CN117247180B

CN117247180B - Device and method for treating high-fluorine high-ammonia-nitrogen wastewater

Info

Publication number: CN117247180B
Application number: CN202311318042.5A
Authority: CN
Inventors: 沈财方; 邓国敢; 李惠林; 安国栋; 沈勘力; 高勇; 张会林; 钱志强
Original assignee: Zhejiang Zhiyuan Environmental Technology Co ltd
Current assignee: Zhejiang Zhiyuan Environmental Technology Co ltd
Priority date: 2023-10-11
Filing date: 2023-10-11
Publication date: 2024-07-09
Anticipated expiration: 2043-10-11
Also published as: CN117247180A

Abstract

A device and a method for treating high-fluorine high-ammonia nitrogen wastewater are disclosed. The device comprises a pretreatment unit, a fluoride ion removal unit, a nitrogen removal unit and a post-treatment unit, and is characterized in that the nitrogen removal unit is a high-efficiency biological carrier reactor, and the high-efficiency biological carrier reactor comprises dissolved oxygen and ORP detection instruments, an electric proportional valve, an oxygenation fan, an aerator and an automatic control component; wherein the automatic control component is used for controlling the oxygenation fan based on the oxygen content and ORP value acquired by the dissolved oxygen and ORP detection instrument. Thus, the treatment and automatic adjustment of the high-fluorine high-ammonia nitrogen wastewater can be realized.

Description

Device and method for treating high-fluorine high-ammonia-nitrogen wastewater

Technical Field

The present application relates to the field of wastewater treatment, and more particularly, to an apparatus and method for treating high fluorine and high ammonia nitrogen wastewater.

Background

With the rapid development of the photovoltaic industry in China, the photovoltaic wastewater treatment is also receiving attention. The photovoltaic wastewater mainly comes from auxiliary waste liquid and cleaning wastewater generated in the processes of cutting, grinding and slicing the silicon rod and grinding, corroding and polishing the silicon wafer, and the treatment difficulty mainly comprises high concentration of organic matters and suspended matters and contains pollutants such as fluoride ions, ammonia nitrogen and the like.

The traditional denitrification process comprises the methods of stripping, magnesium ammonium phosphate method, biological denitrification and the like. The stripping process is suitable for high-concentration ammonia nitrogen wastewater, but needs to be recovered after acid absorption, and has higher equipment investment and operation cost. Struvite produced by the magnesium ammonium phosphate process is commonly used as a fertilizer but cannot treat industrial wastewater containing toxic substances. Therefore, in most cases, a biological denitrification process is employed.

The biological denitrification process mainly utilizes autotrophic aerobic nitrifying bacteria to oxidize ammonia nitrogen into nitrite and nitrate, and then utilizes heterotrophic anaerobic denitrifying bacteria to reduce nitrite and nitrate into nitrogen. This method is relatively economical and environmentally friendly, but still presents some challenges in treating high-fluorine high-ammonia nitrogen wastewater. For example, high fluorine concentrations may inhibit the growth and metabolic activity of microorganisms, affecting the efficiency of the nitrification and denitrification processes and thus the removal of ammonia nitrogen from the wastewater. In addition, high ammonia nitrogen concentrations can also cause microbial overload and excessive nitrogen production, increasing the difficulty and cost of subsequent treatments.

Accordingly, an optimized apparatus for treating high fluorine and high ammonia nitrogen wastewater is desired.

Disclosure of Invention

The present application has been made to solve the above-mentioned technical problems. The embodiment of the application provides a device and a method for treating high-fluorine high-ammonia-nitrogen wastewater. The method can realize the treatment and automatic adjustment of the high-fluorine high-ammonia nitrogen wastewater.

According to one aspect of the present application, there is provided an apparatus for treating high-fluorine high-ammonia nitrogen wastewater, comprising a pretreatment unit, a fluoride ion removal unit, a nitrogen removal unit and a post-treatment unit, wherein the nitrogen removal unit is a high-efficiency bio-carrier reactor comprising dissolved oxygen and ORP detection instruments, an electric proportional valve, an oxygenation fan, an aerator and an automatic control assembly; wherein the automatic control component is used for controlling the oxygenation fan based on the oxygen content and ORP value acquired by the dissolved oxygen and ORP detection instrument.

According to another aspect of the present application, there is provided a method for treating high fluorine and high ammonia nitrogen wastewater, comprising:

Collecting oxygen content and ORP values of the mixed solution at a plurality of predetermined time points within a predetermined time period through a dissolved oxygen and ORP detector disposed in the efficient bio-carrier reactor;

Performing time sequence feature interaction correlation analysis on the oxygen content and ORP values at a plurality of preset time points to obtain time sequence interaction features of the oxygen content-ORP values; and

Based on the oxygen content-ORP value timing interaction characteristics, the power value of the oxygenation fan at the current time point is determined to be increased or decreased.

Compared with the prior art, the device and the method for treating the high-fluorine high-ammonia nitrogen wastewater provided by the application comprise a pretreatment unit, a fluoride ion removal unit, a nitrogen removal unit and a post-treatment unit, and are characterized in that the nitrogen removal unit is a high-efficiency biological carrier reactor, and the high-efficiency biological carrier reactor comprises dissolved oxygen and ORP (oxidation/reduction/oxidation) detection instruments, an electric proportional valve, an oxygenation fan, an aerator and an automatic control component; wherein the automatic control component is used for controlling the oxygenation fan based on the oxygen content and ORP value acquired by the dissolved oxygen and ORP detection instrument. Thus, the treatment and automatic adjustment of the high-fluorine high-ammonia nitrogen wastewater can be realized.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the description of the embodiments will be briefly introduced below, the following drawings not being drawn to scale with respect to actual dimensions, emphasis instead being placed upon illustrating the gist of the present application.

Fig. 1 is a schematic block diagram of the automatic control assembly in an apparatus for treating high fluorine and high ammonia nitrogen wastewater according to an embodiment of the present application.

Fig. 2 is a diagram of an embodiment of the present application. And a block diagram schematic diagram of the data parameter time sequence correlation analysis unit in the device for treating the high-fluorine high-ammonia nitrogen wastewater.

FIG. 3 is a flow chart of a method for treating high fluorine and high ammonia nitrogen wastewater in accordance with an embodiment of the application.

