CN117970987B - Intelligent control system and method for wet desulfurization - Google Patents

Abstract

The application discloses an intelligent control system and method for wet desulfurization, which inputs raw material gas containing hydrogen sulfide into a desulfurization tower, wherein the desulfurization tower uses desulfurization lean solution to carry out desulfurization treatment on the raw material gas containing hydrogen sulfide so as to obtain purified product gas, elemental sulfur and desulfurization rich solution; wherein the inflow rate of the raw material gas is adaptively controlled and adjusted based on the real-time temperature and the real-time pressure of the desulfurization lean solution. In this way, the response correlation between the inflow flow rate time sequence characteristic of the raw material gas and the reaction progress time sequence characteristic can be used as feedback information to guide the self-adaptive control and adjustment of the inflow flow rate of the raw material gas, so as to ensure that the desulfurization reaction is carried out in an optimal state.

Description

Intelligent control system and method for wet desulfurization

Technical Field

The application relates to the technical field of intelligent desulfurization control, in particular to an intelligent control system and method for wet desulfurization.

Background

Sulfur (Sulfur) is a nonmetallic element. Natural gas often contains hydrogen sulfide which can lead to corrosion of the natural gas pipelines and equipment, thereby shortening the service life of the equipment, increasing maintenance costs, and even causing equipment failure. In addition, hydrogen sulfide has a pungent odor and can explode at high concentrations. If the natural gas contains a large amount of hydrogen sulfide, once leakage or accident occurs, serious consequences such as fire and explosion can be caused. For subsequent section applications and safe pipeline transportation, reducing atmospheric pollution, it is generally necessary to desulphurize natural gas.

In the prior art, desulfurization treatment of natural gas is generally classified into wet desulfurization, dry desulfurization and semi-dry desulfurization. Among them, wet desulfurization is generally performed using a solution containing an absorbent, and the desulfurization efficiency is high. However, conventional wet desulfurization also has some problems, such as high reliance on manual control. For example, operators need to adjust the addition of the absorbent according to the concentration of hydrogen sulfide in the natural gas or acid gas and other operating parameters to ensure effective desulfurization. This manual control method may lead to inconsistent control results due to subjectivity, and thus unstable desulfurization effects. Therefore, an intelligent control system and method for wet desulfurization are desired.

Disclosure of Invention

The application provides an intelligent control system and method for wet desulfurization, which inputs raw material gas containing hydrogen sulfide into a desulfurization tower, wherein the desulfurization tower uses desulfurization lean solution to carry out desulfurization treatment on the raw material gas containing hydrogen sulfide so as to obtain purified product gas, elemental sulfur and desulfurization rich solution; wherein the inflow rate of the raw material gas is adaptively controlled and adjusted based on the real-time temperature and the real-time pressure of the desulfurization lean solution. In this way, the response correlation between the inflow flow rate time sequence characteristic of the raw material gas and the reaction progress time sequence characteristic can be used as feedback information to guide the self-adaptive control and adjustment of the inflow flow rate of the raw material gas, so as to ensure that the desulfurization reaction is carried out in an optimal state.

The application also provides an intelligent control method for wet desulfurization, which comprises the following steps: inputting a raw material gas containing hydrogen sulfide into a desulfurization tower, wherein the desulfurization tower uses a desulfurization lean solution to carry out desulfurization treatment on the raw material gas containing hydrogen sulfide to obtain purified product gas, elemental sulfur and desulfurization rich solution; wherein the inflow rate of the raw material gas is adaptively controlled and adjusted based on the real-time temperature and the real-time pressure of the desulfurization lean solution.

In the wet desulfurization intelligent control method, the desulfurization lean solution is a complex molten iron solution.

In the above-described wet desulfurization intelligent control method, the inflow rate of the raw material gas is adaptively controlled and adjusted based on the real-time temperature and the real-time pressure of the desulfurization lean solution, including: acquiring a time sequence of the inlet flow velocity of the raw material gas acquired by the sensor group and a time sequence of the real-time temperature and the real-time pressure of the desulfurization lean solution; performing correlation analysis on the time sequence of the real-time temperature and the real-time pressure of the desulfurization lean solution to obtain a temperature-pressure time sequence correlation characteristic vector; extracting time sequence characteristics of the time sequence of the inflow flow velocity of the raw material gas to obtain an inflow flow velocity time sequence related characteristic vector; an incoming flow control instruction is determined based on characteristic interaction response correlation information between the incoming flow timing correlation characteristic vector and the temperature-pressure timing correlation characteristic vector.

In the above-mentioned intelligent control method for wet desulfurization, performing a correlation analysis on a time series of a real-time temperature and a real-time pressure of the desulfurization lean solution to obtain a temperature-pressure time-series correlation feature vector, including: data normalization is carried out on the time sequence of the real-time temperature and the real-time pressure of the desulfurization lean solution according to the time dimension respectively to obtain a real-time temperature time sequence input vector of the desulfurization lean solution and a real-time pressure time sequence input vector of the desulfurization lean solution; and inputting the real-time temperature time sequence input vector of the desulfurization lean solution and the real-time pressure time sequence input vector of the desulfurization lean solution into a temperature-pressure correlation feature extractor based on a factoring machine to obtain the temperature-pressure time sequence correlation feature vector.

In the above-mentioned intelligent control method for wet desulfurization, inputting the real-time temperature time sequence input vector of the desulfurization lean solution and the real-time pressure time sequence input vector of the desulfurization lean solution into a temperature-pressure correlation feature extractor based on a factorizer to obtain the temperature-pressure time sequence correlation feature vector, comprising: inputting the real-time temperature time sequence input vector of the desulfurization lean solution and the real-time pressure time sequence input vector of the desulfurization lean solution into a temperature-pressure correlation feature extractor based on a factoring machine in the following factoring formula to obtain the temperature-pressure time sequence correlation feature vector; wherein, the factorization formula is: ; wherein/> For the/>, in the temperature-pressure timing-related feature vectorCharacteristic value/>Inputting the first/> in the vector for the real-time temperature time sequence of the desulfurization lean solutionCharacteristic value/>Inputting the/> in the vector for the real-time pressure time sequence of the desulfurization lean solutionCharacteristic value/>Inputting vector dimension for real-time pressure time sequence of the desulfurization lean solution,/>Is a constant bias coefficient,/>For/>First factorization bias coefficient,/>For/>And a second factoring bias factor.

