correction data related to the change in the wastewater index; determining a proper correction method according to the specific problems and the characteristics of data, and correcting the elements of each feature vector; and taking the corrected feature vector as a corrected updated sewage index change semantic feature vector.

Specifically, the aeration amount real-time control unit 372 is configured to pass the corrected updated sewage index change semantic feature vector through a classifier to obtain a classification result, where the classification result is used to indicate that the aeration amount at the current time point should be increased, decreased or maintained. That is, classification processing is performed using the time-series variation characteristic information of the sewage index updated by the local time-series variation characteristic of the aeration amount, thereby performing aeration amount control at the current time point to improve efficiency and quality of sewage treatment. More specifically, passing the corrected updated sewage index change semantic feature vector through a classifier to obtain a classification result for indicating that the aeration amount at the current time point should be increased, decreased or maintained, including: performing full-connection coding on the corrected updated sewage index change semantic feature vector by using a plurality of full-connection layers of the classifier to obtain a coded classification feature vector; and passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.

It should be noted that, in other specific examples of the present application, it may also be determined that the aeration amount at the current time point should be increased, decreased or maintained based on the updated sewage index change semantic feature in other manners, for example: and collecting sewage index data at the current time point. Such data may include various measurements relating to water quality, water level, or other relevant indicators; extracting semantic features from the sewage index data by using a proper feature extraction method; the extracted semantic features are combined into a feature vector. This feature vector will be used to determine whether the aeration amount at the current point in time should be increased, decreased or remain unchanged; training a classification model or a regression model by using the existing training data set, and training the feature vector and the corresponding aeration quantity label; and predicting the feature vector of the current time point by using the trained model. And judging whether the aeration quantity should be increased, decreased or kept unchanged according to the output of the model.

As described above, the intelligent sewage treatment system 300 according to the embodiment of the present application may be implemented in various wireless terminals, such as a server having an intelligent sewage treatment algorithm, or the like. In one possible implementation, the intelligent wastewater treatment system 300 according to embodiments of the present application may be integrated into the wireless terminal as a software module and/or hardware module. For example, the intelligent wastewater treatment system 300 may be a software module in the operating system of the wireless terminal, or may be an application developed for the wireless terminal; of course, the intelligent wastewater treatment system 300 may also be one of a number of hardware modules of the wireless terminal.

Alternatively, in another example, the intelligent wastewater treatment system 300 and the wireless terminal may be separate devices, and the intelligent wastewater treatment system 300 may be connected to the wireless terminal through a wired and/or wireless network and transmit interactive information in a agreed data format.

Further, an intelligent sewage treatment method is also provided.

Fig. 6 is a flow chart of an intelligent wastewater treatment method according to an embodiment of the present application. As shown in fig. 6, the intelligent sewage treatment method according to the embodiment of the present application includes the steps of: s1, acquiring a parameter index of treated sewage at a first time point and a parameter index of the treated sewage at a second time point, wherein the parameter indexes comprise pH value, dissolved oxygen, chemical oxygen demand, ammonia nitrogen, total phosphorus and total nitrogen; s2, acquiring aeration amounts at a plurality of preset time points between the first time point and the second time point; s3, respectively carrying out correlation analysis on the parameter indexes of the treated sewage at the first time point and the parameter indexes at the second time point to obtain a first sewage index feature vector and a second sewage index feature vector; s4, calculating a differential feature vector between the first sewage index feature vector and the second sewage index feature vector to obtain a sewage index change semantic feature vector; s5, after the aeration quantity at the plurality of preset time points is arranged into aeration quantity time sequence input vectors according to the time dimension, carrying out local time sequence feature analysis on the aeration quantity time sequence input vectors to obtain a sequence of aeration quantity local time sequence feature vectors; s6, carrying out feature embedded fusion expression optimization updating on the sequence of the aeration local time sequence feature vector and the sewage index change semantic feature vector to obtain updated sewage index change semantic features; and S7, determining that the aeration amount at the current time point should be increased, decreased or maintained based on the updated sewage index change semantic features.

In summary, the intelligent sewage treatment method according to the embodiment of the application is clarified, by monitoring and collecting various indexes of sewage in real time, introducing a data processing and analyzing algorithm at the rear end to perform time sequence change analysis on the associated information between the parameter indexes at two different time points, and simultaneously performing time sequence change analysis on the aeration quantity between the two time points, so as to mine the association relation between the time sequence change semantics of the sewage indexes and the time sequence change semantics of the aeration quantity, and performing aeration quantity control at the current time point to improve the efficiency and quality of sewage treatment. Therefore, the intelligent and automatic sewage treatment can be realized, the risks of manual intervention and misoperation are reduced, and meanwhile, the resource waste is reduced, so that the method has important significance for environmental protection and sustainable development.

The foregoing description of the embodiments of the present disclosure has been presented for purposes of illustration and description, and is not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the various embodiments described. The terminology used herein was chosen in order to best explain the principles of the embodiments, the practical application, or the improvement of technology in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.