Fig. 4 is a schematic diagram of a system architecture of a method for treating high fluorine and high ammonia nitrogen wastewater according to an embodiment of the application.

Fig. 5 is an application scenario diagram of an apparatus for treating high-fluorine high-ammonia nitrogen wastewater according to an embodiment of the present application.

FIG. 6 is a process flow diagram of a high fluorine and high ammonia nitrogen wastewater treatment device according to another embodiment of the application.

FIG. 7 shows the concentration change of the pollutants after the treatment of the sewage with different concentrations in the chemical coagulation process of example 1.

FIG. 8 shows the change of the pollutant removal rate after the treatment of sewage with different concentrations in the chemical coagulation process of example 1.

FIG. 9 shows the change in contaminant concentration after wastewater treatment at different residence times for the electrochemical process of example 1 under condition 1.

FIG. 10 is a graph showing the change in contaminant concentration after wastewater treatment at different residence times for the electrochemical process of example 1 under condition 2.

FIG. 11 shows the COD removal amount and removal rate change in the sewage treatment cycle of example 2.

FIG. 12 shows the ammonia nitrogen removal rate and the ammonia nitrogen removal rate change in the wastewater treatment cycle of example 2.

FIG. 13 shows the total nitrogen removal and the change in the removal rate in the wastewater treatment cycle of example 2.

Detailed Description

The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some, but not all embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are also within the scope of the application.

As used in the specification and in the claims, the terms "a," "an," "the," and/or "the" are not specific to a singular, but may include a plurality, 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.

A flowchart is used in the present application to describe the operations performed by a system 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, exemplary 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 embodiments of the present application and not all embodiments of the present application, and it should be understood that the present application is not limited by the example embodiments described herein.

Aiming at the technical problems, in the treatment process of high-fluorine high-ammonia nitrogen wastewater, a high-efficiency biological carrier reactor is provided to realize the high-efficiency denitrification process, the high-efficiency biological carrier reactor comprises components such as dissolved oxygen and ORP (oxidation reduction potential) detection instruments, an electric proportional valve, an oxygenation fan, an aerator, a controller and the like, wherein the controller can automatically adjust the power value of the oxygenation fan and the opening value of the electric proportional valve in real time according to the dissolved oxygen and the ORP of the mixed solution.

It will be appreciated that in the treatment of high fluorine and high ammonia nitrogen wastewater, adequate oxygen supply is critical to the normal growth and metabolism of microorganisms. By automatically adjusting the frequency of the fan and the opening of the electric valve, the dissolved oxygen in the wastewater can be ensured to be maintained at a proper level, good oxygen conditions are provided, and the nitrification and denitrification processes are promoted. Also, excessive oxygen supply may cause peroxidation to occur, generating excessive hydroxyl radicals, and causing damage to microorganisms. Through automatically adjusting the frequency of the fan and the opening of the electric valve, the peroxidation can be avoided, and the growth and metabolic activity of microorganisms can be protected. That is, by monitoring the dissolved oxygen and ORP values, the metabolic status and activity level of microorganisms in the wastewater can be known in real time. According to the indexes, the frequency of the fan and the opening of the electric valve can be adaptively adjusted to control the oxygen supply and the aeration effect, so that the growth environment of microorganisms is optimized, the treatment and the automatic adjustment of the high-fluorine high-ammonia nitrogen wastewater are realized, and the high-fluorine high-ammonia nitrogen wastewater treatment device has high treatment efficiency and economy.

Correspondingly, the device for treating the high-fluorine high-ammonia nitrogen wastewater comprises a pretreatment unit, a fluoride ion removal unit, a nitrogen removal unit and a post-treatment unit, wherein the nitrogen removal unit is a high-efficiency biological carrier reactor, and the high-efficiency biological carrier reactor comprises dissolved oxygen and ORP (oxidation-reduction potential) detection instruments, an electric proportional valve, an oxygenation fan, an aerator and an automatic control component; wherein the automatic control component is used for controlling the oxygenation fan based on the oxygen content and ORP value acquired by the dissolved oxygen and ORP detection instrument.

Fig. 1 is a schematic block diagram of the automatic control assembly in an apparatus for treating high fluorine and high ammonia nitrogen wastewater according to an embodiment of the present application. As shown in fig. 1, the apparatus for treating wastewater with high fluorine and high ammonia nitrogen according to the embodiment of the present application, the automatic control assembly 100 includes: a data collection unit 110 for collecting oxygen contents and ORP values of a mixed solution at a plurality of predetermined time points within a predetermined period of time through the dissolved oxygen and ORP detector disposed at the efficient bio-carrier reactor; a data parameter time sequence correlation analysis unit 120, configured to perform time sequence feature interaction correlation analysis on the oxygen content and ORP values at the multiple predetermined time points to obtain oxygen content-ORP value time sequence interaction features; and an oxygenation fan control unit 130 for determining, based on the oxygen content-ORP value timing interaction characteristics, whether the power value of the oxygenation fan at the current point in time should be increased or decreased.

Specifically, in the technical scheme of the application, firstly, the oxygen content and ORP values of a mixed solution collected by a dissolved oxygen and ORP detector deployed in a high-efficiency biological carrier reactor at a plurality of preset time points in a preset time period are obtained, so that the metabolic state of microorganisms is monitored in real time and the oxygen supply is regulated, and the biological denitrification process is optimized. In particular, the dissolved oxygen is the oxygen required for respiration by the microorganism. By monitoring the amount of dissolved oxygen in the mixed solution, the oxygen supply to which the microorganism is exposed can be known. If the dissolved oxygen is too low, it may cause the microorganisms to be anoxic or the nitrification and denitrification reactions may not be performed normally. Too high dissolved oxygen may initiate peroxidation and damage to the microorganisms. Thus, monitoring the amount of dissolved oxygen can help regulate the oxygen supply, maintain a suitable oxygen level, and promote normal metabolic activity of the microorganism. The ORP (oxidation-reduction potential) is a measure of the redox capacity in a solution. During biological denitrification, metabolic activity of microorganisms causes oxidation-reduction reactions in wastewater. By monitoring the ORP value, the redox status during the metabolism of the microorganism can be understood. Changes in ORP values can reflect the activity and metabolic levels of the microorganism. A higher ORP value may indicate a stronger metabolic activity of the microorganism, while a lower ORP value may indicate a weaker metabolic activity of the microorganism. By monitoring ORP values, oxygen supply and aeration effects can be timely regulated, and the growth environment of microorganisms can be optimized.