In the above-mentioned intelligent control method for wet desulfurization, performing time sequence feature extraction on the time sequence of the inflow flow rate of the raw material gas to obtain an inflow flow rate time sequence correlation feature vector, including: the time sequence of the inflow velocity of the raw material gas is regulated according to the time dimension to obtain an inflow velocity time sequence input vector; inputting the input flow time sequence input vector into a flow time sequence correlation mode feature extractor based on one-dimensional expansion convolution to obtain the input flow time sequence correlation feature vector.

In the above-mentioned intelligent control method for wet desulfurization, inputting the inflow velocity time sequence input vector into a velocity time sequence correlation pattern feature extractor based on one-dimensional expansion convolution to obtain the inflow velocity time sequence correlation feature vector, including: processing the incoming flow time sequence input vector by using the following one-dimensional expansion convolution formula to obtain the incoming flow time sequence association characteristic vector; wherein the one-dimensional extended convolution formula is: ; wherein one-dimensional extended convolution kernel/> Is/>，/>For the length of the original convolution kernel,/>For the expansion rate/>Representing the/>, in the input vector, at the input flow rate timingThe length of the characteristic value of each position is/>Time window of/>Is a bias term, and/>，/>Nonlinear activation function,/>Is/>Local convolution encoded eigenvectors,/>For the dimension of the incoming flow timing input vector,/>Is the incoming flow timing related feature vector,/>Representing a cascade.

In the above-described wet desulfurization intelligent control method, determining an inflow rate control instruction based on characteristic interaction response correlation information between the inflow rate timing correlation feature vector and the temperature-pressure timing correlation feature vector includes: inputting the input flow time sequence association characteristic vector and the temperature-pressure time sequence association characteristic vector into a characteristic interaction response association analysis network to obtain an adaptive control posterior time sequence characteristic vector; performing feature distribution correction on the self-adaptive control posterior time sequence feature vector to obtain a corrected self-adaptive control posterior time sequence feature vector; and passing the corrected self-adaptive control posterior time sequence characteristic vector through an inflow flow controller based on a classifier to obtain the inflow flow control instruction, wherein the inflow flow control instruction is used for indicating that the inflow flow rate at the current time point should be increased, decreased or kept unchanged.

In the above-mentioned intelligent control method for wet desulfurization, inputting the inflow flow rate time-series related feature vector and the temperature-pressure time-series related feature vector into a feature interaction response related analysis network to obtain an adaptive control posterior time-series feature vector, comprising: cascading the incoming flow time sequence related feature vector and the temperature-pressure time sequence related feature vector along a channel dimension to obtain a channel cascading feature vector; information concentration is carried out on the channel cascade feature vector by using pooling and activation processing so as to obtain a concentrated channel information characterization feature vector; the concentrated channel information representation feature vector passes through a first full-connection layer and a second full-connection layer respectively to obtain a first channel information representation coding feature vector and a second channel information representation coding feature vector, wherein the first full-connection layer and the second full-connection layer have the same node number; and multiplying the first channel information representation coding feature vector and the second channel information representation coding feature vector with the access flow time sequence association feature vector and the temperature-pressure time sequence association feature vector respectively by using broadcast multiplication, and then adding multiplication results to obtain the self-adaptive control posterior time sequence feature vector.

The application also provides an intelligent control system for wet desulfurization, which comprises the following steps: the desulfurization treatment module is used for inputting the raw material gas containing the hydrogen sulfide into a desulfurization tower, wherein the desulfurization tower uses a desulfurization lean solution to carry out desulfurization treatment on the raw material gas containing the hydrogen sulfide so as to obtain purified product gas, elemental sulfur and desulfurization rich solution; an adaptive control and adjustment module for adaptively controlling and adjusting the inflow rate of the raw material gas based on the real-time temperature and the real-time pressure of the desulfurization lean solution; wherein the adaptive control and adjustment module comprises: acquiring a time sequence of the inlet flow velocity of the raw material gas acquired by the sensor group and a time sequence of the real-time temperature and the real-time pressure of the desulfurization lean solution; performing correlation analysis on the time sequence of the real-time temperature and the real-time pressure of the desulfurization lean solution to obtain a temperature-pressure time sequence correlation characteristic vector; extracting time sequence characteristics of the time sequence of the inflow flow velocity of the raw material gas to obtain an inflow flow velocity time sequence related characteristic vector; an incoming flow control instruction is determined based on characteristic interaction response correlation information between the incoming flow timing correlation characteristic vector and the temperature-pressure timing correlation characteristic vector.

Compared with the prior art, the intelligent control system and the intelligent control method for wet desulfurization are characterized in that raw material gas containing hydrogen sulfide is input into a desulfurization tower, wherein the desulfurization tower uses desulfurization lean solution to carry out desulfurization treatment on the raw material gas containing hydrogen sulfide to obtain purified product gas, elemental sulfur and desulfurization rich solution; wherein the inflow rate of the raw material gas is adaptively controlled and adjusted based on the real-time temperature and the real-time pressure of the desulfurization lean solution. In this way, the response correlation between the inflow flow rate time sequence characteristic of the raw material gas and the reaction progress time sequence characteristic can be used as feedback information to guide the self-adaptive control and adjustment of the inflow flow rate of the raw material gas, so as to ensure that the desulfurization reaction is carried out in an optimal state.

Drawings

In order to more clearly illustrate the embodiments of the application or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

In the drawings: fig. 1 is a flowchart of an intelligent control method for wet desulfurization provided in an embodiment of the present application.

Fig. 2 is a flowchart of a sub-step of step S120 in the intelligent control method for wet desulfurization according to an embodiment of the present application.

Fig. 3 is a block diagram of an intelligent control system for wet desulfurization according to an embodiment of the present application.

Fig. 4 is a process flow diagram of wet oxidation desulfurization and sulfur recovery provided in an embodiment of the present application.

Fig. 5 is a block diagram of an electronic device, according to an example embodiment.

Fig. 6 is an application scenario diagram of an intelligent control method for wet desulfurization provided in an embodiment of the present application.