Claims (8)

1. An intelligent sewage treatment system, comprising:

the parameter index data acquisition module is used for acquiring parameter indexes of the treated sewage at a first time point and parameter indexes of the treated sewage at a second time point, wherein the parameter indexes comprise pH value, dissolved oxygen, chemical oxygen demand, ammonia nitrogen, total phosphorus and total nitrogen;

an aeration amount acquisition module for acquiring aeration amounts at a plurality of predetermined time points between the first time point and the second time point;

the sewage index inter-correlation coding module is used for respectively carrying out the inter-sewage index correlation analysis on the parameter index of the treated sewage at the first time point and the parameter index at the second time point so as to obtain a first sewage index feature vector and a second sewage index feature vector;

the sewage index change semantic feature extraction module is used for calculating a difference feature vector between the first sewage index feature vector and the second sewage index feature vector to obtain a sewage index change semantic feature vector;

the aeration local time sequence feature extraction module is used for carrying out local time sequence feature analysis on the aeration time sequence input vectors after arranging the aeration at a plurality of preset time points into the aeration time sequence input vectors according to the time dimension so as to obtain a sequence of the aeration local time sequence feature vectors;

the feature embedded fusion updating module is used for carrying out feature embedded fusion expression optimization updating on the sequence of the aeration local time sequence feature vector and the sewage index change semantic feature vector so as to obtain updated sewage index change semantic features;

and the aeration control module is used for determining that the aeration amount at the current time point should be increased, decreased or maintained based on the updated sewage index change semantic features.

2. The intelligent wastewater treatment system of claim 1, wherein the wastewater index inter-correlation encoding module is configured to: and respectively arranging the parameter indexes of the treated sewage at a first time point and the parameter indexes at a second time point into input vectors, and then obtaining the first sewage index feature vector and the second sewage index feature vector through a sewage index feature extractor based on a full connection layer.

3. The intelligent wastewater treatment system of claim 2, wherein the aeration local time sequence feature extraction module comprises:

the aeration local time sequence vector segmentation unit is used for carrying out vector segmentation on the aeration time sequence input vectors after arranging the aeration quantity of the plurality of preset time points into the aeration time sequence input vectors according to the time dimension so as to obtain a sequence of the aeration local time sequence input vectors;

and the aeration local time sequence characteristic analysis unit is used for enabling the sequence of the aeration local time sequence input vector to pass through an aeration local time sequence characteristic extractor based on a one-dimensional convolution layer to obtain the sequence of the aeration local time sequence characteristic vector.

4. The intelligent wastewater treatment system of claim 3, wherein the feature embedded fusion update module is configured to: and performing feature expression optimization on the sewage index change semantic feature vector based on aeration information based on the sequence of the aeration quantity local time sequence feature vector to obtain an updated sewage index change semantic feature vector so as to obtain the updated sewage index change semantic feature.

5. The intelligent wastewater treatment system of claim 4, wherein the feature embedded fusion update module is configured to: the sequence is used for optimizing the characteristic expression based on aeration information of the sewage index change semantic feature vector based on the sequence of the aeration quantity local time sequence feature vector according to the following formula to obtain the updated sewage index change semantic feature vectorWherein the formula is:wherein (1)>Representing the change semantic feature vector of sewage indexes, +.>Representation->Matrix of->Equal to the scale of the sewage index variation semantic feature vector,/->Is->Matrix of->Equal to the number of aeration local time sequence eigenvectors in the sequence of aeration local time sequence eigenvectors,/->Is a Sigmoid function->Is a weight coefficient>And->Convolution operation representing a 1 x 1 convolution kernel, < >>Each aeration amount local time sequence characteristic vector in the sequence of aeration amount local time sequence characteristic vectors,and representing the scale of each aeration local time sequence characteristic vector in the sequence of the aeration local time sequence characteristic vectors.

6. The intelligent wastewater treatment system of claim 5, wherein the aeration quantity control module comprises:

the characteristic correction unit is used for carrying out characteristic correction on the updated sewage index change semantic feature vector so as to obtain a corrected updated sewage index change semantic feature vector;

and the aeration quantity real-time control unit is used for passing the corrected updated sewage index change semantic feature vector through a classifier to obtain a classification result, wherein the classification result is used for indicating that the aeration quantity at the current time point should be increased, decreased or maintained.

7. The intelligent sewage treatment system according to claim 6, wherein the characteristic correcting unit includes:

a correspondence corrector unit, configured to perform correspondence correction on the sequence of the aeration local time sequence feature vector and the sewage index change semantic feature vector to obtain a correspondence correction feature vector;

and the correction feature fusion subunit is used for fusing the corresponding correction feature vector with the updated sewage index change semantic feature vector to obtain the corrected updated sewage index change semantic feature vector.

8. An intelligent sewage treatment method is characterized by comprising the following steps:

acquiring a parameter index of the treated sewage at a first time point and a parameter index of the treated sewage at a second time point, wherein the parameter indexes comprise pH value, dissolved oxygen, chemical oxygen demand, ammonia nitrogen, total phosphorus and total nitrogen;

acquiring aeration amounts at a plurality of predetermined time points between the first time point and the second time point;

respectively carrying out correlation analysis on the parameter indexes of the treated sewage at a first time point and the parameter indexes at a second time point to obtain a first sewage index feature vector and a second sewage index feature vector;

calculating a differential feature vector between the first sewage index feature vector and the second sewage index feature vector to obtain a sewage index change semantic feature vector;

after arranging the aeration quantity at a plurality of preset time points into aeration quantity time sequence input vectors according to a time dimension, carrying out local time sequence feature analysis on the aeration quantity time sequence input vectors to obtain a sequence of aeration quantity local time sequence feature vectors;

performing feature embedded fusion expression optimization updating on the sequence of the aeration local time sequence feature vector and the sewage index change semantic feature vector to obtain updated sewage index change semantic features;

based on the updated sewage index change semantic features, it is determined that the aeration amount at the current time point should be increased, decreased or maintained.