Next, considering that the oxygen content and the ORP value of the mixed solution have a dynamic change rule of time sequence in the time dimension, and have a cooperative association relationship of time sequence in the time dimension, and the two have an influence on the metabolic state of the microorganism and the supply of regulated oxygen together, in the technical scheme of the application, the oxygen content and the ORP value at a plurality of preset time points need to be respectively arranged into an oxygen content time sequence input vector and an ORP value time sequence input vector according to the time dimension, so that the distribution information of the oxygen content and the ORP value of the mixed solution in time sequence is respectively integrated.

Then, in order to improve the capability of capturing the characteristic of the time sequence fine change of the oxygen content and the ORP value of the mixed solution, in the technical scheme of the application, the oxygen content time sequence input vector and the ORP value time sequence input vector are further subjected to up-sampling processing to obtain an up-sampling oxygen content time sequence input vector and an up-sampling ORP value time sequence input vector, so that the density and the smoothness of the oxygen content and the ORP value data of the mixed solution in time sequence are increased, and the time sequence characteristics of the oxygen content and the ORP value are conveniently and better represented later.

Further, considering that the oxygen content and the ORP value of the mixed solution have fluctuation and uncertainty in the time dimension, the oxygen content and the ORP value exhibit different time-series dynamic change characteristics in different time period spans in the time dimension, and the oxygen content and the ORP value may be weak in time-series change, so that it is difficult to perform sufficient time-series characteristic extraction by a conventional characteristic extraction method. Therefore, in the technical scheme of the application, the up-sampling oxygen content time sequence input vector and the up-sampling ORP value time sequence input vector are further subjected to feature mining in a time sequence feature extractor based on a multi-scale neighborhood feature extraction module, so that time sequence multi-scale neighborhood associated feature information of the oxygen content and the ORP value of the mixed solution under different time spans is extracted respectively, and thus the oxygen content time sequence multi-scale feature vector and the ORP value time sequence multi-scale feature vector are obtained.

And then, performing feature interaction based on an attention mechanism on the oxygen content time sequence multi-scale feature vector and the ORP value time sequence multi-scale feature vector by using an inter-feature attention layer to obtain an oxygen content-ORP value time sequence interaction feature vector, so as to capture the association and interaction between the oxygen content time sequence change feature and the time sequence change feature of the OPR value. It should be appreciated that since the goal of the traditional attention mechanism is to learn an attention weight matrix, a greater weight is given to important features and a lesser weight is given to secondary features, thereby selecting more critical information to the current task goal. This approach is more focused on weighting the importance of individual features, while ignoring the dependency between features. The attention layer between the features can capture the correlation and the mutual influence between the time sequence change features of the oxygen content and the time sequence change features of the OPR value through the feature interaction based on an attention mechanism, learn the dependency relationship between different features, and interact and integrate the features according to the dependency relationship, so that an oxygen content-ORP value time sequence interaction feature vector is obtained, and the microbial metabolism state in the solution is analyzed more accurately and the supply of oxygen is regulated.

Accordingly, as shown in fig. 2, the data parameter timing relationship analysis unit 120 includes: a data parameter time sequence arrangement subunit 121, configured to arrange the oxygen content and ORP values at the plurality of predetermined time points into an oxygen content time sequence input vector and an ORP value time sequence input vector according to a time dimension, respectively; an up-sampling subunit 122, configured to up-sample the oxygen content time-series input vector and the ORP value time-series input vector to obtain an up-sampled oxygen content time-series input vector and an up-sampled ORP value time-series input vector; a parameter multi-scale time sequence feature extraction subunit 123, configured to perform multi-scale time sequence feature extraction on the up-sampled oxygen content time sequence input vector and the up-sampled ORP value time sequence input vector by using a time sequence feature extractor based on a deep neural network model, so as to obtain an oxygen content time sequence multi-scale feature vector and an ORP value time sequence multi-scale feature vector; and a multi-parameter time sequence feature interaction correlation analysis subunit 124, configured to perform feature interaction correlation analysis on the oxygen content time sequence multi-scale feature vector and the ORP value time sequence multi-scale feature vector to obtain an oxygen content-ORP value time sequence interaction feature vector as the oxygen content-ORP value time sequence interaction feature. It should be understood that the data parameter timing arrangement subunit 121 arranges the data in time order for subsequent processing. In the upsampling subunit 122, upsampling refers to increasing the density of data points, typically by interpolation or the like, in order to represent the data more finely. The parameter multi-scale time sequence feature extraction subunit 123 extracts features of different scales through a neural network model to capture a time sequence mode and related information of input data, and obtains an oxygen content time sequence multi-scale feature vector and an ORP value time sequence multi-scale feature vector. The multi-parameter timing feature cross-correlation analysis subunit 124 reveals the timing relationships and interactions between them by analyzing and correlating multi-scale features of oxygen content and ORP values, resulting in a more comprehensive and detailed representation of the features. The combination and functionality of these subunits can help to analyze and correlate the time series data of oxygen content and ORP values, thereby extracting time series interaction characteristics of oxygen content-ORP values, and providing useful information for subsequent data analysis and decision-making.

More specifically, in the parametric multi-scale temporal feature extraction subunit 123, the deep neural network model is a multi-scale neighborhood feature extraction module, and the multi-scale neighborhood feature extraction module includes a first convolution layer and a second convolution layer in parallel, and a multi-scale feature fusion layer connected to the first convolution layer and the second convolution layer, where the first convolution layer and the second convolution layer use one-dimensional convolution kernels having different scales.