Wherein, 1, a desulfurizing tower; 2. a rich liquid pump; 3. a lean liquid pump; 4. a regeneration tank; 5. a buffer tank; 6. a filtrate tank; 7. a filtrate pump; 8. A filter press; 9. a blower; 10. a separator.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the embodiments of the present application will be described in further detail with reference to the accompanying drawings. The exemplary embodiments of the present application and their descriptions herein are for the purpose of explaining the present application, but are not to be construed as limiting the application.

Unless defined otherwise, all technical and scientific terms used in the embodiments of the application have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the present application is for the purpose of describing particular embodiments only and is not intended to limit the scope of the present application.

In describing embodiments of the present application, unless otherwise indicated and limited thereto, the term "connected" should be construed broadly, for example, it may be an electrical connection, or may be a communication between two elements, or may be a direct connection, or may be an indirect connection via an intermediate medium, and it will be understood by those skilled in the art that the specific meaning of the term may be interpreted according to circumstances.

It should be noted that, the term "first\second\third" related to the embodiment of the present application is merely to distinguish similar objects, and does not represent a specific order for the objects, it is to be understood that "first\second\third" may interchange a specific order or sequence where allowed. It is to be understood that the "first\second\third" distinguishing objects may be interchanged where appropriate such that embodiments of the application described herein may be practiced in sequences other than those illustrated or described herein.

Sulfur (Sulfur) is a nonmetallic element, chemical symbol S, atomic number 16. Elemental sulfur is typically yellow crystals, also known as sulfur. The allotropes of the sulfur simple substance are various, and include orthorhombic sulfur, monoclinic sulfur, elastic sulfur and the like. Elemental sulfur is commonly present in nature in the form of sulfides, sulfates, or simple substances. Elemental sulfur is poorly soluble in water, slightly soluble in ethanol, and readily soluble in carbon disulfide.

The primary uses of sulfur are for the manufacture of sulfur compounds such as sulfuric acid, sulfites, thiosulfates, thiocyanates, sulfur dioxide, carbon disulfide, phosphorus oxychloride, sulfides of phosphorus, metal sulfides, and the like. The sulfur consumed for producing sulfuric acid every year in the world accounts for more than 80% of the total annual consumption of sulfur. Sulfur is also used in large quantities for the production of vulcanized rubber. Sulfur is also required for the manufacture of black powder and matches, and is one of the main raw materials for producing pyrotechnics, and sulfur can also be used for producing sulfur dyes and pigments. The bleaching industry and the pharmaceutical industry also consume a portion of the sulfur.

The natural gas of the oil field contains hydrogen sulfide or organic sulfur, and the natural gas is generally required to be removed for the application and pipeline transportation of the subsequent working section to reduce the atmospheric pollution. The method and the system mainly remove the hydrogen sulfide in the hydrogen sulfide-containing gas and oxidize the hydrogen sulfide into elemental sulfur to prepare sulfur paste or sulfur cake, wherein the general sulfur content is 67%, the water content is 27%, and the dust and other impurities content is 6%, so that the method and the system have certain utilization value in the fields of ceramics, building materials and farmlands, and can also further prepare the sulfur paste into refined sulfur products.

In one embodiment of the present application, fig. 1 is a flowchart of a method for intelligently controlling wet desulfurization provided in the embodiment of the present application. As shown in fig. 1, the intelligent control method for wet desulfurization according to the embodiment of the application comprises the following steps: s110, inputting a raw material gas containing hydrogen sulfide into a desulfurizing tower, wherein the desulfurizing tower uses a desulfurizing lean solution to carry out desulfurizing treatment on the raw material gas containing hydrogen sulfide so as to obtain a purified product gas, a sulfur simple substance and a desulfurizing rich solution; wherein, S120, the inflow flow rate of the raw material gas is adaptively controlled and adjusted based on the real-time temperature and the real-time pressure of the desulfurization lean solution.

In a specific example of the present application, the desulfurization lean solution is a complex molten iron solution.

In this process, it is important to control the flow rate of the raw material gas. This is because the inflow rate directly affects the desulfurization efficiency and the contact time of the raw material gas with the desulfurization lean liquid in the reaction tower. Specifically, if the feed gas inflow rate is too fast, the contact time of the feed gas with the desulfurization lean solution will be reduced, which may result in insufficient reaction of hydrogen sulfide with the desulfurization lean solution, thereby affecting the quality of the purified product gas. Meanwhile, too fast inflow speed can also cause pressure increase in the desulfurizing tower, and the risk of equipment operation is increased. Conversely, if the inflow rate is too slow, although the contact time of the gas and liquid may be increased, the processing capacity and efficiency of the desulfurization system may be reduced, and the concentration of the dissolved oxidizing agent in the desulfurization lean solution may be reduced, affecting the generation of elemental sulfur. Therefore, it is particularly important to reasonably control the inflow rate of the raw material gas. However, the manner of manual control may lead to inconsistent control results due to subjectivity, and thus to unstable desulfurization effects.

In consideration of the reaction progress and desulfurization effect of the hydrogen sulfide-containing raw material gas and the desulfurization lean solution in the process of inputting the hydrogen sulfide-containing raw material gas into the desulfurization tower, the reaction progress and desulfurization effect can be taken as important basis for guiding the inflow flow rate of the raw material gas. The reaction progress and the desulfurization effect of the raw material gas containing the hydrogen sulfide and the desulfurization lean solution are reflected in the real-time temperature and the real-time pressure data of the desulfurization lean solution. That is, the integrated timing information of the real-time temperature and real-time pressure data of the desulfurization lean solution can characterize the reaction progress and desulfurization effect of the hydrogen sulfide-containing raw material gas and the desulfurization lean solution. Thus, in the technical idea of the present application, it is expected that the self-adaptive control and adjustment of the inflow flow rate of the raw material gas is guided by using the responsive correlation between the inflow flow rate timing characteristic of the raw material gas and the reaction progress timing characteristic as feedback information, thereby ensuring that the desulfurization reaction is performed in an optimal state.

Fig. 2 is a flowchart of a sub-step of step S120 in the intelligent control method for wet desulfurization according to an embodiment of the present application. As shown in fig. 2, S120, the inflow rate of the raw material gas is adaptively controlled and adjusted based on the real-time temperature and the real-time pressure of the desulfurization lean solution, including: s121, acquiring a time sequence of the inlet flow velocity of the raw material gas acquired by a sensor group and a time sequence of the real-time temperature and the real-time pressure of the desulfurization lean solution; s122, performing correlation analysis on the time sequence of the real-time temperature and the real-time pressure of the desulfurization lean solution to obtain a temperature-pressure time sequence correlation feature vector; s123, carrying out time sequence feature extraction on the time sequence of the inflow velocity of the raw material gas to obtain an inflow velocity time sequence associated feature vector; s124, determining an incoming flow control instruction based on characteristic interaction response association information between the incoming flow time sequence association characteristic vector and the temperature-pressure time sequence association characteristic vector.

Based on the above, in the technical scheme of the application, the specific steps of adaptively controlling and adjusting the inflow flow rate of the raw material gas based on the real-time temperature and the real-time pressure of the desulfurization lean solution comprise the following steps: first, a time series of the inflow flow rate of the raw material gas collected by the sensor group and a time series of the real-time temperature and the real-time pressure of the desulfurization lean solution are acquired. Here, the real-time monitoring of the operation state of the desulfurization system and the desulfurization reaction can be realized by collecting the time series of the inflow flow rate of the raw material gas, the real-time temperature and the real-time pressure of the desulfurization lean solution by the sensor group.

In one embodiment of the present application, performing a correlation analysis on a time series of real-time temperature and real-time pressure of the desulfurization lean solution to obtain a temperature-pressure time-series correlation feature vector, includes: data normalization is carried out on the time sequence of the real-time temperature and the real-time pressure of the desulfurization lean solution according to the time dimension respectively to obtain a real-time temperature time sequence input vector of the desulfurization lean solution and a real-time pressure time sequence input vector of the desulfurization lean solution; and inputting the real-time temperature time sequence input vector of the desulfurization lean solution and the real-time pressure time sequence input vector of the desulfurization lean solution into a temperature-pressure correlation feature extractor based on a factoring machine to obtain the temperature-pressure time sequence correlation feature vector.

And then, respectively carrying out data normalization on the time sequence of the real-time temperature and the real-time pressure of the desulfurization lean solution according to the time dimension to obtain a real-time temperature time sequence input vector of the desulfurization lean solution and a real-time pressure time sequence input vector of the desulfurization lean solution. In this way, the discrete distribution of temperature data and pressure data is converted into a structured vector representation in a data-structured manner according to the time dimension, and the time data is retained while further analysis and processing of the subsequent model are facilitated. Then, the real-time temperature time sequence input vector of the desulfurization lean solution and the real-time pressure time sequence input vector of the desulfurization lean solution are input into a temperature-pressure correlation feature extractor based on a factoring machine to obtain a temperature-pressure time sequence correlation feature vector. Wherein the factorer (Factorization Machines) model captures complex relationships between features, as compared to a common linear model, which typically builds the relationship between features by simple weighted summation only. Specifically, the factorizer model introduces deep association mining capability among feature components on the basis of linear regression, and complex association among features can be better captured by modeling interaction relationship among the features, so that expressive force and generalization capability of the model are improved. The factorizer model can effectively learn the higher-order interaction relation between the features by introducing hidden factors, so that the factorizer model is better suitable for the complexity of data. That is, the factorizer model has an advantage in that interactions between features can be modeled while maintaining linear complexity, thereby exhibiting excellent performance in processing high-dimensional sparse data. This makes the factoring machine a powerful tool when processing data sets with complex feature interactions. In the technical scheme of the application, a temperature-pressure correlation characteristic extractor based on a factoring machine is utilized to effectively model and capture a deep complex time sequence hiding correlation mode and a dynamic rule between time sequence temperature data and time sequence pressure data of the desulfurization lean solution.

In one embodiment of the present application, inputting the desulfation lean solution real-time temperature time sequence input vector and the desulfation lean solution real-time pressure time sequence input vector into a temperature-pressure correlation feature extractor based on a factorizer to obtain the temperature-pressure time sequence correlation feature vector comprises: inputting the real-time temperature time sequence input vector of the desulfurization lean solution and the real-time pressure time sequence input vector of the desulfurization lean solution into a temperature-pressure correlation feature extractor based on a factoring machine in the following factoring formula to obtain the temperature-pressure time sequence correlation feature vector; wherein, the factorization formula is: ; wherein/> For the/>, in the temperature-pressure timing-related feature vectorCharacteristic value/>Inputting the first/> in the vector for the real-time temperature time sequence of the desulfurization lean solutionCharacteristic value/>Inputting the/> in the vector for the real-time pressure time sequence of the desulfurization lean solutionCharacteristic value/>Inputting vector dimension for real-time pressure time sequence of the desulfurization lean solution,/>Is a constant bias coefficient,/>For/>First factorization bias coefficient,/>For/>And a second factoring bias factor.

In a specific embodiment of the present application, the time series of the inflow velocity of the raw material gas is subjected to time series feature extraction to obtain an inflow velocity time series associated feature vector, which includes: the time sequence of the inflow velocity of the raw material gas is regulated according to the time dimension to obtain an inflow velocity time sequence input vector; inputting the input flow time sequence input vector into a flow time sequence correlation mode feature extractor based on one-dimensional expansion convolution to obtain the input flow time sequence correlation feature vector.

At the same time, the time sequence of the inflow velocity of the raw material gas is regulated according to the time dimension to obtain an inflow velocity time sequence input vector; and inputting the input flow time sequence input vector into a flow time sequence correlation mode feature extractor based on one-dimensional expansion convolution to obtain an input flow time sequence correlation feature vector. Here, the incoming flow rate of the raw gas is likewise subjected to data normalization processing to convert the discretely distributed incoming flow rate data into a vector representation to meet the input requirements of the flow rate time-series correlation pattern feature extractor based on one-dimensional expansion convolution. Among them, one-dimensional spread convolution is a convolution operation in a Convolutional Neural Network (CNN), which is generally used to process one-dimensional data, such as data with time information or signal data. Unlike conventional two-dimensional convolution, one-dimensional convolution operation performs sliding window convolution computation in one dimension, and can effectively capture local patterns and features in sequence data. Compared with the common one-dimensional convolution, the one-dimensional expansion convolution introduces a super-parameter named expansion rate to control the number of the hollow values 0 in the convolution kernel. The expansion rate defines the receptive field size of the convolution kernel when processing one-dimensional data, so that the adaptive time sequence neighborhood receptive field can be adjusted according to actual application scenes, implicit data modes under different spans are captured, and the structure and the characteristics of the data are better understood. In this way, the flow velocity time sequence correlation mode feature extractor based on the one-dimensional expansion convolution is used for feature extraction, so that the time sequence implicit distribution mode of the inflow flow velocity of the raw material gas under a larger receptive field can be captured, and the dynamic change of the inflow flow velocity can be understood.