More specifically, the multi-parameter timing feature cross-correlation analysis subunit 124 is configured to: and performing feature interaction based on an attention mechanism on the oxygen content time sequence multi-scale feature vector and the ORP value time sequence multi-scale feature vector by using an inter-feature attention layer to obtain the oxygen content-ORP value time sequence interaction feature vector. It is worth mentioning that the attention mechanism is a mechanism for assigning a different weight or attention weight to each input, given a set of inputs, in order to weight aggregate or focus the inputs. It can help the model automatically learn the degree of attention to different inputs as it processes sequence or aggregate data, thereby better capturing the relationships and importance between inputs. In the multiparameter timing feature interaction correlation analysis subunit 124, the inter-feature attention layer is used to perform attention-mechanism-based feature interactions on the oxygen content timing multiscale feature vector and the ORP value timing multiscale feature vector. This means that the model will learn to assign attention weights to different features in order to pay more attention to important features for the relation between oxygen content and ORP values during feature interaction. Through the attention mechanism, the model can automatically learn the correlation and dependency between input features. Thus, time sequence interaction information between the oxygen content and the ORP value can be better captured, and modeling capacity of the model on the relationship between the oxygen content and the ORP value is improved. The attention weight can be dynamically adjusted according to the input context, so that the model can be better adapted to different input conditions and task requirements. In other words, the attention mechanism can help the model pay attention to important features better in the feature interaction process, and the expressive power and performance of the model are improved.

And then, the oxygen content-ORP value time sequence interaction characteristic vector is passed through a classifier to obtain a classification result, wherein the classification result is used for indicating that the power value of the oxygenation fan at the current time point is increased or decreased. That is, the real-time self-adaptive control of the power value of the oxygenation fan is performed by utilizing the interactive correlation characteristic information between the time sequence change characteristic of the oxygen content and the time sequence change characteristic of the OPR value, and the opening value of the electric valve is controlled in real time to control the oxygen supply and the aeration effect, so that the growth environment of microorganisms is optimized, the treatment and the automatic regulation of the high-fluorine high-ammonia nitrogen wastewater are realized, and the high-fluorine high-ammonia nitrogen wastewater treatment efficiency and the high economical efficiency are realized.

Accordingly, the oxygenation fan control unit 130 is configured to: and passing the oxygen content-ORP value time sequence interaction feature vector through a classifier to obtain a classification result, wherein the classification result is used for indicating that the power value of the oxygenation fan at the current time point is increased or decreased.

More specifically, the oxygenation fan control unit 130 is further configured to: performing full-connection coding on the oxygen content-ORP value time sequence interaction feature vector by using a full-connection layer of the classifier to obtain a coding classification feature vector; and inputting the coding classification feature vector into a Softmax classification function of the classifier to obtain the classification result.

That is, in the technical solution of the present disclosure, the labels of the classifier include that the power value of the oxygenation fan at the current time point should be increased (first label), and that the power value of the oxygenation fan at the current time point should be decreased (second label), wherein the classifier determines to which classification label the oxygen content-ORP value time sequence interaction feature vector belongs through a soft maximum function. It should be noted that the first tag p1 and the second tag p2 do not include the concept of artificial setting, and in fact, during the training process, the computer model does not have the concept of "the power value of the oxygenation fan at the current time point should be increased or decreased", which is simply that there are two kinds of classification tags and the probability that the output characteristic is the sum of the two classification tags sign, that is, p1 and p2 is one. Therefore, the classification result that the power value of the oxygenation fan at the current time point should be increased or decreased is actually converted into the classified probability distribution conforming to the two classifications of the natural law through the classification label, and the physical meaning of the natural probability distribution of the label is essentially used instead of the language text meaning of "the power value of the oxygenation fan at the current time point should be increased or decreased".

It should be appreciated that the role of the classifier is to learn the classification rules and classifier using a given class, known training data, and then classify (or predict) the unknown data. Logistic regression (logistics), SVM, etc. are commonly used to solve the classification problem, and for multi-classification problems (multi-class classification), logistic regression or SVM can be used as well, but multiple bi-classifications are required to compose multiple classifications, but this is error-prone and inefficient, and the commonly used multi-classification method is the Softmax classification function.

Further, in the technical scheme of the application, the device for treating the high-fluorine high-ammonia nitrogen wastewater further comprises a time sequence feature extractor based on the multi-scale neighborhood feature extraction module, the inter-feature attention layer and the classifier.

More specifically, the training module includes: the training data acquisition unit is used for acquiring training data, wherein the training data comprise training oxygen content and training ORP values of the mixed solution at a plurality of preset time points in a preset time period, and a true value that the power of the oxygenation fan should be increased or decreased; the training data time sequence arrangement unit is used for arranging the training oxygen content and the training ORP values of the plurality of preset time points into a training oxygen content time sequence input vector and a training ORP value time sequence input vector according to the time dimension; the training data time sequence up-sampling unit is used for up-sampling the training oxygen content time sequence input vector and the training ORP value time sequence input vector respectively to obtain a training up-sampling oxygen content time sequence input vector and a training up-sampling ORP value time sequence input vector; the training parameter time sequence multi-scale feature extraction unit is used for enabling the training up-sampling oxygen content time sequence input vector and the training up-sampling ORP value time sequence input vector to pass through the time sequence feature extractor based on the multi-scale neighborhood feature extraction module respectively so as to obtain a training oxygen content time sequence multi-scale feature vector and a training ORP value time sequence multi-scale feature vector; the training parameter time sequence feature interaction association unit is used for carrying out feature interaction based on an attention mechanism on the training oxygen content time sequence multi-scale feature vector and the training ORP value time sequence multi-scale feature vector by using the inter-feature attention layer so as to obtain a training oxygen content-ORP value time sequence interaction feature vector; the classification loss unit is used for enabling the training oxygen content-ORP value time sequence interaction feature vector to pass through the classifier to obtain a classification loss function value; and a model training unit, configured to train the time sequence feature extractor, the inter-feature attention layer and the classifier based on the multi-scale neighborhood feature extraction module based on the classification loss function value and through propagation in a gradient descent direction, where, when each weight matrix iteration of the training is performed, weight space exploration constraint optimization based on regularization of a class matrix is performed on the training oxygen content-ORP value time sequence interaction feature vector.

In particular, in the technical scheme of the application, the training oxygen content time series multiscale feature vector and the training ORP value time series multiscale feature vector express the time series multiscale neighborhood associated features of the training oxygen content and the training ORP value respectively, so that after the feature interaction based on an attention mechanism is carried out on the training oxygen content time series multiscale feature vector and the training ORP value time series multiscale feature vector by using an attention layer, the training oxygen content-ORP value time series interaction feature vector comprises the time series multiscale neighborhood associated features of the training oxygen content and the training ORP value, and also comprises the dependency relationship features of the training oxygen content and the training ORP value in the time series distribution direction, namely, the training oxygen content-ORP value time series interaction feature vector simultaneously comprises the feature representation of a diversified time series associated dimension corresponding to the time series association in a sample and the time series association between samples, and the time series associated feature representation of the time series associated feature vector of the training oxygen content-ORP value under the cross dimension can also cause the cross-correlation feature vector of the oxygen content-ORP value to affect the cross-classification effect of the cross-classification of the feature vector in the cross-domain classifier when the cross-domain feature distribution is carried out in the classifier.