In a specific embodiment of the present application, inputting the incoming flow timing input vector into a flow timing correlation pattern feature extractor based on one-dimensional spread convolution to obtain the incoming flow timing correlation feature vector includes: processing the incoming flow time sequence input vector by using the following one-dimensional expansion convolution formula to obtain the incoming flow time sequence association characteristic vector; wherein the one-dimensional extended convolution formula is: ; wherein one-dimensional extended convolution kernel/> Is/>，/>For the length of the original convolution kernel,/>For the expansion rate/>Representing the/>, in the input vector, at the input flow rate timingThe length of the characteristic value of each position is/>Time window of/>Is a bias term, and/>，/>Nonlinear activation function,/>Is/>Local convolution encoded eigenvectors,/>For the dimension of the incoming flow timing input vector,/>Is the incoming flow timing related feature vector,/>Representing a cascade.

And then inputting the incoming flow time sequence correlation characteristic vector and the temperature-pressure time sequence correlation characteristic vector into a characteristic interaction response correlation analysis network to obtain an adaptive control posterior time sequence characteristic vector. The characteristic interaction response correlation analysis network is used for describing and describing dynamic changes and interaction relations between the temperature-pressure time sequence correlation mode distribution of the desulfurization lean solution and the inflow flow time sequence implicit correlation characteristic distribution of the raw material gas, so that the overall state of the desulfurization system is comprehensively understood. In the encoding process of the characteristic interaction response association analysis network, unlike the traditional method of directly and simply adding characteristic graphs, the characteristic interaction response association analysis network distributes weights by integrating global information of the incoming flow time sequence association characteristic vector and the temperature-pressure time sequence association characteristic vector in a channel dimension, so that the network can select and strengthen important characteristic information according to the input characteristic distribution information of the incoming flow time sequence association characteristic vector and the temperature-pressure time sequence association characteristic vector, redundancy and invalid characteristics are reduced, and the association interaction response relationship between the two is fully expressed. And then, the self-adaptive control posterior time sequence characteristic vector passes through an inlet flow rate controller based on a classifier to obtain an inlet flow rate control instruction, wherein the inlet flow rate control instruction is used for indicating that the inlet flow rate at the current time point should be increased, decreased or kept unchanged.

In one particular embodiment of the application, determining an incoming flow control instruction based on characteristic interaction response correlation information between the incoming flow timing correlation characteristic vector and the temperature-pressure timing correlation characteristic vector comprises: inputting the input flow time sequence association characteristic vector and the temperature-pressure time sequence association characteristic vector into a characteristic interaction response association analysis network to obtain an adaptive control posterior time sequence characteristic vector; performing feature distribution correction on the self-adaptive control posterior time sequence feature vector to obtain a corrected self-adaptive control posterior time sequence feature vector; and passing the corrected self-adaptive control posterior time sequence characteristic vector through an inflow flow controller based on a classifier to obtain the inflow flow control instruction, wherein the inflow flow control instruction is used for indicating that the inflow flow rate at the current time point should be increased, decreased or kept unchanged.

Further, inputting the incoming flow rate timing related feature vector and the temperature-pressure timing related feature vector into a feature interaction response correlation analysis network to obtain an adaptive control posterior timing feature vector, comprising: cascading the incoming flow time sequence related feature vector and the temperature-pressure time sequence related feature vector along a channel dimension to obtain a channel cascading feature vector; information concentration is carried out on the channel cascade feature vector by using pooling and activation processing so as to obtain a concentrated channel information characterization feature vector; the concentrated channel information representation feature vector passes through a first full-connection layer and a second full-connection layer respectively to obtain a first channel information representation coding feature vector and a second channel information representation coding feature vector, wherein the first full-connection layer and the second full-connection layer have the same node number; and multiplying the first channel information representation coding feature vector and the second channel information representation coding feature vector with the access flow time sequence association feature vector and the temperature-pressure time sequence association feature vector respectively by using broadcast multiplication, and then adding multiplication results to obtain the self-adaptive control posterior time sequence feature vector.

In the technical scheme of the application, the inflow time sequence related feature vector and the temperature-pressure time sequence related feature vector respectively express local time sequence related features of the inflow flow rate of the raw material gas and time sequence related features of the real-time temperature and the real-time pressure of the desulfurization lean solution, so that the inflow time sequence related feature vector and the temperature-pressure time sequence related feature vector also have obvious time sequence feature distribution differences in consideration of source time sequence distribution differences and corresponding feature extraction dimension differences of the inflow flow rate of the raw material gas and the real-time temperature and the real-time pressure of the desulfurization lean solution, and therefore, when the inflow time sequence related feature vector and the temperature-pressure time sequence related feature vector are input into a feature interaction response related analysis network, the mapping effect of the inflow time sequence related feature vector and the temperature-pressure time sequence related feature vector to feature distribution domains after feature interaction response related fusion is expected to be improved.

Accordingly, the applicant of the present application further performs an optimized fusion of the inflow timing-related feature vector and the temperature-pressure timing-related feature vector, specifically expressed as: carrying out optimized fusion on the input flow time sequence related characteristic vector and the temperature-pressure time sequence related characteristic vector by using the following optimized formula to obtain a corrected characteristic vector; wherein, the optimization formula is: ; wherein/> Is the incoming flow timing related feature vector,/>Is the temperature-pressure time sequence associated eigenvector, eigenvector/>And/>Having the same length,/>And/>Feature vector/>, respectivelyMean and standard deviation of corresponding feature sets,/>And/>Feature vector/>, respectivelyMean and standard deviation of corresponding feature sets,/>Representing the position-by-position evolution of the feature vector, and/>As a logarithmic function with 2 as the base,/>Is a correction feature vector,/>Representing per-position addition,/>Representing the multiplication by position; and fusing the corrected characteristic vector with the self-adaptive control posterior time sequence characteristic vector to obtain a corrected self-adaptive control posterior time sequence characteristic vector.

Here, in order to promote the mapping effect of the incoming flow time sequence related feature vector and the temperature-pressure time sequence related feature vector to the fusion feature distribution domain under the feature fusion scene, on the basis that the traditional weighted fusion mode has limitation on deducing the semantic space evolution diffusion mode based on feature superposition, the fusion effect of the incoming flow time sequence related feature vector and the temperature-pressure time sequence related feature vector is promoted by adopting a mode of combining a low-order superposition fusion mode and a high-order superposition fusion mode of the space and simulating the evolution center and the evolution track through the statistical feature interaction relation of the incoming flow time sequence related feature vector and the temperature-pressure time sequence related feature vector so as to reconstruct the semantic space evolution diffusion under the fusion scene based on asynchronous evolution under the action of different evolution diffusion speed fields. Thus, the correction feature vector for optimizing fusion is usedAnd the self-adaptive control posterior time sequence feature vector fusion is carried out, so that the feature interaction response association fusion expression effect of the self-adaptive control posterior time sequence feature vector is improved, and the accuracy of a classification result obtained by the self-adaptive control posterior time sequence feature vector through a classifier is improved.

In summary, the charging management method of the charging pile according to the embodiment of the present application is explained, and is expected to utilize the responsive correlation between the inflow flow rate timing characteristic of the raw material gas and the reaction progress timing characteristic as feedback information to guide the adaptive control and adjustment of the inflow flow rate of the raw material gas, thereby ensuring that the desulfurization reaction is performed in an optimal state.

Fig. 3 is a block diagram of an intelligent control system for wet desulfurization according to an embodiment of the present application. As shown in fig. 3, the intelligent control system 200 for wet desulfurization includes: a desulfurization treatment module 210, configured to input a raw gas containing hydrogen sulfide into a desulfurization tower, where the desulfurization tower uses a desulfurization lean solution to perform desulfurization treatment on the raw gas containing hydrogen sulfide to obtain a purified product gas, elemental sulfur, and a desulfurization rich solution; an adaptive control and adjustment module 220 for adaptively controlling and adjusting the incoming flow rate of the feed gas based on the real-time temperature and real-time pressure of the sweet lean solution.

In one embodiment of the present application, the adaptive control and adjustment module includes: acquiring a time sequence of the inlet flow velocity of the raw material gas acquired by the sensor group and a time sequence of the real-time temperature and the real-time pressure of the desulfurization lean solution; performing correlation analysis on the time sequence of the real-time temperature and the real-time pressure of the desulfurization lean solution to obtain a temperature-pressure time sequence correlation characteristic vector; extracting time sequence characteristics of the time sequence of the inflow flow velocity of the raw material gas to obtain an inflow flow velocity time sequence related characteristic vector; an incoming flow control instruction is determined based on characteristic interaction response correlation information between the incoming flow timing correlation characteristic vector and the temperature-pressure timing correlation characteristic vector.

It will be appreciated by those skilled in the art that the specific operations of the respective steps in the above-described wet desulfurization intelligent control system have been described in detail in the above description of the wet desulfurization intelligent control method with reference to fig. 1 to 2, and thus, repetitive descriptions thereof will be omitted.

As described above, the wet desulfurization intelligent control system 200 according to the embodiment of the present application can be implemented in various terminal devices, such as a server or the like for wet desulfurization intelligent control. In one example, the wet desulfurization intelligent control system 200 according to an embodiment of the present application can be integrated into the terminal device as one software module and/or hardware module. For example, the wet desulfurization intelligent control system 200 can be a software module in the operating system of the terminal device, or can be an application developed for the terminal device; of course, the intelligent control system for wet desulfurization 200 can also be one of a plurality of hardware modules of the terminal device.

Alternatively, in another example, the wet desulfurization intelligent control system 200 and the terminal device may be separate devices, and the wet desulfurization intelligent control system 200 may be connected to the terminal device through a wired and/or wireless network and transmit interactive information in a agreed data format.

In one embodiment, the application relates to a method and a system for desulfurization and sulfur recovery by a wet oxidation method, in particular to a method for oxidizing and recovering hydrogen sulfide generated in a gas containing hydrogen sulfide into elemental sulfur by a wet oxidation method sulfur recovery system consisting of a desulfurizing tower, a lean solution pump, a rich solution pump, a regeneration tank, a filter press and the like.

The wet oxidation desulfurization and sulfur recovery system of the present application comprises: a desulfurizing tower 1; a rich liquid pump 2; a lean liquid pump 3; a regeneration tank 4; a buffer tank 5; a filtrate tank 6; a filtrate pump 7; a filter press 8; a fan 9; a separator 10; and connecting lines and associated auxiliary valves, etc.

The method of the application aims at the hydrogen sulfide-containing gas to remove hydrogen sulfide and oxidize the hydrogen sulfide into elemental sulfur. The gas containing hydrogen sulfide (such as natural gas containing hydrogen sulfide or acid gas containing hydrogen sulfide) enters the bottom of the desulfurizing tower to be in countercurrent contact with the desulfurizing lean liquid sprayed from the tower top, and the gas containing hydrogen sulfide removes hydrogen sulfide to become purified product gas to be out of the boundary. The desulfurization lean solution oxidizes the hydrogen sulfide in the hydrogen sulfide-containing gas into elemental sulfur, the elemental sulfur is reduced into desulfurization rich solution, the elemental sulfur enters the desulfurization rich solution to form suspension, and the suspension is discharged from the bottom of the tower and conveyed to a regeneration tank through a rich solution pump. The regeneration tank is introduced into a fan to convey oxygen in the air to oxidize the rich liquid into lean liquid, and the lean liquid and the suspension formed by elemental sulfur enter a buffer tank together. Pumping the suspension in the buffer tank into a filter press through a filtrate pump, filtering and pressing the sulfur simple substance into a sulfur cake product, returning the remained desulfurization lean solution to the filtrate tank for buffering, and circulating the remained desulfurization lean solution into a desulfurization tower through a lean solution pump for desulfurization.