Based on the above, when classifying the training oxygen content-ORP value time sequence interaction feature vector by a classifier, the applicant performs weight space exploration constraint based on regularization of a class matrix on the training oxygen content-ORP value time sequence interaction feature vector at each iteration of the weight matrix.

Accordingly, in a specific example, at each weight matrix iteration of the training, performing weight space exploration constraint optimization based on class matrix regularization on the training oxygen content-ORP value time sequence interaction feature vector by using the following optimization formula to obtain an optimized training oxygen content-ORP value time sequence interaction feature vector; wherein, the optimization formula is:

Wherein V is the training oxygen content-ORP value time sequence interaction feature vector, V ' is the optimized training oxygen content-ORP value time sequence interaction feature vector, V is a column vector, V ' is a row vector, M _t∈R^L×L is a learnable domain transfer matrix, such as a diagonal matrix formed by diagonal elements of weight matrix M that may be initially set as the last iteration, M represents the weight matrix of the last iteration, M ' represents the weight matrix after the iteration, Representing matrix multiplication, V' represents the optimal training oxygen content-ORP value timing interaction feature vector.

Here, considering the domain difference (domain gap) between the weight space domain of the weight matrix and the probability distribution domain of the classification result of the training oxygen content-ORP value time sequence interaction feature vector V, the regularized representation of the class matrix of the weight matrix M relative to the training oxygen content-ORP value time sequence interaction feature vector V is used as an inter-domain migration agent (inter-domain TRANSFERRING AGENT) to transfer the probability distribution of the valuable label constraint into the weight space, so that the excessive exploration (over-explat) of the weight distribution in the weight space by the rich label (rich labeled) probability distribution domain in the classification process based on the weight space is avoided, the convergence effect of the weight matrix is improved, and the training effect of the training oxygen content-ORP value time sequence interaction feature vector in the classification regression through the classifier is also improved. Therefore, the metabolic state and activity level of microorganisms in the wastewater can be known in real time by monitoring the dissolved oxygen and ORP values, and the oxygen supply and aeration effect can be controlled by adaptively adjusting the power of a fan and the opening of an electric valve, so that the growth environment of the microorganisms is optimized, the treatment and automatic adjustment of the high-fluorine high-ammonia nitrogen wastewater are realized, and the high-fluorine high-ammonia nitrogen wastewater treatment device has high treatment efficiency and economy.

In summary, the device for treating the high-fluorine high-ammonia-nitrogen wastewater, which is based on the embodiment of the application, is illustrated, and can realize the treatment and automatic adjustment of the high-fluorine high-ammonia-nitrogen wastewater.

As described above, the apparatus for treating high-fluorine and high-ammonia-nitrogen wastewater according to the embodiment of the present application may be implemented in various terminal devices, for example, a server or the like having an algorithm for treating high-fluorine and high-ammonia-nitrogen wastewater according to the embodiment of the present application. In one example, the apparatus for treating high-fluorine high-ammonia nitrogen wastewater according to the embodiment of the present application may be integrated into a terminal device as one software module and/or hardware module. For example, the apparatus for treating high-fluorine high-ammonia nitrogen wastewater according to the embodiment of the present application may be a software module in an operating system of the terminal device, or may be an application program developed for the terminal device; of course, the device for treating high-fluorine high-ammonia nitrogen wastewater according to the embodiment of the application can be one of a plurality of hardware modules of the terminal equipment.

Alternatively, in another example, the apparatus for treating high fluorine and high ammonia nitrogen wastewater may be connected to the terminal device through a wired and/or wireless network and transmit the interactive information in a agreed data format.

FIG. 3 is a flow chart of a method for treating high fluorine and high ammonia nitrogen wastewater in accordance with an embodiment of the application. Fig. 4 is a schematic diagram of a system architecture of a method for treating high fluorine and high ammonia nitrogen wastewater according to an embodiment of the application. As shown in fig. 3 and 4, a method for treating wastewater with high fluorine and high ammonia nitrogen according to an embodiment of the present application includes: s110, collecting oxygen content and ORP values of the mixed solution at a plurality of preset time points in a preset time period through a dissolved oxygen and ORP detector arranged in the high-efficiency biological carrier reactor; s120, performing time sequence feature interaction correlation analysis on the oxygen content and ORP values at a plurality of preset time points to obtain time sequence interaction features of the oxygen content-ORP values; and S130, determining that the power value of the oxygenation fan at the current time point is increased or decreased based on the oxygen content-ORP value time sequence interaction characteristic.

Here, it will be understood by those skilled in the art that the specific operations of the respective steps in the above-described method for treating high-fluorine high-ammonia-nitrogen wastewater have been described in detail in the above description of the apparatus for treating high-fluorine high-ammonia-nitrogen wastewater with reference to fig. 1 to 2, and thus, repetitive descriptions thereof will be omitted.

Fig. 5 is an application scenario diagram of an apparatus for treating high-fluorine high-ammonia nitrogen wastewater according to an embodiment of the present application. As shown in fig. 5, in this application scenario, first, the oxygen content (e.g., D1 illustrated in fig. 5) and the ORP value (e.g., D2 illustrated in fig. 5) of a plurality of predetermined time points of the mixed solution within a predetermined period of time are collected by the dissolved oxygen and ORP detector disposed in the high-efficiency bio-carrier reactor, and then the oxygen content and ORP value of the plurality of predetermined time points are input to the server (e.g., S illustrated in fig. 5) disposed with the algorithm for treating the high-fluorine high-ammonia nitrogen wastewater, wherein the server is capable of processing the oxygen content and ORP value of the plurality of predetermined time points using the algorithm for treating the high-fluorine high-ammonia nitrogen wastewater to obtain a classification result indicating that the power value of the oxygenation fan at the current time point should be increased or decreased.

It is worth mentioning that, in order to solve the existing defluorination and nitrogen removal process with high costs, the disadvantage that the treatment efficiency is not high enough, the application provides a system that can remove high fluorine and high ammonia nitrogen waste water steadily with high efficiency. The method is characterized in that the water quality and the water quantity are comprehensively collected in respective collecting tanks through fluorine-containing wastewater, and the water quality and the water quantity are homogenized and uniform. The effluent is pumped into a homogenizing mixing tank. The fluoride ion removing unit is a wastewater treatment and defluorination process based on the combination of chemical agent coagulation and precipitation and electrochemistry, and is characterized in that the chemical agent reduces high-concentration fluoride ions in water into low-concentration fluoride ions (less than 100 mg/L) and performs precipitation separation on flocs and water. Adopting a three-stage reactant materialization method defluorination system, controlling the pH value in a first-stage reaction tank to be within a range of 6-7, adding lime milk and CaCl ₂, and continuously adding PAC and PAM in a subsequent two-stage reaction tank; after precipitation, the effluent enters a precipitation tank to complete mud-water separation, then an electrochemical device is adopted, after metal cations are blended, residual fluoride ions are trapped and adsorbed by utilizing the high activity of nascent metal ions, and the effluent enters a secondary precipitation tank to perform mud-water separation again and clarify; then adopting an electrochemical treatment process to reduce the concentration of fluoride ions to below 8mg/L, and carrying out precipitation separation on the flocs and water. Continuous water inflow is adopted, water is lifted to a mixing and stirring pool by using a water pump, the stirring pool is three-stage, and lime water, calcium chloride, polyaluminium chloride and polyacrylamide are sequentially added by stirring through a mechanical paddle stirrer. The adding amount of polyaluminum chloride is 0.1%, the adding amount of PAM is 0.001%, the concentration of lime water is between 0.5% and 2%, and the concentration of CaCl ₂ solution is 20g to 30g per 100g of water, so that the fluorine ions with high concentration in the water are reduced to low concentration (less than 100 mg/L). After the stirring reaction is completed, the wastewater flows into a high-efficiency materialized sedimentation tank, and the separation of sludge and clean water is completed in the tank. Wherein the surface load is from 1.0m3/h to 2.0m3/h. The effluent of the sedimentation tank is lifted to an electrochemical treatment device by a water pump, the voltage of the device is changed to 15V, the current density is 160A/m < 2 >, and the residence time is 10min. The active metal ions generated by the electrochemical device can react with fluorine ions in water, the fluorine ions are trapped and produced into flocculent precipitate, and the flocculent precipitate is separated from water in a subsequent precipitation unit.

After the wastewater subjected to electrochemical treatment is mixed with domestic sewage and high ammonia nitrogen wastewater, the biochemical treatment section consists of an anoxic zone, an aerobic zone and a sedimentation tank, and the aerobic zone and the anoxic zone both adopt high-efficiency biological carrier reactors. The microbial community on the high-efficiency biological carrier in the ammonia removal unit can adsorb COD and ammonia nitrogen in the wastewater. Microorganisms on the surface layer can hydrolyze organic matters to change macromolecules into small molecules, gradually oxidize the macromolecules into carbon dioxide and water, and oxidize ammonium ions into nitrate and nitrite. Microorganisms in the inner layer can further oxidize organic matters transferred from the surface layer and reduce nitrate and nitrite into nitrogen. In addition, raw water animals and metazoans on the biological carrier can also be provided with bacteria and fungi to form longer food chains, thereby reducing the generation of sludge.

The high-efficiency biological carrier reactor is provided with dissolved oxygen and ORP detection instruments, an electric proportional valve, an oxygenation fan and an aerator, and can be automatically adjusted; the automatic control component automatically adjusts the frequency of the fan and the opening degree of the electric valve according to the dissolved oxygen and the ORP.

The dissolved oxygen of the nitrogen removal unit is controlled to be 0.5 to 2mg/L, and the whole biological carrier is in a circulating fluidization state. The COD treatment load of the high-efficiency biological carrier reactor can reach 2kgCOD/m3/d. The nitrogen treatment load of the high-efficiency biological carrier reactor can reach 0.6kgN/m3/d. The residence time of the sewage in the high-efficiency biological carrier is 12-48 h, water is continuously fed and discharged in the mode, and a plurality of reactors can be arranged in series according to the concentration of pollutants, so that the effluent can reach the discharge standard. The sewage treated by the high-efficiency biological carrier reactor needs to be further separated into mud and water in a secondary sedimentation tank. The secondary sedimentation tank can select a radial sedimentation tank or a vertical sedimentation tank according to the water quantity.

It should be understood that the sewage treatment method is a sewage treatment method with low energy consumption and high efficiency for biological denitrification, under the action of the method, the high-abundance polysaccharide bacteria biological film is domesticated, and under the action of the high-abundance polysaccharide bacteria biological film, the ammonia nitrogen adsorbing filler can repeatedly adsorb ammonia nitrogen pollutants in sewage; the biological membrane based on the high-abundance polysaccharide bacteria can maximally utilize BOD in the sewage for denitrification, so that a higher total nitrogen removal rate is obtained in the treatment of the sewage with low carbon nitrogen ratio. The sewage treatment process disclosed by the application has the advantages of small occupied area of required treatment equipment, small energy consumption, low operation cost and simple post-maintenance. Correspondingly, the high-fluorine high-ammonia nitrogen wastewater treatment device has the following advantages: the method is designed aiming at the characteristics of high-fluorine high-ammonia nitrogen wastewater, so that pollutants in the wastewater can be effectively removed, and the treatment efficiency is improved; the defluorination unit has less sludge production amount and medicament use amount, the nitrogen removal unit has higher removal rate and less carbon source addition amount; the treatment residence time of each unit can be selected according to actual conditions, so that the wastewater treatment requirements of different types and scales can be met; simple structure, convenient operation possess higher stability and reliability.

The inherent good mixing and mass transfer characteristics of fluidization make this technology advantageous in both domestic and industrial wastewater treatment processes. Both laboratory and pilot plant stage studies have demonstrated that more than 80% of the carbon source and more than 80% of the nitrogen source in the wastewater can be removed and that the sludge yield is less than 1/3 of the activated sludge process. Due to the high efficiency of the technology, the technology has the advantage of small occupied area. Meanwhile, the improved process dynamic load test (the change of water inflow and water inflow concentration) also shows good impact resistance and recovery performance.