The process flow diagram of the present application is shown in fig. 4. The application realizes the removal and oxidation of the hydrogen sulfide in the hydrogen sulfide-containing gas into the sulfur simple substance product, greatly improves the comprehensive utilization of the hydrogen sulfide-containing natural gas, has low investment and good economic benefit compared with the traditional Claus sulfur recovery device, avoids the pollution of the direct utilization or discharge of the hydrogen sulfide-containing gas to the environment, and has important significance for protecting the natural environment. The application has simple process and strong operability.

Referring now to fig. 5, a schematic diagram of an electronic device 600 suitable for use in implementing embodiments of the present disclosure is shown. The terminal devices in the embodiments of the present disclosure may include, but are not limited to, mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and the like, and stationary terminals such as digital TVs, desktop computers, and the like. The electronic device shown in fig. 5 is merely an example and should not be construed to limit the functionality and scope of use of the disclosed embodiments.

As shown in fig. 5, the electronic device 600 may include a processing means (e.g., a central processing unit, a graphics processor, etc.) 601, which may perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM) 602 or a program loaded from a storage means 608 into a Random Access Memory (RAM) 603. In the RAM603, various programs and data required for the operation of the electronic apparatus 600 are also stored. The processing device 601, the ROM602, and the RAM603 are connected to each other through a bus 604. An input/output (I/O) interface 605 is also connected to bus 604.

In general, the following devices may be connected to the I/O interface 605: input devices 606 including, for example, a touch screen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, and the like; an output device 607 including, for example, a Liquid Crystal Display (LCD), a speaker, a vibrator, and the like; storage 608 including, for example, magnetic tape, hard disk, etc.; and a communication device 609. The communication means 609 may allow the electronic device 600 to communicate with other devices wirelessly or by wire to exchange data. While fig. 5 shows an electronic device 600 having various means, it is to be understood that not all of the illustrated means are required to be implemented or provided. More or fewer devices may be implemented or provided instead.

In particular, according to embodiments of the present disclosure, the processes described above with reference to flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a non-transitory computer readable medium, the computer program comprising program code for performing the method shown in the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network via communication means 609, or from storage means 608, or from ROM 602. The above-described functions defined in the methods of the embodiments of the present disclosure are performed when the computer program is executed by the processing device 601.

It should be noted that the computer readable medium described in the present disclosure may be a computer readable signal medium or a computer readable storage medium, or any combination of the two. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this disclosure, a computer-readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In the present disclosure, however, the computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, with the computer-readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, fiber optic cables, RF (radio frequency), and the like, or any suitable combination of the foregoing.

In some embodiments, the clients, servers may communicate using any currently known or future developed network protocol, such as HTTP (HyperText TransferProtocol ), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), the internet (e.g., the internet), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future developed networks.

The computer readable medium may be contained in the electronic device; or may exist alone without being incorporated into the electronic device.

Computer program code for carrying out operations of the present disclosure may be written in one or more programming languages, including, but not limited to, an object oriented programming language such as Java, smalltalk, C ++ and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the Internet using an Internet service provider).

The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The modules described in the embodiments of the present disclosure may be implemented in software or hardware. The name of the module is not limited to the module itself in some cases, and for example, the test parameter obtaining module may also be described as "a module for obtaining the device test parameter corresponding to the target device".

The functions described above herein may be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: a Field Programmable Gate Array (FPGA), an Application Specific Integrated Circuit (ASIC), an Application Specific Standard Product (ASSP), a system on a chip (SOC), a Complex Programmable Logic Device (CPLD), and the like.

In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

Fig. 6 is an application scenario diagram of an intelligent control method for wet desulfurization provided in an embodiment of the present application. As shown in fig. 6, in this application scenario, first, a time series of the inflow flow rate of the raw material gas collected by the sensor group and a time series of the real-time temperature (e.g., C1 as illustrated in fig. 6) and the real-time pressure (e.g., C2 as illustrated in fig. 6) of the desulfurization lean solution are acquired; the obtained real-time temperature and real-time pressure are then input into a server (e.g., S as illustrated in fig. 6) deployed with a wet-desulfurization intelligent control algorithm, wherein the server is capable of processing the real-time temperature and the real-time pressure based on the wet-desulfurization intelligent control algorithm to determine an incoming flow control command based on characteristic interaction response correlation information between the incoming flow timing correlation feature vector and the temperature-pressure timing correlation feature vector.

The foregoing description of the embodiments has been provided for the purpose of illustrating the general principles of the application, and is not meant to limit the scope of the application, but to limit the application to the particular embodiments, and any modifications, equivalents, improvements, etc. that fall within the spirit and principles of the application are intended to be included within the scope of the application.

Claims (5)

1. An intelligent control method for wet desulfurization is characterized by comprising the following steps:

Inputting a raw material gas containing hydrogen sulfide into a desulfurization tower, wherein the desulfurization tower uses a desulfurization lean solution to carry out desulfurization treatment on the raw material gas containing hydrogen sulfide to obtain purified product gas, elemental sulfur and desulfurization rich solution;

wherein the inflow flow rate of the raw material gas is adaptively controlled and adjusted based on the real-time temperature and the real-time pressure of the desulfurization lean solution, comprising:

Acquiring a time sequence of the inlet flow velocity of the raw material gas acquired by the sensor group and a time sequence of the real-time temperature and the real-time pressure of the desulfurization lean solution;

Performing correlation analysis on the time sequence of the real-time temperature and the real-time pressure of the desulfurization lean solution to obtain a temperature-pressure time sequence correlation characteristic vector;

extracting time sequence characteristics of the time sequence of the inflow flow velocity of the raw material gas to obtain an inflow flow velocity time sequence related characteristic vector;

determining an incoming flow control instruction based on feature interaction response correlation information between the incoming flow timing correlation feature vector and the temperature-pressure timing correlation feature vector;

performing correlation analysis on a time sequence of real-time temperature and real-time pressure of the desulfurization lean solution to obtain a temperature-pressure time sequence correlation feature vector, wherein the method comprises the following steps of:

Data normalization is carried out on the time sequence of the real-time temperature and the real-time pressure of the desulfurization lean solution according to the time dimension respectively to obtain a real-time temperature time sequence input vector of the desulfurization lean solution and a real-time pressure time sequence input vector of the desulfurization lean solution;

inputting the real-time temperature time sequence input vector of the desulfurization lean solution and the real-time pressure time sequence input vector of the desulfurization lean solution into a temperature-pressure correlation feature extractor based on a factor decomposition machine to obtain a temperature-pressure time sequence correlation feature vector;

Wherein inputting the desulfation lean solution real-time temperature time sequence input vector and the desulfation lean solution real-time pressure time sequence input vector into a temperature-pressure correlation feature extractor based on a factoring machine to obtain the temperature-pressure time sequence correlation feature vector comprises: inputting the real-time temperature time sequence input vector of the desulfurization lean solution and the real-time pressure time sequence input vector of the desulfurization lean solution into a temperature-pressure correlation feature extractor based on a factoring machine in the following factoring formula to obtain the temperature-pressure time sequence correlation feature vector;

Wherein, the factorization formula is:

，

Wherein, For the/>, in the temperature-pressure timing-related feature vectorCharacteristic value/>Inputting the first/> in the vector for the real-time temperature time sequence of the desulfurization lean solutionCharacteristic value/>Inputting the/> in the vector for the real-time pressure time sequence of the desulfurization lean solutionCharacteristic value/>Inputting vector dimension for real-time pressure time sequence of the desulfurization lean solution,/>Is a constant bias coefficient,/>Is the firstFirst factorization bias coefficient,/>For/>A second factorization bias factor;

Wherein determining an incoming flow control instruction based on characteristic interaction response correlation information between the incoming flow timing correlation characteristic vector and the temperature-pressure timing correlation characteristic vector comprises:

inputting the input flow time sequence association characteristic vector and the temperature-pressure time sequence association characteristic vector into a characteristic interaction response association analysis network to obtain an adaptive control posterior time sequence characteristic vector;

Performing feature distribution correction on the self-adaptive control posterior time sequence feature vector to obtain a corrected self-adaptive control posterior time sequence feature vector;

The corrected self-adaptive control posterior time sequence feature vector passes through an inflow flow controller based on a classifier to obtain an inflow flow control instruction, wherein the inflow flow control instruction is used for indicating that the inflow flow rate at the current time point should be increased, decreased or kept unchanged;

wherein inputting the incoming flow rate timing related feature vector and the temperature-pressure timing related feature vector into a feature interaction response related analysis network to obtain an adaptive control posterior timing feature vector, comprising:

Cascading the incoming flow time sequence related feature vector and the temperature-pressure time sequence related feature vector along a channel dimension to obtain a channel cascading feature vector;

Information concentration is carried out on the channel cascade feature vector by using pooling and activation processing so as to obtain a concentrated channel information characterization feature vector;

The concentrated channel information representation feature vector passes through a first full-connection layer and a second full-connection layer respectively to obtain a first channel information representation coding feature vector and a second channel information representation coding feature vector, wherein the first full-connection layer and the second full-connection layer have the same node number;

And multiplying the first channel information representation coding feature vector and the second channel information representation coding feature vector with the access flow time sequence association feature vector and the temperature-pressure time sequence association feature vector respectively by using broadcast multiplication, and then adding multiplication results to obtain the self-adaptive control posterior time sequence feature vector.

2. The intelligent control method for wet desulfurization according to claim 1, wherein the desulfurization lean solution is a complex molten iron solution.

3. The intelligent control method for wet desulfurization according to claim 2, characterized by performing time series feature extraction on the time series of the inflow flow rate of the raw material gas to obtain an inflow flow rate time series-related feature vector, comprising:

The time sequence of the inflow velocity of the raw material gas is regulated according to the time dimension to obtain an inflow velocity time sequence input vector;

Inputting the input flow time sequence input vector into a flow time sequence correlation mode feature extractor based on one-dimensional expansion convolution to obtain the input flow time sequence correlation feature vector.

4. The method according to claim 3, wherein inputting the inflow timing input vector into a flow timing correlation pattern feature extractor based on one-dimensional spread convolution to obtain the inflow timing correlation feature vector comprises:

Processing the incoming flow time sequence input vector by using the following one-dimensional expansion convolution formula to obtain the incoming flow time sequence association characteristic vector; wherein the one-dimensional extended convolution formula is:

，

wherein the one-dimensional extended convolution kernel Is/>，/>For the length of the original convolution kernel,/>For the expansion rate/>Representing the/>, in the input vector, at the input flow rate timingThe length of the characteristic value of each position is/>Time window of/>Is a bias term, and/>，/>Nonlinear activation function,/>Is/>Local convolution encoded eigenvectors,/>For the dimension of the incoming flow timing input vector,/>Is the incoming flow timing related feature vector,/>Representing a cascade.

5. An intelligent control system for wet desulfurization using the intelligent control method for wet desulfurization according to claim 1, comprising:

The desulfurization treatment module is used for inputting the raw material gas containing the hydrogen sulfide into a desulfurization tower, wherein the desulfurization tower uses a desulfurization lean solution to carry out desulfurization treatment on the raw material gas containing the hydrogen sulfide so as to obtain purified product gas, elemental sulfur and desulfurization rich solution;

an adaptive control and adjustment module for adaptively controlling and adjusting the inflow rate of the raw material gas based on the real-time temperature and the real-time pressure of the desulfurization lean solution;

Wherein the adaptive control and adjustment module comprises:

Acquiring a time sequence of the inlet flow velocity of the raw material gas acquired by the sensor group and a time sequence of the real-time temperature and the real-time pressure of the desulfurization lean solution;

Performing correlation analysis on the time sequence of the real-time temperature and the real-time pressure of the desulfurization lean solution to obtain a temperature-pressure time sequence correlation characteristic vector;

extracting time sequence characteristics of the time sequence of the inflow flow velocity of the raw material gas to obtain an inflow flow velocity time sequence related characteristic vector;

An incoming flow control instruction is determined based on characteristic interaction response correlation information between the incoming flow timing correlation characteristic vector and the temperature-pressure timing correlation characteristic vector.