Further, in another embodiment of the application, the process flow of the high-fluorine high-ammonia nitrogen wastewater treatment device is shown in fig. 6. The implementation process uses a pilot scale to simplify the simulation. The primary sedimentation reaction tank is simulated by adopting a triple stirrer, and the primary sedimentation tank beaker is simulated by adopting natural sedimentation for 30min. The electrochemical reaction tank is simulated by using a 240 x 65 x 150mm organic glass device, a polar plate is made of stainless steel and an aluminum plate, the polar plate area is 15cm x 15cm, the reaction residence time is 5 to 10min, and finally natural sedimentation is carried out for 30min. The biological carrier reactor adopts a 5L plastic beaker, the specific surface area of the high-efficiency biological carrier is more than 600m2/m3, the specific gravity is 0.94 to 0.97, the surface is specially treated, and the biological film has extremely strong adhesive capability and is made of PE. The three-sedimentation tank is simulated by adopting a beaker for natural sedimentation for 30min, the final reaction tank is formed by adopting a triple stirrer, and the sedimentation process is performed by adopting the beaker for natural sedimentation for 30min.

The activated sludge is taken from an aerobic aeration tank of a municipal sewage treatment plant, the fluorine-containing wastewater adopts the wastewater of a photovoltaic production line, and the concentration of fluorine ions is properly regulated to a specified concentration. The nitrogen-containing wastewater was configured in a laboratory. The quality of the prepared inflow water is shown in tables 1 and 2, and glucose is used as a COD index and a carbon source. The pH value is kept between 7 and 8 in the using process, the pH value in the experimental process of the example 1 is adjusted by sulfuric acid and liquid alkali, the pH value in the process is adjusted by sodium dihydrogen phosphate and sodium carbonate, and sodium acetate is used as a carbon source for supplementing in the process. The nitrogen-containing wastewater is configured using ammonium chloride and glucose.

TABLE 1 quality of influent wastewater from fluorine-containing wastewater

Sequence number	Index (I)	Unit (B)	Limit value
				1	pH	---	7-8
2	CODcr	mg/L	500±50mg/L
				3	F-	mg/L	3000±50mg/L

TABLE 2 quality of influent wastewater containing nitrogen

The method for calculating the removal rate comprises the following steps:

Example 1:

(1) In this example, the chemical coagulation agent was calcium hydroxide to prepare a 40g/L suspension. Adding the waste liquid with different concentrations according to the specified proportion. Quick stirring is carried out for 5 minutes, then 0.05% polyaluminum chloride and 0.01% PAM solution are added, slow stirring is carried out for 1 minute, natural sedimentation is carried out for 30 minutes, and the supernatant is taken for measurement.

(2) In this example, wastewater after chemical coagulation was diluted to a F ion concentration of 100mg/L. And (3) putting a certain amount of the solution into an electrochemical oxidation device, starting a power supply, adjusting the solution to a specified voltage, recording voltage and current, sampling once every 1min, and measuring the concentration of F ions.

Wherein, table 3 shows the pollutant concentrations after the sewage treatment of different concentrations in the chemical coagulation process of example 1, table 4 shows the pollutant removal rates after the sewage treatment of different concentrations in the chemical coagulation process of example 1, table 5 shows the pollutant concentrations after the sewage treatment of different residence times in the electrochemical process of example 1 under condition 1, and table 6 shows the pollutant concentrations after the sewage treatment of different residence times in the electrochemical process of example 1 under condition 2.

Table 3: example 1 concentration of pollutants after treatment of wastewater of different concentrations in chemical coagulation

Table 4: example 1 pollutant removal Rate after treatment of wastewater of different concentrations in chemical coagulation Process

Table 5: example 1 electrochemical Process contaminant concentration after wastewater treatment at different residence times under condition 1

Table 6: example 1 electrochemical Process contaminant concentration after wastewater treatment at different residence times under condition 2

10V2A	F-concentration mg/L	F-removal rate
			0min	100	0％
1min	91	9％
			2min	80	11％
3min	69	11％
			4min	57	12％
5min	43	14％
			6min	33	10％
7min	25	8％
			8min	17	8％
9min	9	8％
			10min	5	4％

Accordingly, fig. 7 shows the change of the concentration of the pollutants after the treatment of the sewage with different concentrations in the chemical coagulation process of example 1; FIG. 8 shows the change in contaminant removal rate after treatment of wastewater of different concentrations during chemical coagulation in example 1; FIG. 9 shows the change in contaminant concentration after wastewater treatment at 15V-3A for different residence times in the electrochemical process of example 1; FIG. 10 shows the change in contaminant concentration after wastewater treatment at 10V-2A for different residence times in the electrochemical process of example 1.

Example 2:

In the embodiment, the prepared wastewater is added into and extracted from the high-efficiency biological carrier reactor according to the specified flow, and the primary and the secondary are used in series. The flow rate of the primary reaction tank is set according to the residence time of 48h, and the flow rate of the secondary reaction tank is set according to the residence time of 24 h. And (3) after mud-water separation at the tail end, returning the sludge to the reaction tank, and sampling and measuring supernatant after coagulating sedimentation.

Accordingly, table 7 shows the removal amount and removal rate of COD in the sewage treatment cycle of example 2; table 8 shows the ammonia nitrogen removal amount and removal rate for the wastewater treatment cycle of example 2; table 9 shows the ammonia nitrogen removal and removal rate for the wastewater treatment cycle of example 2.

Table 7: COD removal amount and removal rate in the wastewater treatment cycle of example 2

Table 8: ammonia nitrogen removal amount and ammonia nitrogen removal rate in sewage treatment period of example 2

Table 9: ammonia nitrogen removal amount and ammonia nitrogen removal rate in sewage treatment period of example 2

Accordingly, fig. 11 shows the COD removal amount and removal rate change in the sewage treatment cycle of example 2; FIG. 12 shows the ammonia nitrogen removal rate and the change in the removal rate during the wastewater treatment cycle of example 2; fig. 13 shows the total nitrogen removal and removal rate variation of the sewage treatment cycle of example 2.

Furthermore, those skilled in the art will appreciate that the various aspects of the application are illustrated and described in the context of a number of patentable categories or circumstances, including any novel and useful procedures, machines, products, or materials, or any novel and useful modifications thereof. Accordingly, aspects of the application may be performed entirely by hardware, entirely by software (including firmware, resident software, micro-code, etc.) or by a combination of hardware and software. The above hardware or software may be referred to as a "data block," module, "" engine, "" unit, "" component, "or" system. Furthermore, aspects of the application may take the form of a computer product, comprising computer-readable program code, embodied in one or more computer-readable media.

Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.

The foregoing is illustrative of the present invention and is not to be construed as limiting thereof. Although a few exemplary embodiments of this invention have been described, those skilled in the art will readily appreciate that many modifications are possible in the exemplary embodiments without materially departing from the novel teachings and advantages of this invention. Accordingly, all such modifications are intended to be included within the scope of this invention as defined in the following claims. It is to be understood that the foregoing is illustrative of the present invention and is not to be construed as limited to the specific embodiments disclosed, and that modifications to the disclosed embodiments, as well as other embodiments, are intended to be included within the scope of the appended claims. The invention is defined by the claims and their equivalents.

Claims

1. A method for treating high fluorine and high ammonia nitrogen wastewater, comprising:

Performing time sequence feature interaction correlation analysis on the oxygen content and ORP values at a plurality of preset time points to obtain time sequence interaction features of the oxygen content-ORP values; determining whether the power value of the oxygenation fan at the current time point is increased or decreased based on the oxygen content-ORP value time sequence interaction characteristic;

The method further comprises a training step for training a time sequence feature extractor, an inter-feature attention layer and a classifier based on the multi-scale neighborhood feature extraction module;

The training step comprises the following steps:

Acquiring training data, wherein the training data comprises training oxygen content and training ORP values of a mixed solution at a plurality of preset time points in a preset time period, and a true value that the power of the oxygenation fan should be increased or decreased;

respectively arranging the training oxygen content and the training ORP values of the plurality of preset time points into a training oxygen content time sequence input vector and a training ORP value time sequence input vector according to a time dimension;

respectively carrying out up-sampling processing on the training oxygen content time sequence input vector and the training ORP value time sequence input vector to obtain a training up-sampling oxygen content time sequence input vector and a training up-sampling ORP value time sequence input vector;

respectively passing the training up-sampling oxygen content time sequence input vector and the training up-sampling ORP value time sequence input vector through a time sequence feature extractor based on a multi-scale neighborhood feature extraction module to obtain a training oxygen content time sequence multi-scale feature vector and a training ORP value time sequence multi-scale feature vector;

Performing feature interaction based on an attention mechanism on the training oxygen content time sequence multi-scale feature vector and the training ORP value time sequence multi-scale feature vector by using the inter-feature attention layer to obtain a training oxygen content-ORP value time sequence interaction feature vector;

Passing the training oxygen content-ORP value time sequence interaction feature vector through the classifier to obtain a classification loss function value; training a time sequence feature extractor based on a multi-scale neighborhood feature extraction module, the inter-feature attention layer and the classifier based on the classification loss function value and through gradient descent direction propagation, wherein weight space exploration constraint optimization based on class matrix regularization is carried out on the training oxygen content-ORP value time sequence interaction feature vector when each weight matrix iteration of the training is carried out;

when the weight matrix of the training is iterated each time, carrying out weight space exploration constraint optimization on the training oxygen content-ORP value time sequence interaction feature vector based on class matrix regularization by using the following optimization formula to obtain an optimized training oxygen content-ORP value time sequence interaction feature vector;

Wherein, the optimization formula is:

wherein, Is the training oxygen content-ORP value time sequence interaction characteristic vector,Is the time sequence interaction characteristic vector of the optimized training oxygen content-ORP value, andAs a column vector of the column-wise vector,Is the vector of the row and,In order to be a domain transfer matrix that can be learned,Representing the weight matrix of the last iteration,Representing the weight matrix after the iteration,Representing a matrix multiplication of the number of bits,Representing the optimal training oxygen content-ORP value time sequence interaction characteristic vector.

2. The method for treating high-fluorine high-ammonia-nitrogen wastewater according to claim 1, wherein performing a time series characteristic cross-correlation analysis on the oxygen content and the ORP value at the plurality of predetermined time points to obtain an oxygen content-ORP value time series cross-correlation, comprises:

Arranging the oxygen content and the ORP values of the plurality of preset time points into an oxygen content time sequence input vector and an ORP value time sequence input vector according to a time dimension respectively;

Respectively carrying out up-sampling treatment on the oxygen content time sequence input vector and the ORP value time sequence input vector to obtain an up-sampling oxygen content time sequence input vector and an up-sampling ORP value time sequence input vector;

Performing multi-scale time sequence feature extraction on the up-sampling oxygen content time sequence input vector and the up-sampling ORP value time sequence input vector through a time sequence feature extractor based on a deep neural network model to obtain an oxygen content time sequence multi-scale feature vector and an ORP value time sequence multi-scale feature vector; and performing feature interaction correlation analysis on the oxygen content time sequence multi-scale feature vector and the ORP value time sequence multi-scale feature vector to obtain an oxygen content-ORP value time sequence interaction feature vector as the oxygen content-ORP value time sequence interaction feature.

3. The method for treating high-fluorine high-ammonia-nitrogen wastewater according to claim 2, wherein the deep neural network model is a multi-scale neighborhood feature extraction module comprising a first convolution layer and a second convolution layer in parallel, and a multi-scale feature fusion layer connected with the first convolution layer and the second convolution layer, wherein the first convolution layer and the second convolution layer use one-dimensional convolution kernels with different scales.

4. The method for treating high-fluorine high-ammonia nitrogen wastewater according to claim 3, wherein performing feature cross correlation analysis on the oxygen content time series multi-scale feature vector and the ORP value time series multi-scale feature vector to obtain an oxygen content-ORP value time series cross feature vector as the oxygen content-ORP value time series cross feature, comprises:

And performing feature interaction based on an attention mechanism on the oxygen content time sequence multi-scale feature vector and the ORP value time sequence multi-scale feature vector by using an inter-feature attention layer to obtain the oxygen content-ORP value time sequence interaction feature vector.

5. The method for treating high fluorine and high ammonia nitrogen wastewater of claim 4, wherein determining the power value of the oxygenation fan at the current point in time should be increased or decreased based on the oxygen content-ORP value timing interaction characteristics, comprising:

and passing the oxygen content-ORP value time sequence interaction feature vector through a classifier to obtain a classification result, wherein the classification result is used for indicating that the power value of the oxygenation fan at the current time point is increased or decreased.